sample conclusion and recommendation in research

How to Write a Conclusion for Research Papers (with Examples)

The conclusion of a research paper is a crucial section that plays a significant role in the overall impact and effectiveness of your research paper. However, this is also the section that typically receives less attention compared to the introduction and the body of the paper. The conclusion serves to provide a concise summary of the key findings, their significance, their implications, and a sense of closure to the study. Discussing how can the findings be applied in real-world scenarios or inform policy, practice, or decision-making is especially valuable to practitioners and policymakers. The research paper conclusion also provides researchers with clear insights and valuable information for their own work, which they can then build on and contribute to the advancement of knowledge in the field.

The research paper conclusion should explain the significance of your findings within the broader context of your field. It restates how your results contribute to the existing body of knowledge and whether they confirm or challenge existing theories or hypotheses. Also, by identifying unanswered questions or areas requiring further investigation, your awareness of the broader research landscape can be demonstrated.

Remember to tailor the research paper conclusion to the specific needs and interests of your intended audience, which may include researchers, practitioners, policymakers, or a combination of these.

Table of Contents

What is a conclusion in a research paper, summarizing conclusion, editorial conclusion, externalizing conclusion, importance of a good research paper conclusion, how to write a conclusion for your research paper, research paper conclusion examples.

• How to write a research paper conclusion with Paperpal? 

Frequently Asked Questions

A conclusion in a research paper is the final section where you summarize and wrap up your research, presenting the key findings and insights derived from your study. The research paper conclusion is not the place to introduce new information or data that was not discussed in the main body of the paper. When working on how to conclude a research paper, remember to stick to summarizing and interpreting existing content. The research paper conclusion serves the following purposes: 1

• Warn readers of the possible consequences of not attending to the problem.
• Recommend specific course(s) of action.
• Restate key ideas to drive home the ultimate point of your research paper.
• Provide a “take-home” message that you want the readers to remember about your study.

Types of conclusions for research papers

In research papers, the conclusion provides closure to the reader. The type of research paper conclusion you choose depends on the nature of your study, your goals, and your target audience. I provide you with three common types of conclusions:

A summarizing conclusion is the most common type of conclusion in research papers. It involves summarizing the main points, reiterating the research question, and restating the significance of the findings. This common type of research paper conclusion is used across different disciplines.

An editorial conclusion is less common but can be used in research papers that are focused on proposing or advocating for a particular viewpoint or policy. It involves presenting a strong editorial or opinion based on the research findings and offering recommendations or calls to action.

An externalizing conclusion is a type of conclusion that extends the research beyond the scope of the paper by suggesting potential future research directions or discussing the broader implications of the findings. This type of conclusion is often used in more theoretical or exploratory research papers.

Align your conclusion’s tone with the rest of your research paper. Start Writing with Paperpal Now!  

The conclusion in a research paper serves several important purposes:

• Offers Implications and Recommendations : Your research paper conclusion is an excellent place to discuss the broader implications of your research and suggest potential areas for further study. It’s also an opportunity to offer practical recommendations based on your findings.
• Provides Closure : A good research paper conclusion provides a sense of closure to your paper. It should leave the reader with a feeling that they have reached the end of a well-structured and thought-provoking research project.
• Leaves a Lasting Impression : Writing a well-crafted research paper conclusion leaves a lasting impression on your readers. It’s your final opportunity to leave them with a new idea, a call to action, or a memorable quote.

Writing a strong conclusion for your research paper is essential to leave a lasting impression on your readers. Here’s a step-by-step process to help you create and know what to put in the conclusion of a research paper: 2

• Research Statement : Begin your research paper conclusion by restating your research statement. This reminds the reader of the main point you’ve been trying to prove throughout your paper. Keep it concise and clear.
• Key Points : Summarize the main arguments and key points you’ve made in your paper. Avoid introducing new information in the research paper conclusion. Instead, provide a concise overview of what you’ve discussed in the body of your paper.
• Address the Research Questions : If your research paper is based on specific research questions or hypotheses, briefly address whether you’ve answered them or achieved your research goals. Discuss the significance of your findings in this context.
• Significance : Highlight the importance of your research and its relevance in the broader context. Explain why your findings matter and how they contribute to the existing knowledge in your field.
• Implications : Explore the practical or theoretical implications of your research. How might your findings impact future research, policy, or real-world applications? Consider the “so what?” question.
• Future Research : Offer suggestions for future research in your area. What questions or aspects remain unanswered or warrant further investigation? This shows that your work opens the door for future exploration.
• Closing Thought : Conclude your research paper conclusion with a thought-provoking or memorable statement. This can leave a lasting impression on your readers and wrap up your paper effectively. Avoid introducing new information or arguments here.
• Proofread and Revise : Carefully proofread your conclusion for grammar, spelling, and clarity. Ensure that your ideas flow smoothly and that your conclusion is coherent and well-structured.

Write your research paper conclusion 2x faster with Paperpal. Try it now!

Remember that a well-crafted research paper conclusion is a reflection of the strength of your research and your ability to communicate its significance effectively. It should leave a lasting impression on your readers and tie together all the threads of your paper. Now you know how to start the conclusion of a research paper and what elements to include to make it impactful, let’s look at a research paper conclusion sample.

How to write a research paper conclusion with Paperpal?

A research paper conclusion is not just a summary of your study, but a synthesis of the key findings that ties the research together and places it in a broader context. A research paper conclusion should be concise, typically around one paragraph in length. However, some complex topics may require a longer conclusion to ensure the reader is left with a clear understanding of the study’s significance. Paperpal, an AI writing assistant trusted by over 800,000 academics globally, can help you write a well-structured conclusion for your research paper. 

• Sign Up or Log In: Create a new Paperpal account or login with your details.  
• Navigate to Features : Once logged in, head over to the features’ side navigation pane. Click on Templates and you’ll find a suite of generative AI features to help you write better, faster.  
• Generate an outline: Under Templates, select ‘Outlines’. Choose ‘Research article’ as your document type.  
• Select your section: Since you’re focusing on the conclusion, select this section when prompted.  
• Choose your field of study: Identifying your field of study allows Paperpal to provide more targeted suggestions, ensuring the relevance of your conclusion to your specific area of research. 
• Provide a brief description of your study: Enter details about your research topic and findings. This information helps Paperpal generate a tailored outline that aligns with your paper’s content. 
• Generate the conclusion outline: After entering all necessary details, click on ‘generate’. Paperpal will then create a structured outline for your conclusion, to help you start writing and build upon the outline.  
• Write your conclusion: Use the generated outline to build your conclusion. The outline serves as a guide, ensuring you cover all critical aspects of a strong conclusion, from summarizing key findings to highlighting the research’s implications. 
• Refine and enhance: Paperpal’s ‘Make Academic’ feature can be particularly useful in the final stages. Select any paragraph of your conclusion and use this feature to elevate the academic tone, ensuring your writing is aligned to the academic journal standards. 

By following these steps, Paperpal not only simplifies the process of writing a research paper conclusion but also ensures it is impactful, concise, and aligned with academic standards. Sign up with Paperpal today and write your research paper conclusion 2x faster .  

The research paper conclusion is a crucial part of your paper as it provides the final opportunity to leave a strong impression on your readers. In the research paper conclusion, summarize the main points of your research paper by restating your research statement, highlighting the most important findings, addressing the research questions or objectives, explaining the broader context of the study, discussing the significance of your findings, providing recommendations if applicable, and emphasizing the takeaway message. The main purpose of the conclusion is to remind the reader of the main point or argument of your paper and to provide a clear and concise summary of the key findings and their implications. All these elements should feature on your list of what to put in the conclusion of a research paper to create a strong final statement for your work.

A strong conclusion is a critical component of a research paper, as it provides an opportunity to wrap up your arguments, reiterate your main points, and leave a lasting impression on your readers. Here are the key elements of a strong research paper conclusion: 1. Conciseness : A research paper conclusion should be concise and to the point. It should not introduce new information or ideas that were not discussed in the body of the paper. 2. Summarization : The research paper conclusion should be comprehensive enough to give the reader a clear understanding of the research’s main contributions. 3 . Relevance : Ensure that the information included in the research paper conclusion is directly relevant to the research paper’s main topic and objectives; avoid unnecessary details. 4 . Connection to the Introduction : A well-structured research paper conclusion often revisits the key points made in the introduction and shows how the research has addressed the initial questions or objectives. 5. Emphasis : Highlight the significance and implications of your research. Why is your study important? What are the broader implications or applications of your findings? 6 . Call to Action : Include a call to action or a recommendation for future research or action based on your findings.

The length of a research paper conclusion can vary depending on several factors, including the overall length of the paper, the complexity of the research, and the specific journal requirements. While there is no strict rule for the length of a conclusion, but it’s generally advisable to keep it relatively short. A typical research paper conclusion might be around 5-10% of the paper’s total length. For example, if your paper is 10 pages long, the conclusion might be roughly half a page to one page in length.

In general, you do not need to include citations in the research paper conclusion. Citations are typically reserved for the body of the paper to support your arguments and provide evidence for your claims. However, there may be some exceptions to this rule: 1. If you are drawing a direct quote or paraphrasing a specific source in your research paper conclusion, you should include a citation to give proper credit to the original author. 2. If your conclusion refers to or discusses specific research, data, or sources that are crucial to the overall argument, citations can be included to reinforce your conclusion’s validity.

The conclusion of a research paper serves several important purposes: 1. Summarize the Key Points 2. Reinforce the Main Argument 3. Provide Closure 4. Offer Insights or Implications 5. Engage the Reader. 6. Reflect on Limitations

Remember that the primary purpose of the research paper conclusion is to leave a lasting impression on the reader, reinforcing the key points and providing closure to your research. It’s often the last part of the paper that the reader will see, so it should be strong and well-crafted.

• Makar, G., Foltz, C., Lendner, M., & Vaccaro, A. R. (2018). How to write effective discussion and conclusion sections. Clinical spine surgery, 31(8), 345-346.
• Bunton, D. (2005). The structure of PhD conclusion chapters.  Journal of English for academic purposes ,  4 (3), 207-224.

Paperpal is a comprehensive AI writing toolkit that helps students and researchers achieve 2x the writing in half the time. It leverages 21+ years of STM experience and insights from millions of research articles to provide in-depth academic writing, language editing, and submission readiness support to help you write better, faster.  

Get accurate academic translations, rewriting support, grammar checks, vocabulary suggestions, and generative AI assistance that delivers human precision at machine speed. Try for free or upgrade to Paperpal Prime starting at US$19 a month to access premium features, including consistency, plagiarism, and 30+ submission readiness checks to help you succeed.  

Experience the future of academic writing – Sign up to Paperpal and start writing for free!  

Related Reads:

• 5 Reasons for Rejection After Peer Review
• Ethical Research Practices For Research with Human Subjects

7 Ways to Improve Your Academic Writing Process

• Paraphrasing in Academic Writing: Answering Top Author Queries

Preflight For Editorial Desk: The Perfect Hybrid (AI + Human) Assistance Against Compromised Manuscripts

You may also like, how to cite in apa format (7th edition):..., how to write your research paper in apa..., how to choose a dissertation topic, how to write a phd research proposal, how to write an academic paragraph (step-by-step guide), research funding basics: what should a grant proposal..., how to write an abstract in research papers..., how to write dissertation acknowledgements, how to write the first draft of a..., mla works cited page: format, template & examples.

• Privacy Policy

Home » Research Paper Conclusion – Writing Guide and Examples

Research Paper Conclusion – Writing Guide and Examples

Table of Contents

Research Paper Conclusion

Definition:

A research paper conclusion is the final section of a research paper that summarizes the key findings, significance, and implications of the research. It is the writer’s opportunity to synthesize the information presented in the paper, draw conclusions, and make recommendations for future research or actions.

The conclusion should provide a clear and concise summary of the research paper, reiterating the research question or problem, the main results, and the significance of the findings. It should also discuss the limitations of the study and suggest areas for further research.

Parts of Research Paper Conclusion

The parts of a research paper conclusion typically include:

Restatement of the Thesis

The conclusion should begin by restating the thesis statement from the introduction in a different way. This helps to remind the reader of the main argument or purpose of the research.

Summary of Key Findings

The conclusion should summarize the main findings of the research, highlighting the most important results and conclusions. This section should be brief and to the point.

Implications and Significance

In this section, the researcher should explain the implications and significance of the research findings. This may include discussing the potential impact on the field or industry, highlighting new insights or knowledge gained, or pointing out areas for future research.

Limitations and Recommendations

It is important to acknowledge any limitations or weaknesses of the research and to make recommendations for how these could be addressed in future studies. This shows that the researcher is aware of the potential limitations of their work and is committed to improving the quality of research in their field.

Concluding Statement

The conclusion should end with a strong concluding statement that leaves a lasting impression on the reader. This could be a call to action, a recommendation for further research, or a final thought on the topic.

How to Write Research Paper Conclusion

Here are some steps you can follow to write an effective research paper conclusion:

• Restate the research problem or question: Begin by restating the research problem or question that you aimed to answer in your research. This will remind the reader of the purpose of your study.
• Summarize the main points: Summarize the key findings and results of your research. This can be done by highlighting the most important aspects of your research and the evidence that supports them.
• Discuss the implications: Discuss the implications of your findings for the research area and any potential applications of your research. You should also mention any limitations of your research that may affect the interpretation of your findings.
• Provide a conclusion : Provide a concise conclusion that summarizes the main points of your paper and emphasizes the significance of your research. This should be a strong and clear statement that leaves a lasting impression on the reader.
• Offer suggestions for future research: Lastly, offer suggestions for future research that could build on your findings and contribute to further advancements in the field.

Remember that the conclusion should be brief and to the point, while still effectively summarizing the key findings and implications of your research.

Example of Research Paper Conclusion

Here’s an example of a research paper conclusion:

Conclusion :

In conclusion, our study aimed to investigate the relationship between social media use and mental health among college students. Our findings suggest that there is a significant association between social media use and increased levels of anxiety and depression among college students. This highlights the need for increased awareness and education about the potential negative effects of social media use on mental health, particularly among college students.

Despite the limitations of our study, such as the small sample size and self-reported data, our findings have important implications for future research and practice. Future studies should aim to replicate our findings in larger, more diverse samples, and investigate the potential mechanisms underlying the association between social media use and mental health. In addition, interventions should be developed to promote healthy social media use among college students, such as mindfulness-based approaches and social media detox programs.

Overall, our study contributes to the growing body of research on the impact of social media on mental health, and highlights the importance of addressing this issue in the context of higher education. By raising awareness and promoting healthy social media use among college students, we can help to reduce the negative impact of social media on mental health and improve the well-being of young adults.

Purpose of Research Paper Conclusion

The purpose of a research paper conclusion is to provide a summary and synthesis of the key findings, significance, and implications of the research presented in the paper. The conclusion serves as the final opportunity for the writer to convey their message and leave a lasting impression on the reader.

The conclusion should restate the research problem or question, summarize the main results of the research, and explain their significance. It should also acknowledge the limitations of the study and suggest areas for future research or action.

Overall, the purpose of the conclusion is to provide a sense of closure to the research paper and to emphasize the importance of the research and its potential impact. It should leave the reader with a clear understanding of the main findings and why they matter. The conclusion serves as the writer’s opportunity to showcase their contribution to the field and to inspire further research and action.

When to Write Research Paper Conclusion

The conclusion of a research paper should be written after the body of the paper has been completed. It should not be written until the writer has thoroughly analyzed and interpreted their findings and has written a complete and cohesive discussion of the research.

Before writing the conclusion, the writer should review their research paper and consider the key points that they want to convey to the reader. They should also review the research question, hypotheses, and methodology to ensure that they have addressed all of the necessary components of the research.

Once the writer has a clear understanding of the main findings and their significance, they can begin writing the conclusion. The conclusion should be written in a clear and concise manner, and should reiterate the main points of the research while also providing insights and recommendations for future research or action.

Characteristics of Research Paper Conclusion

The characteristics of a research paper conclusion include:

• Clear and concise: The conclusion should be written in a clear and concise manner, summarizing the key findings and their significance.
• Comprehensive: The conclusion should address all of the main points of the research paper, including the research question or problem, the methodology, the main results, and their implications.
• Future-oriented : The conclusion should provide insights and recommendations for future research or action, based on the findings of the research.
• Impressive : The conclusion should leave a lasting impression on the reader, emphasizing the importance of the research and its potential impact.
• Objective : The conclusion should be based on the evidence presented in the research paper, and should avoid personal biases or opinions.
• Unique : The conclusion should be unique to the research paper and should not simply repeat information from the introduction or body of the paper.

Advantages of Research Paper Conclusion

The advantages of a research paper conclusion include:

• Summarizing the key findings : The conclusion provides a summary of the main findings of the research, making it easier for the reader to understand the key points of the study.
• Emphasizing the significance of the research: The conclusion emphasizes the importance of the research and its potential impact, making it more likely that readers will take the research seriously and consider its implications.
• Providing recommendations for future research or action : The conclusion suggests practical recommendations for future research or action, based on the findings of the study.
• Providing closure to the research paper : The conclusion provides a sense of closure to the research paper, tying together the different sections of the paper and leaving a lasting impression on the reader.
• Demonstrating the writer’s contribution to the field : The conclusion provides the writer with an opportunity to showcase their contribution to the field and to inspire further research and action.

Limitations of Research Paper Conclusion

While the conclusion of a research paper has many advantages, it also has some limitations that should be considered, including:

• I nability to address all aspects of the research: Due to the limited space available in the conclusion, it may not be possible to address all aspects of the research in detail.
• Subjectivity : While the conclusion should be objective, it may be influenced by the writer’s personal biases or opinions.
• Lack of new information: The conclusion should not introduce new information that has not been discussed in the body of the research paper.
• Lack of generalizability: The conclusions drawn from the research may not be applicable to other contexts or populations, limiting the generalizability of the study.
• Misinterpretation by the reader: The reader may misinterpret the conclusions drawn from the research, leading to a misunderstanding of the findings.

About the author

Muhammad Hassan

Researcher, Academic Writer, Web developer

You may also like

Research Approach – Types Methods and Examples

Scope of the Research – Writing Guide and...

Figures in Research Paper – Examples and Guide

How to Publish a Research Paper – Step by Step...

Conceptual Framework – Types, Methodology and...

APA Table of Contents – Format and Example

Have a language expert improve your writing

Run a free plagiarism check in 10 minutes, automatically generate references for free.

• Knowledge Base
• Dissertation
• How to Write a Thesis or Dissertation Conclusion

How to Write a Dissertation Conclusion | Checklist and Examples

Published on 9 September 2022 by Tegan George and Shona McCombes. Revised on 10 October 2022.

The conclusion is the very last part of your thesis or dissertation . It should be concise and engaging, leaving your reader with a clear understanding of your main findings, as well as the answer to your research question .

In it, you should:

• Clearly state the answer to your main research question
• Summarise and reflect on your research process
• Make recommendations for future work on your topic
• Show what new knowledge you have contributed to your field
• Wrap up your thesis or dissertation

Instantly correct all language mistakes in your text

Be assured that you'll submit flawless writing. Upload your document to correct all your mistakes.

Table of contents

Discussion vs. conclusion, how long should your conclusion be, step 1: answer your research question, step 2: summarise and reflect on your research, step 3: make future recommendations, step 4: emphasise your contributions to your field, step 5: wrap up your thesis or dissertation, full conclusion example, conclusion checklist, frequently asked questions about conclusion sections.

While your conclusion contains similar elements to your discussion section , they are not the same thing.

Your conclusion should be shorter and more general than your discussion. Instead of repeating literature from your literature review , discussing specific research results , or interpreting your data in detail, concentrate on making broad statements that sum up the most important insights of your research.

As a rule of thumb, your conclusion should not introduce new data, interpretations, or arguments.

Prevent plagiarism, run a free check.

Depending on whether you are writing a thesis or dissertation, your length will vary. Generally, a conclusion should make up around 5–7% of your overall word count.

An empirical scientific study will often have a short conclusion, concisely stating the main findings and recommendations for future research. A humanities topic or systematic review , on the other hand, might require more space to conclude its analysis, tying all the previous sections together in an overall argument.

Your conclusion should begin with the main question that your thesis or dissertation aimed to address. This is your final chance to show that you’ve done what you set out to do, so make sure to formulate a clear, concise answer.

• Don’t repeat a list of all the results that you already discussed
• Do synthesise them into a final takeaway that the reader will remember.

An empirical thesis or dissertation conclusion may begin like this:

A case study –based thesis or dissertation conclusion may begin like this:

In the second example, the research aim is not directly restated, but rather added implicitly to the statement. To avoid repeating yourself, it is helpful to reformulate your aims and questions into an overall statement of what you did and how you did it.

Your conclusion is an opportunity to remind your reader why you took the approach you did, what you expected to find, and how well the results matched your expectations.

To avoid repetition , consider writing more reflectively here, rather than just writing a summary of each preceding section. Consider mentioning the effectiveness of your methodology , or perhaps any new questions or unexpected insights that arose in the process.

You can also mention any limitations of your research, but only if you haven’t already included these in the discussion. Don’t dwell on them at length, though – focus on the positives of your work.

• While x limits the generalisability of the results, this approach provides new insight into y .
• This research clearly illustrates x , but it also raises the question of y .

The only proofreading tool specialized in correcting academic writing

The academic proofreading tool has been trained on 1000s of academic texts and by native English editors. Making it the most accurate and reliable proofreading tool for students.

Correct my document today

You may already have made a few recommendations for future research in your discussion section, but the conclusion is a good place to elaborate and look ahead, considering the implications of your findings in both theoretical and practical terms.

• Based on these conclusions, practitioners should consider …
• To better understand the implications of these results, future studies could address …
• Further research is needed to determine the causes of/effects of/relationship between …

When making recommendations for further research, be sure not to undermine your own work. Relatedly, while future studies might confirm, build on, or enrich your conclusions, they shouldn’t be required for your argument to feel complete. Your work should stand alone on its own merits.

Just as you should avoid too much self-criticism, you should also avoid exaggerating the applicability of your research. If you’re making recommendations for policy, business, or other practical implementations, it’s generally best to frame them as ‘shoulds’ rather than ‘musts’. All in all, the purpose of academic research is to inform, explain, and explore – not to demand.

Make sure your reader is left with a strong impression of what your research has contributed to the state of your field.

Some strategies to achieve this include:

• Returning to your problem statement to explain how your research helps solve the problem
• Referring back to the literature review and showing how you have addressed a gap in knowledge
• Discussing how your findings confirm or challenge an existing theory or assumption

Again, avoid simply repeating what you’ve already covered in the discussion in your conclusion. Instead, pick out the most important points and sum them up succinctly, situating your project in a broader context.

The end is near! Once you’ve finished writing your conclusion, it’s time to wrap up your thesis or dissertation with a few final steps:

• It’s a good idea to write your abstract next, while the research is still fresh in your mind.
• Next, make sure your reference list is complete and correctly formatted. To speed up the process, you can use our free APA citation generator .
• Once you’ve added any appendices , you can create a table of contents and title page .
• Finally, read through the whole document again to make sure your thesis is clearly written and free from language errors. You can proofread it yourself , ask a friend, or consider Scribbr’s proofreading and editing service .

Here is an example of how you can write your conclusion section. Notice how it includes everything mentioned above:

V. Conclusion

The current research aimed to identify acoustic speech characteristics which mark the beginning of an exacerbation in COPD patients.

The central questions for this research were as follows: 1. Which acoustic measures extracted from read speech differ between COPD speakers in stable condition and healthy speakers? 2. In what ways does the speech of COPD patients during an exacerbation differ from speech of COPD patients during stable periods?

All recordings were aligned using a script. Subsequently, they were manually annotated to indicate respiratory actions such as inhaling and exhaling. The recordings of 9 stable COPD patients reading aloud were then compared with the recordings of 5 healthy control subjects reading aloud. The results showed a significant effect of condition on the number of in- and exhalations per syllable, the number of non-linguistic in- and exhalations per syllable, and the ratio of voiced and silence intervals. The number of in- and exhalations per syllable and the number of non-linguistic in- and exhalations per syllable were higher for COPD patients than for healthy controls, which confirmed both hypotheses.

However, the higher ratio of voiced and silence intervals for COPD patients compared to healthy controls was not in line with the hypotheses. This unpredicted result might have been caused by the different reading materials or recording procedures for both groups, or by a difference in reading skills. Moreover, there was a trend regarding the effect of condition on the number of syllables per breath group. The number of syllables per breath group was higher for healthy controls than for COPD patients, which was in line with the hypothesis. There was no effect of condition on pitch, intensity, center of gravity, pitch variability, speaking rate, or articulation rate.

This research has shown that the speech of COPD patients in exacerbation differs from the speech of COPD patients in stable condition. This might have potential for the detection of exacerbations. However, sustained vowels rarely occur in spontaneous speech. Therefore, the last two outcome measures might have greater potential for the detection of beginning exacerbations, but further research on the different outcome measures and their potential for the detection of exacerbations is needed due to the limitations of the current study.

Checklist: Conclusion

I have clearly and concisely answered the main research question .

I have summarized my overall argument or key takeaways.

I have mentioned any important limitations of the research.

I have given relevant recommendations .

I have clearly explained what my research has contributed to my field.

I have  not introduced any new data or arguments.

You've written a great conclusion! Use the other checklists to further improve your dissertation.

In a thesis or dissertation, the discussion is an in-depth exploration of the results, going into detail about the meaning of your findings and citing relevant sources to put them in context.

The conclusion is more shorter and more general: it concisely answers your main research question and makes recommendations based on your overall findings.

While it may be tempting to present new arguments or evidence in your thesis or disseration conclusion , especially if you have a particularly striking argument you’d like to finish your analysis with, you shouldn’t. Theses and dissertations follow a more formal structure than this.

All your findings and arguments should be presented in the body of the text (more specifically in the discussion section and results section .) The conclusion is meant to summarize and reflect on the evidence and arguments you have already presented, not introduce new ones.

For a stronger dissertation conclusion , avoid including:

• Generic concluding phrases (e.g. “In conclusion…”)
• Weak statements that undermine your argument (e.g. “There are good points on both sides of this issue.”)

Your conclusion should leave the reader with a strong, decisive impression of your work.

The conclusion of your thesis or dissertation shouldn’t take up more than 5-7% of your overall word count.

The conclusion of your thesis or dissertation should include the following:

• A restatement of your research question
• A summary of your key arguments and/or results
• A short discussion of the implications of your research

Cite this Scribbr article

If you want to cite this source, you can copy and paste the citation or click the ‘Cite this Scribbr article’ button to automatically add the citation to our free Reference Generator.

George, T. & McCombes, S. (2022, October 10). How to Write a Dissertation Conclusion | Checklist and Examples. Scribbr. Retrieved 30 August 2024, from https://www.scribbr.co.uk/thesis-dissertation/conclusion/

Is this article helpful?

Tegan George

Other students also liked, how to write a thesis or dissertation introduction, how to write a discussion section | tips & examples, how to write an abstract | steps & examples.

• USC Libraries
• Research Guides

Organizing Your Social Sciences Research Paper

• 9. The Conclusion
• Purpose of Guide
• Design Flaws to Avoid
• Independent and Dependent Variables
• Glossary of Research Terms
• Reading Research Effectively
• Narrowing a Topic Idea
• Broadening a Topic Idea
• Extending the Timeliness of a Topic Idea
• Academic Writing Style
• Applying Critical Thinking
• Choosing a Title
• Making an Outline
• Paragraph Development
• Research Process Video Series
• Executive Summary
• The C.A.R.S. Model
• Background Information
• The Research Problem/Question
• Theoretical Framework
• Citation Tracking
• Content Alert Services
• Evaluating Sources
• Primary Sources
• Secondary Sources
• Tiertiary Sources
• Scholarly vs. Popular Publications
• Qualitative Methods
• Quantitative Methods
• Insiderness
• Using Non-Textual Elements
• Limitations of the Study
• Common Grammar Mistakes
• Writing Concisely
• Avoiding Plagiarism
• Footnotes or Endnotes?
• Further Readings
• Generative AI and Writing
• USC Libraries Tutorials and Other Guides
• Bibliography

The conclusion is intended to help the reader understand why your research should matter to them after they have finished reading the paper. A conclusion is not merely a summary of the main topics covered or a re-statement of your research problem, but a synthesis of key points derived from the findings of your study and, if applicable based on your analysis, explain new areas for future research. For most college-level research papers, two or three well-developed paragraphs is sufficient for a conclusion, although in some cases, more paragraphs may be required in describing the key findings and highlighting their significance.

Conclusions. The Writing Center. University of North Carolina; Conclusions. The Writing Lab and The OWL. Purdue University.

Importance of a Good Conclusion

A well-written conclusion provides important opportunities to demonstrate to the reader your understanding of the research problem. These include:

• Presenting the last word on the issues you raised in your paper . Just as the introduction gives a first impression to your reader, the conclusion offers a chance to leave a lasting impression. Do this, for example, by highlighting key findings in your analysis that advance new understanding about the research problem, that are unusual or unexpected, or that have important implications applied to practice.
• Summarizing your thoughts and conveying the larger significance of your study . The conclusion is an opportunity to succinctly re-emphasize  your answer to the "So What?" question by placing the study within the context of how your research advances past studies about the topic.
• Identifying how a gap in the literature has been addressed . The conclusion can be where you describe how a previously identified gap in the literature [first identified in your literature review section] has been addressed by your research and why this contribution is significant.
• Demonstrating the importance of your ideas . Don't be shy. The conclusion offers an opportunity to elaborate on the impact and significance of your findings. This is particularly important if your study approached examining the research problem from an unusual or innovative perspective.
• Introducing possible new or expanded ways of thinking about the research problem . This does not refer to introducing new information [which should be avoided], but to offer new insight and creative approaches for framing or contextualizing the research problem based on the results of your study.

Bunton, David. “The Structure of PhD Conclusion Chapters.” Journal of English for Academic Purposes 4 (July 2005): 207–224; Conclusions. The Writing Center. University of North Carolina; Kretchmer, Paul. Twelve Steps to Writing an Effective Conclusion. San Francisco Edit, 2003-2008; Conclusions. The Writing Lab and The OWL. Purdue University; Assan, Joseph. "Writing the Conclusion Chapter: The Good, the Bad and the Missing." Liverpool: Development Studies Association (2009): 1-8.

Structure and Writing Style

I.  General Rules

The general function of your paper's conclusion is to restate the main argument . It reminds the reader of your main argument(s) strengths and reiterates the most important evidence supporting those argument(s). Do this by clearly summarizing the context, background, and the necessity of examining the research problem in relation to an issue, controversy, or a gap found in the literature. However, make sure that your conclusion is not simply a repetitive summary of the findings. This reduces the impact of the argument(s) you have developed in your paper.

When writing the conclusion to your paper, follow these general rules:

• Present your conclusions in clear, concise language. Re-state the purpose of your study, then describe how your findings differ or support those of other studies and why [i.e., describe what were the unique, new, or crucial contributions your study made to the overall research about your topic].
• Do not simply reiterate your findings or the discussion of your results. Provide a synthesis of arguments presented in the paper to show how these converge to address the research problem and the overall objectives of your study.
• Indicate opportunities for future research if you haven't already done so in the discussion section of your paper. Highlighting the need for further research provides the reader with evidence that you have an in-depth awareness of the research problem but that further analysis should take place beyond the scope of your investigation.

Consider the following points to help ensure your conclusion is presented well:

• If the argument or purpose of your paper is complex, you may need to summarize the argument for your reader.
• If, prior to your conclusion, you have not yet explained the significance of your findings or if you are proceeding inductively, use the end of your paper to describe your main points and explain their significance.
• Move from a detailed to a general level of consideration that returns the topic to the context provided by the introduction or within a new context that emerges from the data [this is opposite of the introduction, which begins with general discussion of the context and ends with a detailed description of the research problem]. 

The conclusion also provides a place for you to persuasively and succinctly restate the research problem, given that the reader has now been presented with all the information about the topic . Depending on the discipline you are writing in, the concluding paragraph may contain your reflections on the evidence presented. However, the nature of being introspective about the research you have conducted will depend on the topic and whether your professor wants you to express your observations in this way. If asked to think introspectively about the topic, do not delve into idle speculation. Being introspective means looking within yourself as an author to try and understand an issue more deeply, not to guess at possible outcomes or make up scenarios not supported by the evidence.

II.  Developing a Compelling Conclusion

Although an effective conclusion needs to be clear and succinct, it does not need to be written passively or lack a compelling narrative. Strategies to help you move beyond merely summarizing the key points of your research paper may include any of the following:

• If your paper addresses a critical, contemporary problem, warn readers of the possible consequences of not attending to the problem proactively based on the evidence presented in your study.
• Recommend a specific course or courses of action that, if adopted, could address a specific problem in practice or in the development of new knowledge leading to positive change.
• Cite a relevant quotation or expert opinion already noted in your paper in order to lend authority and support to the conclusion(s) you have reached [a good source would be from a source cited in your literature review].
• Explain the consequences of your research in a way that elicits action or demonstrates urgency in seeking change.
• Restate a key statistic, fact, or visual image to emphasize the most important finding of your paper.
• If your discipline encourages personal reflection, illustrate your concluding point by drawing from your own life experiences.
• Return to an anecdote, an example, or a quotation that you presented in your introduction, but add further insight derived from the findings of your study; use your interpretation of results from your study to recast it in new or important ways.
• Provide a "take-home" message in the form of a succinct, declarative statement that you want the reader to remember about your study.

III. Problems to Avoid

Failure to be concise Your conclusion section should be concise and to the point. Conclusions that are too lengthy often have unnecessary information in them. The conclusion is not the place for details about your methodology or results. Although you should give a summary of what was learned from your research, this summary should be relatively brief, since the emphasis in the conclusion is on the implications, evaluations, insights, and other forms of analysis that you make. Strategies for writing concisely can be found here .

Failure to comment on larger, more significant issues In the introduction, your task was to move from the general [topic studied within the field of study] to the specific [the research problem]. However, in the conclusion, your task is to move the discussion from specific [your research problem] back to a general discussion framed around the implications and significance of your findings [i.e., how your research contributes new understanding or fills an important gap in the literature]. In short, the conclusion is where you should place your research within a larger context [visualize the structure of your paper as an hourglass--start with a broad introduction and review of the literature, move to the specific method of analysis and the discussion, conclude with a broad summary of the study's implications and significance].

Failure to reveal problems and negative results Negative aspects of the research process should never be ignored. These are problems, deficiencies, or challenges encountered during your study. They should be summarized as a way of qualifying your overall conclusions. If you encountered negative or unintended results [i.e., findings that are validated outside the research context in which they were generated], you must report them in the results section and discuss their implications in the discussion section of your paper. In the conclusion, use negative or surprising results as an opportunity to explain their possible significance and/or how they may form the basis for future research.

Failure to provide a clear summary of what was learned In order to discuss how your research fits within your field of study [and possibly the world at large], you need to summarize briefly and succinctly how it contributes to new knowledge or a new understanding about the research problem. This element of your conclusion may be only a few sentences long, but it often represents the key takeaway for your reader.

Failure to match the objectives of your research Often research objectives in the social and behavioral sciences change while the research is being carried out due to unforeseen factors or unanticipated variables. This is not a problem unless you forget to go back and refine the original objectives in your introduction. As these changes emerge they must be documented so that they accurately reflect what you were trying to accomplish in your research [not what you thought you might accomplish when you began].

Resist the urge to apologize If you've immersed yourself in studying the research problem, you presumably should know a good deal about it [perhaps even more than your professor!]. Nevertheless, by the time you have finished writing, you may be having some doubts about what you have produced. Repress those doubts! Don't undermine your authority as a researcher by saying something like, "This is just one approach to examining this problem; there may be other, much better approaches that...." The overall tone of your conclusion should convey confidence to the reader concerning the validity and realiability of your research.

Assan, Joseph. "Writing the Conclusion Chapter: The Good, the Bad and the Missing." Liverpool: Development Studies Association (2009): 1-8; Concluding Paragraphs. College Writing Center at Meramec. St. Louis Community College; Conclusions. The Writing Center. University of North Carolina; Conclusions. The Writing Lab and The OWL. Purdue University; Freedman, Leora  and Jerry Plotnick. Introductions and Conclusions. The Lab Report. University College Writing Centre. University of Toronto; Leibensperger, Summer. Draft Your Conclusion. Academic Center, the University of Houston-Victoria, 2003; Make Your Last Words Count. The Writer’s Handbook. Writing Center. University of Wisconsin Madison; Miquel, Fuster-Marquez and Carmen Gregori-Signes. “Chapter Six: ‘Last but Not Least:’ Writing the Conclusion of Your Paper.” In Writing an Applied Linguistics Thesis or Dissertation: A Guide to Presenting Empirical Research . John Bitchener, editor. (Basingstoke,UK: Palgrave Macmillan, 2010), pp. 93-105; Tips for Writing a Good Conclusion. Writing@CSU. Colorado State University; Kretchmer, Paul. Twelve Steps to Writing an Effective Conclusion. San Francisco Edit, 2003-2008; Writing Conclusions. Writing Tutorial Services, Center for Innovative Teaching and Learning. Indiana University; Writing: Considering Structure and Organization. Institute for Writing Rhetoric. Dartmouth College.

Writing Tip

Don't Belabor the Obvious!

Avoid phrases like "in conclusion...," "in summary...," or "in closing...." These phrases can be useful, even welcome, in oral presentations. But readers can see by the tell-tale section heading and number of pages remaining that they are reaching the end of your paper. You'll irritate your readers if you belabor the obvious.

Assan, Joseph. "Writing the Conclusion Chapter: The Good, the Bad and the Missing." Liverpool: Development Studies Association (2009): 1-8.

Another Writing Tip

New Insight, Not New Information!

Don't surprise the reader with new information in your conclusion that was never referenced anywhere else in the paper. This is why the conclusion rarely has citations to sources that haven't been referenced elsewhere in your paper. If you have new information to present, add it to the discussion or other appropriate section of the paper. Note that, although no new information is introduced, the conclusion, along with the discussion section, is where you offer your most "original" contributions in the paper; the conclusion is where you describe the value of your research, demonstrate that you understand the material that you have presented, and position your findings within the larger context of scholarship on the topic, including describing how your research contributes new insights to that scholarship.

Assan, Joseph. "Writing the Conclusion Chapter: The Good, the Bad and the Missing." Liverpool: Development Studies Association (2009): 1-8; Conclusions. The Writing Center. University of North Carolina.

• << Previous: Limitations of the Study
• Next: Appendices >>
• Last Updated: Aug 30, 2024 10:02 AM
• URL: https://libguides.usc.edu/writingguide

How To Write The Conclusion Chapter

A Simple Explainer With Examples + Free Template

By: Jenna Crossley (PhD) | Reviewed By: Dr. Eunice Rautenbach | September 2021

So, you’ve wrapped up your results and discussion chapters, and you’re finally on the home stretch – the conclusion chapter . In this post, we’ll discuss everything you need to know to craft a high-quality conclusion chapter for your dissertation or thesis project.

Overview: The Conclusion Chapter

• What the thesis/dissertation conclusion chapter is
• What to include in your conclusion
• How to structure and write up your conclusion
• A few tips  to help you ace the chapter
• FREE conclusion template

What is the conclusion chapter?

The conclusion chapter is typically the final major chapter of a dissertation or thesis. As such, it serves as a concluding summary of your research findings and wraps up the document. While some publications such as journal articles and research reports combine the discussion and conclusion sections, these are typically separate chapters in a dissertation or thesis. As always, be sure to check what your university’s structural preference is before you start writing up these chapters.

So, what’s the difference between the discussion and the conclusion chapter?

Well, the two chapters are quite similar , as they both discuss the key findings of the study. However, the conclusion chapter is typically more general and high-level in nature. In your discussion chapter, you’ll typically discuss the intricate details of your study, but in your conclusion chapter, you’ll take a   broader perspective, reporting on the main research outcomes and how these addressed your research aim (or aims) .

A core function of the conclusion chapter is to synthesise all major points covered in your study and to tell the reader what they should take away from your work. Basically, you need to tell them what you found , why it’s valuable , how it can be applied , and what further research can be done.

Whatever you do, don’t just copy and paste what you’ve written in your discussion chapter! The conclusion chapter should not be a simple rehash of the discussion chapter. While the two chapters are similar, they have distinctly different functions.  

What should I include in the conclusion chapter?

To understand what needs to go into your conclusion chapter, it’s useful to understand what the chapter needs to achieve. In general, a good dissertation conclusion chapter should achieve the following:

• Summarise the key findings of the study
• Explicitly answer the research question(s) and address the research aims
• Inform the reader of the study’s main contributions
• Discuss any limitations or weaknesses of the study
• Present recommendations for future research

Therefore, your conclusion chapter needs to cover these core components. Importantly, you need to be careful not to include any new findings or data points. Your conclusion chapter should be based purely on data and analysis findings that you’ve already presented in the earlier chapters. If there’s a new point you want to introduce, you’ll need to go back to your results and discussion chapters to weave the foundation in there.

In many cases, readers will jump from the introduction chapter directly to the conclusions chapter to get a quick overview of the study’s purpose and key findings. Therefore, when you write up your conclusion chapter, it’s useful to assume that the reader hasn’t consumed the inner chapters of your dissertation or thesis. In other words, craft your conclusion chapter such that there’s a strong connection and smooth flow between the introduction and conclusion chapters, even though they’re on opposite ends of your document.

Need a helping hand?

How to write the conclusion chapter

Now that you have a clearer view of what the conclusion chapter is about, let’s break down the structure of this chapter so that you can get writing. Keep in mind that this is merely a typical structure – it’s not set in stone or universal. Some universities will prefer that you cover some of these points in the discussion chapter , or that you cover the points at different levels in different chapters.

Step 1: Craft a brief introduction section

As with all chapters in your dissertation or thesis, the conclusions chapter needs to start with a brief introduction. In this introductory section, you’ll want to tell the reader what they can expect to find in the chapter, and in what order . Here’s an example of what this might look like:

This chapter will conclude the study by summarising the key research findings in relation to the research aims and questions and discussing the value and contribution thereof. It will also review the limitations of the study and propose opportunities for future research.

Importantly, the objective here is just to give the reader a taste of what’s to come (a roadmap of sorts), not a summary of the chapter. So, keep it short and sweet – a paragraph or two should be ample.

Step 2: Discuss the overall findings in relation to the research aims

The next step in writing your conclusions chapter is to discuss the overall findings of your study , as they relate to the research aims and research questions . You would have likely covered similar ground in the discussion chapter, so it’s important to zoom out a little bit here and focus on the broader findings – specifically, how these help address the research aims .

In practical terms, it’s useful to start this section by reminding your reader of your research aims and research questions, so that the findings are well contextualised. In this section, phrases such as, “This study aimed to…” and “the results indicate that…” will likely come in handy. For example, you could say something like the following:

This study aimed to investigate the feeding habits of the naked mole-rat. The results indicate that naked mole rats feed on underground roots and tubers. Further findings show that these creatures eat only a part of the plant, leaving essential parts to ensure long-term food stability.

Be careful not to make overly bold claims here. Avoid claims such as “this study proves that” or “the findings disprove existing the existing theory”. It’s seldom the case that a single study can prove or disprove something. Typically, this is achieved by a broader body of research, not a single study – especially not a dissertation or thesis which will inherently have significant  limitations . We’ll discuss those limitations a little later.

Step 3: Discuss how your study contributes to the field

Next, you’ll need to discuss how your research has contributed to the field – both in terms of theory and practice . This involves talking about what you achieved in your study, highlighting why this is important and valuable, and how it can be used or applied.

In this section you’ll want to:

• Mention any research outputs created as a result of your study (e.g., articles, publications, etc.)
• Inform the reader on just how your research solves your research problem , and why that matters
• Reflect on gaps in the existing research and discuss how your study contributes towards addressing these gaps
• Discuss your study in relation to relevant theories . For example, does it confirm these theories or constructively challenge them?
• Discuss how your research findings can be applied in the real world . For example, what specific actions can practitioners take, based on your findings?

Be careful to strike a careful balance between being firm but humble in your arguments here. It’s unlikely that your one study will fundamentally change paradigms or shake up the discipline, so making claims to this effect will be frowned upon . At the same time though, you need to present your arguments with confidence, firmly asserting the contribution your research has made, however small that contribution may be. Simply put, you need to keep it balanced .

Step 4: Reflect on the limitations of your study

Now that you’ve pumped your research up, the next step is to critically reflect on the limitations and potential shortcomings of your study. You may have already covered this in the discussion chapter, depending on your university’s structural preferences, so be careful not to repeat yourself unnecessarily.

There are many potential limitations that can apply to any given study. Some common ones include:

• Sampling issues that reduce the generalisability of the findings (e.g., non-probability sampling )
• Insufficient sample size (e.g., not getting enough survey responses ) or limited data access
• Low-resolution data collection or analysis techniques
• Researcher bias or lack of experience
• Lack of access to research equipment
• Time constraints that limit the methodology (e.g. cross-sectional vs longitudinal time horizon)
• Budget constraints that limit various aspects of the study

Discussing the limitations of your research may feel self-defeating (no one wants to highlight their weaknesses, right), but it’s a critical component of high-quality research. It’s important to appreciate that all studies have limitations (even well-funded studies by expert researchers) – therefore acknowledging these limitations adds credibility to your research by showing that you understand the limitations of your research design .

That being said, keep an eye on your wording and make sure that you don’t undermine your research . It’s important to strike a balance between recognising the limitations, but also highlighting the value of your research despite those limitations. Show the reader that you understand the limitations, that these were justified given your constraints, and that you know how they can be improved upon – this will get you marks.

Next, you’ll need to make recommendations for future studies. This will largely be built on the limitations you just discussed. For example, if one of your study’s weaknesses was related to a specific data collection or analysis method, you can make a recommendation that future researchers undertake similar research using a more sophisticated method.

Another potential source of future research recommendations is any data points or analysis findings that were interesting or surprising , but not directly related to your study’s research aims and research questions. So, if you observed anything that “stood out” in your analysis, but you didn’t explore it in your discussion (due to a lack of relevance to your research aims), you can earmark that for further exploration in this section.

Essentially, this section is an opportunity to outline how other researchers can build on your study to take the research further and help develop the body of knowledge. So, think carefully about the new questions that your study has raised, and clearly outline these for future researchers to pick up on.

Step 6: Wrap up with a closing summary

Tips for a top-notch conclusion chapter

Now that we’ve covered the what , why and how of the conclusion chapter, here are some quick tips and suggestions to help you craft a rock-solid conclusion.

• Don’t ramble . The conclusion chapter usually consumes 5-7% of the total word count (although this will vary between universities), so you need to be concise. Edit this chapter thoroughly with a focus on brevity and clarity.
• Be very careful about the claims you make in terms of your study’s contribution. Nothing will make the marker’s eyes roll back faster than exaggerated or unfounded claims. Be humble but firm in your claim-making.
• Use clear and simple language that can be easily understood by an intelligent layman. Remember that not every reader will be an expert in your field, so it’s important to make your writing accessible. Bear in mind that no one knows your research better than you do, so it’s important to spell things out clearly for readers.

Hopefully, this post has given you some direction and confidence to take on the conclusion chapter of your dissertation or thesis with confidence. If you’re still feeling a little shaky and need a helping hand, consider booking a free initial consultation with a friendly Grad Coach to discuss how we can help you with hands-on, private coaching.

Psst... there’s more!

This post was based on one of our popular Research Bootcamps . If you're working on a research project, you'll definitely want to check this out ...

17 Comments

Really you team are doing great!

Your guide on writing the concluding chapter of a research is really informative especially to the beginners who really do not know where to start. Im now ready to start. Keep it up guys

Really your team are doing great!

Very helpful guidelines, timely saved. Thanks so much for the tips.

This post was very helpful and informative. Thank you team.

A very enjoyable, understandable and crisp presentation on how to write a conclusion chapter. I thoroughly enjoyed it. Thanks Jenna.

This was a very helpful article which really gave me practical pointers for my concluding chapter. Keep doing what you are doing! It meant a lot to me to be able to have this guide. Thank you so much.

Nice content dealing with the conclusion chapter, it’s a relief after the streneous task of completing discussion part.Thanks for valuable guidance

Thanks for your guidance

I get all my doubts clarified regarding the conclusion chapter. It’s really amazing. Many thanks.

Very helpful tips. Thanks so much for the guidance

Thank you very much for this piece. It offers a very helpful starting point in writing the conclusion chapter of my thesis.

It’s awesome! Most useful and timely too. Thanks a million times

Bundle of thanks for your guidance. It was greatly helpful.

Wonderful, clear, practical guidance. So grateful to read this as I conclude my research. Thank you.

Submit a Comment Cancel reply

Your email address will not be published. Required fields are marked *

Save my name, email, and website in this browser for the next time I comment.

• Print Friendly

Research Recommendations – Guiding policy-makers for evidence-based decision making

Research recommendations play a crucial role in guiding scholars and researchers toward fruitful avenues of exploration. In an era marked by rapid technological advancements and an ever-expanding knowledge base, refining the process of generating research recommendations becomes imperative.

But, what is a research recommendation?

Research recommendations are suggestions or advice provided to researchers to guide their study on a specific topic . They are typically given by experts in the field. Research recommendations are more action-oriented and provide specific guidance for decision-makers, unlike implications that are broader and focus on the broader significance and consequences of the research findings. However, both are crucial components of a research study.

Difference Between Research Recommendations and Implication

Although research recommendations and implications are distinct components of a research study, they are closely related. The differences between them are as follows:

Types of Research Recommendations

Recommendations in research can take various forms, which are as follows:

These recommendations aim to assist researchers in navigating the vast landscape of academic knowledge.

Let us dive deeper to know about its key components and the steps to write an impactful research recommendation.

Key Components of Research Recommendations

The key components of research recommendations include defining the research question or objective, specifying research methods, outlining data collection and analysis processes, presenting results and conclusions, addressing limitations, and suggesting areas for future research. Here are some characteristics of research recommendations:

Research recommendations offer various advantages and play a crucial role in ensuring that research findings contribute to positive outcomes in various fields. However, they also have few limitations which highlights the significance of a well-crafted research recommendation in offering the promised advantages.

The importance of research recommendations ranges in various fields, influencing policy-making, program development, product development, marketing strategies, medical practice, and scientific research. Their purpose is to transfer knowledge from researchers to practitioners, policymakers, or stakeholders, facilitating informed decision-making and improving outcomes in different domains.

How to Write Research Recommendations?

Research recommendations can be generated through various means, including algorithmic approaches, expert opinions, or collaborative filtering techniques. Here is a step-wise guide to build your understanding on the development of research recommendations.

1. Understand the Research Question:

Understand the research question and objectives before writing recommendations. Also, ensure that your recommendations are relevant and directly address the goals of the study.

2. Review Existing Literature:

Familiarize yourself with relevant existing literature to help you identify gaps , and offer informed recommendations that contribute to the existing body of research.

3. Consider Research Methods:

Evaluate the appropriateness of different research methods in addressing the research question. Also, consider the nature of the data, the study design, and the specific objectives.

4. Identify Data Collection Techniques:

Gather dataset from diverse authentic sources. Include information such as keywords, abstracts, authors, publication dates, and citation metrics to provide a rich foundation for analysis.

5. Propose Data Analysis Methods:

Suggest appropriate data analysis methods based on the type of data collected. Consider whether statistical analysis, qualitative analysis, or a mixed-methods approach is most suitable.

6. Consider Limitations and Ethical Considerations:

Acknowledge any limitations and potential ethical considerations of the study. Furthermore, address these limitations or mitigate ethical concerns to ensure responsible research.

7. Justify Recommendations:

Explain how your recommendation contributes to addressing the research question or objective. Provide a strong rationale to help researchers understand the importance of following your suggestions.

8. Summarize Recommendations:

Provide a concise summary at the end of the report to emphasize how following these recommendations will contribute to the overall success of the research project.

By following these steps, you can create research recommendations that are actionable and contribute meaningfully to the success of the research project.

Download now to unlock some tips to improve your journey of writing research recommendations.

Example of a Research Recommendation

Here is an example of a research recommendation based on a hypothetical research to improve your understanding.

Research Recommendation: Enhancing Student Learning through Integrated Learning Platforms

Background:

The research study investigated the impact of an integrated learning platform on student learning outcomes in high school mathematics classes. The findings revealed a statistically significant improvement in student performance and engagement when compared to traditional teaching methods.

Recommendation:

In light of the research findings, it is recommended that educational institutions consider adopting and integrating the identified learning platform into their mathematics curriculum. The following specific recommendations are provided:

• Implementation of the Integrated Learning Platform:

Schools are encouraged to adopt the integrated learning platform in mathematics classrooms, ensuring proper training for teachers on its effective utilization.

• Professional Development for Educators:

Develop and implement professional programs to train educators in the effective use of the integrated learning platform to address any challenges teachers may face during the transition.

• Monitoring and Evaluation:

Establish a monitoring and evaluation system to track the impact of the integrated learning platform on student performance over time.

• Resource Allocation:

Allocate sufficient resources, both financial and technical, to support the widespread implementation of the integrated learning platform.

By implementing these recommendations, educational institutions can harness the potential of the integrated learning platform and enhance student learning experiences and academic achievements in mathematics.

This example covers the components of a research recommendation, providing specific actions based on the research findings, identifying the target audience, and outlining practical steps for implementation.

Using AI in Research Recommendation Writing

Enhancing research recommendations is an ongoing endeavor that requires the integration of cutting-edge technologies, collaborative efforts, and ethical considerations. By embracing data-driven approaches and leveraging advanced technologies, the research community can create more effective and personalized recommendation systems. However, it is accompanied by several limitations. Therefore, it is essential to approach the use of AI in research with a critical mindset, and complement its capabilities with human expertise and judgment.

Here are some limitations of integrating AI in writing research recommendation and some ways on how to counter them.

1. Data Bias

AI systems rely heavily on data for training. If the training data is biased or incomplete, the AI model may produce biased results or recommendations.

How to tackle: Audit regularly the model’s performance to identify any discrepancies and adjust the training data and algorithms accordingly.

2. Lack of Understanding of Context:

AI models may struggle to understand the nuanced context of a particular research problem. They may misinterpret information, leading to inaccurate recommendations.

How to tackle: Use AI to characterize research articles and topics. Employ them to extract features like keywords, authorship patterns and content-based details.

3. Ethical Considerations:

AI models might stereotype certain concepts or generate recommendations that could have negative consequences for certain individuals or groups.

How to tackle: Incorporate user feedback mechanisms to reduce redundancies. Establish an ethics review process for AI models in research recommendation writing.

4. Lack of Creativity and Intuition:

AI may struggle with tasks that require a deep understanding of the underlying principles or the ability to think outside the box.

How to tackle: Hybrid approaches can be employed by integrating AI in data analysis and identifying patterns for accelerating the data interpretation process.

5. Interpretability:

Many AI models, especially complex deep learning models, lack transparency on how the model arrived at a particular recommendation.

How to tackle: Implement models like decision trees or linear models. Provide clear explanation of the model architecture, training process, and decision-making criteria.

6. Dynamic Nature of Research:

Research fields are dynamic, and new information is constantly emerging. AI models may struggle to keep up with the rapidly changing landscape and may not be able to adapt to new developments.

How to tackle: Establish a feedback loop for continuous improvement. Regularly update the recommendation system based on user feedback and emerging research trends.

The integration of AI in research recommendation writing holds great promise for advancing knowledge and streamlining the research process. However, navigating these concerns is pivotal in ensuring the responsible deployment of these technologies. Researchers need to understand the use of responsible use of AI in research and must be aware of the ethical considerations.

Exploring research recommendations plays a critical role in shaping the trajectory of scientific inquiry. It serves as a compass, guiding researchers toward more robust methodologies, collaborative endeavors, and innovative approaches. Embracing these suggestions not only enhances the quality of individual studies but also contributes to the collective advancement of human understanding.

Frequently Asked Questions

The purpose of recommendations in research is to provide practical and actionable suggestions based on the study's findings, guiding future actions, policies, or interventions in a specific field or context. Recommendations bridges the gap between research outcomes and their real-world application.

To make a research recommendation, analyze your findings, identify key insights, and propose specific, evidence-based actions. Include the relevance of the recommendations to the study's objectives and provide practical steps for implementation.

Begin a recommendation by succinctly summarizing the key findings of the research. Clearly state the purpose of the recommendation and its intended impact. Use a direct and actionable language to convey the suggested course of action.

Rate this article Cancel Reply

Your email address will not be published.

Enago Academy's Most Popular Articles

• Promoting Research

Graphical Abstracts Vs. Infographics: Best practices for using visual illustrations for increased research impact

Dr. Sarah Chen stared at her computer screen, her eyes staring at her recently published…

• Publishing Research

10 Tips to Prevent Research Papers From Being Retracted

Research paper retractions represent a critical event in the scientific community. When a published article…

• Industry News

Google Releases 2024 Scholar Metrics, Evaluates Impact of Scholarly Articles

Google has released its 2024 Scholar Metrics, assessing scholarly articles from 2019 to 2023. This…

• Career Corner
• Reporting Research

How to Create a Poster That Stands Out: Tips for a smooth poster presentation

It was the conference season. Judy was excited to present her first poster! She had…

• Diversity and Inclusion

6 Reasons Why There is a Decline in Higher Education Enrollment: Action plan to overcome this crisis

Over the past decade, colleges and universities across the globe have witnessed a concerning trend…

Academic Essay Writing Made Simple: 4 types and tips

How to Effectively Cite a PDF (APA, MLA, AMA, and Chicago Style)

How to Optimize Your Research Process: A step-by-step guide

Sign-up to read more

Subscribe for free to get unrestricted access to all our resources on research writing and academic publishing including:

• 2000+ blog articles
• 50+ Webinars
• 10+ Expert podcasts
• 50+ Infographics
• 10+ Checklists
• Research Guides

We hate spam too. We promise to protect your privacy and never spam you.

• AI in Academia
• Infographics
• Expert Video Library
• Other Resources
• Enago Learn
• Upcoming & On-Demand Webinars
• Peer Review Week 2024
• Open Access Week 2023
• Conference Videos
• Enago Report
• Journal Finder
• Enago Plagiarism & AI Grammar Check
• Editing Services
• Publication Support Services
• Research Impact
• Translation Services
• Publication solutions
• AI-Based Solutions
• Thought Leadership
• Call for Articles
• Call for Speakers
• Author Training
• Edit Profile

I am looking for Editing/ Proofreading services for my manuscript Tentative date of next journal submission:

In your opinion, what is the most effective way to improve integrity in the peer review process?

Writing the parts of scientific reports

22 Writing the conclusion & recommendations

There are probably some overlaps between the Conclusion and the Discussion section. Nevertheless, this section gives you the opportunity to highlight the most important points in your report, and is sometimes the only section read. Think about what your research/ study has achieved, and the most important findings and ideas you want the reader to know. As all studies have limitations also think about what you were not able to cover (this shows that you are able to evaluate your own work objectively).

Possible structure of this section:

Use present perfect to sum up/ evaluate:

This study has explored/ has attempted …

Use past tense to state what your aim was and to refer to actions you carried out:

• This study was intended to analyse …
• The aim of this study was to …

Use present tense to evaluate your study and to state the generalizations and implications that you draw from your findings.

• The results add to the knowledge of …
• These findings s uggest that …

You can either use present tense or past tense to summarize your results.

• The findings reveal …
• It was found that …

Achievements of this study (positive)

• This study provides evidence that …
• This work has contributed to a number of key issues in the field such as …

Limitations of the study (negative)

• Several limitations should be noted. First …

Combine positive and negative remarks to give a balanced assessment:

• Although this research is somewhat limited in scope, its findings can provide a basis for future studies.
• Despite the limitations, findings from the present study can help us understand …

Use more cautious language (modal verbs may, can, could)

• There are a number of possible extensions of this research …
• The findings suggest the possibility for future research on …
• These results may be important for future studies on …
• Examining a wider context could/ would lead …

Or indicate that future research is needed

• There is still a need for future research to determine …
• Further studies should be undertaken to discover…
• It would be worthwhile to investigate …

Academic Writing in a Swiss University Context Copyright © 2018 by Irene Dietrichs. All Rights Reserved.

How to write a strong conclusion for your research paper

Last updated

17 February 2024

Reviewed by

Short on time? Get an AI generated summary of this article instead

Writing a research paper is a chance to share your knowledge and hypothesis. It's an opportunity to demonstrate your many hours of research and prove your ability to write convincingly.

Ideally, by the end of your research paper, you'll have brought your readers on a journey to reach the conclusions you've pre-determined. However, if you don't stick the landing with a good conclusion, you'll risk losing your reader’s trust.

Writing a strong conclusion for your research paper involves a few important steps, including restating the thesis and summing up everything properly.

Find out what to include and what to avoid, so you can effectively demonstrate your understanding of the topic and prove your expertise.

• Why is a good conclusion important?

A good conclusion can cement your paper in the reader’s mind. Making a strong impression in your introduction can draw your readers in, but it's the conclusion that will inspire them.

• What to include in a research paper conclusion

There are a few specifics you should include in your research paper conclusion. Offer your readers some sense of urgency or consequence by pointing out why they should care about the topic you have covered. Discuss any common problems associated with your topic and provide suggestions as to how these problems can be solved or addressed.

The conclusion should include a restatement of your initial thesis. Thesis statements are strengthened after you’ve presented supporting evidence (as you will have done in the paper), so make a point to reintroduce it at the end.

Finally, recap the main points of your research paper, highlighting the key takeaways you want readers to remember. If you've made multiple points throughout the paper, refer to the ones with the strongest supporting evidence.

• Steps for writing a research paper conclusion

Many writers find the conclusion the most challenging part of any research project . By following these three steps, you'll be prepared to write a conclusion that is effective and concise.

• Step 1: Restate the problem

Always begin by restating the research problem in the conclusion of a research paper. This serves to remind the reader of your hypothesis and refresh them on the main point of the paper. 

When restating the problem, take care to avoid using exactly the same words you employed earlier in the paper.

• Step 2: Sum up the paper

After you've restated the problem, sum up the paper by revealing your overall findings. The method for this differs slightly, depending on whether you're crafting an argumentative paper or an empirical paper.

Argumentative paper: Restate your thesis and arguments

Argumentative papers involve introducing a thesis statement early on. In crafting the conclusion for an argumentative paper, always restate the thesis, outlining the way you've developed it throughout the entire paper.

It might be appropriate to mention any counterarguments in the conclusion, so you can demonstrate how your thesis is correct or how the data best supports your main points.

Empirical paper: Summarize research findings

Empirical papers break down a series of research questions. In your conclusion, discuss the findings your research revealed, including any information that surprised you.

Be clear about the conclusions you reached, and explain whether or not you expected to arrive at these particular ones.

• Step 3: Discuss the implications of your research

Argumentative papers and empirical papers also differ in this part of a research paper conclusion. Here are some tips on crafting conclusions for argumentative and empirical papers.

Argumentative paper: Powerful closing statement

In an argumentative paper, you'll have spent a great deal of time expressing the opinions you formed after doing a significant amount of research. Make a strong closing statement in your argumentative paper's conclusion to share the significance of your work.

You can outline the next steps through a bold call to action, or restate how powerful your ideas turned out to be.

Empirical paper: Directions for future research

Empirical papers are broader in scope. They usually cover a variety of aspects and can include several points of view.

To write a good conclusion for an empirical paper, suggest the type of research that could be done in the future, including methods for further investigation or outlining ways other researchers might proceed.

If you feel your research had any limitations, even if they were outside your control, you could mention these in your conclusion.

After you finish outlining your conclusion, ask someone to read it and offer feedback. In any research project you're especially close to, it can be hard to identify problem areas. Having a close friend or someone whose opinion you value read the research paper and provide honest feedback can be invaluable. Take note of any suggested edits and consider incorporating them into your paper if they make sense.

• Things to avoid in a research paper conclusion

Keep these aspects to avoid in mind as you're writing your conclusion and refer to them after you've created an outline.

Dry summary

Writing a memorable, succinct conclusion is arguably more important than a strong introduction. Take care to avoid just rephrasing your main points, and don't fall into the trap of repeating dry facts or citations.

You can provide a new perspective for your readers to think about or contextualize your research. Either way, make the conclusion vibrant and interesting, rather than a rote recitation of your research paper’s highlights.

Clichéd or generic phrasing

Your research paper conclusion should feel fresh and inspiring. Avoid generic phrases like "to sum up" or "in conclusion." These phrases tend to be overused, especially in an academic context and might turn your readers off.

The conclusion also isn't the time to introduce colloquial phrases or informal language. Retain a professional, confident tone consistent throughout your paper’s conclusion so it feels exciting and bold.

New data or evidence

While you should present strong data throughout your paper, the conclusion isn't the place to introduce new evidence. This is because readers are engaged in actively learning as they read through the body of your paper.

By the time they reach the conclusion, they will have formed an opinion one way or the other (hopefully in your favor!). Introducing new evidence in the conclusion will only serve to surprise or frustrate your reader.

Ignoring contradictory evidence

If your research reveals contradictory evidence, don't ignore it in the conclusion. This will damage your credibility as an expert and might even serve to highlight the contradictions.

Be as transparent as possible and admit to any shortcomings in your research, but don't dwell on them for too long.

Ambiguous or unclear resolutions

The point of a research paper conclusion is to provide closure and bring all your ideas together. You should wrap up any arguments you introduced in the paper and tie up any loose ends, while demonstrating why your research and data are strong.

Use direct language in your conclusion and avoid ambiguity. Even if some of the data and sources you cite are inconclusive or contradictory, note this in your conclusion to come across as confident and trustworthy.

• Examples of research paper conclusions

Your research paper should provide a compelling close to the paper as a whole, highlighting your research and hard work. While the conclusion should represent your unique style, these examples offer a starting point:

Ultimately, the data we examined all point to the same conclusion: Encouraging a good work-life balance improves employee productivity and benefits the company overall. The research suggests that when employees feel their personal lives are valued and respected by their employers, they are more likely to be productive when at work. In addition, company turnover tends to be reduced when employees have a balance between their personal and professional lives. While additional research is required to establish ways companies can support employees in creating a stronger work-life balance, it's clear the need is there.

Social media is a primary method of communication among young people. As we've seen in the data presented, most young people in high school use a variety of social media applications at least every hour, including Instagram and Facebook. While social media is an avenue for connection with peers, research increasingly suggests that social media use correlates with body image issues. Young girls with lower self-esteem tend to use social media more often than those who don't log onto social media apps every day. As new applications continue to gain popularity, and as more high school students are given smartphones, more research will be required to measure the effects of prolonged social media use.

What are the different kinds of research paper conclusions?

There are no formal types of research paper conclusions. Ultimately, the conclusion depends on the outline of your paper and the type of research you’re presenting. While some experts note that research papers can end with a new perspective or commentary, most papers should conclude with a combination of both. The most important aspect of a good research paper conclusion is that it accurately represents the body of the paper.

Can I present new arguments in my research paper conclusion?

Research paper conclusions are not the place to introduce new data or arguments. The body of your paper is where you should share research and insights, where the reader is actively absorbing the content. By the time a reader reaches the conclusion of the research paper, they should have formed their opinion. Introducing new arguments in the conclusion can take a reader by surprise, and not in a positive way. It might also serve to frustrate readers.

How long should a research paper conclusion be?

There's no set length for a research paper conclusion. However, it's a good idea not to run on too long, since conclusions are supposed to be succinct. A good rule of thumb is to keep your conclusion around 5 to 10 percent of the paper's total length. If your paper is 10 pages, try to keep your conclusion under one page.

What should I include in a research paper conclusion?

A good research paper conclusion should always include a sense of urgency, so the reader can see how and why the topic should matter to them. You can also note some recommended actions to help fix the problem and some obstacles they might encounter. A conclusion should also remind the reader of the thesis statement, along with the main points you covered in the paper. At the end of the conclusion, add a powerful closing statement that helps cement the paper in the mind of the reader.

Should you be using a customer insights hub?

Do you want to discover previous research faster?

Do you share your research findings with others?

Do you analyze research data?

Start for free today, add your research, and get to key insights faster

Editor’s picks

Last updated: 18 April 2023

Last updated: 27 February 2023

Last updated: 22 August 2024

Last updated: 5 February 2023

Last updated: 16 August 2024

Last updated: 9 March 2023

Last updated: 30 April 2024

Last updated: 12 December 2023

Last updated: 11 March 2024

Last updated: 4 July 2024

Last updated: 6 March 2024

Last updated: 5 March 2024

Last updated: 13 May 2024

Latest articles

Related topics, .css-je19u9{-webkit-align-items:flex-end;-webkit-box-align:flex-end;-ms-flex-align:flex-end;align-items:flex-end;display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-flex-direction:row;-ms-flex-direction:row;flex-direction:row;-webkit-box-flex-wrap:wrap;-webkit-flex-wrap:wrap;-ms-flex-wrap:wrap;flex-wrap:wrap;-webkit-box-pack:center;-ms-flex-pack:center;-webkit-justify-content:center;justify-content:center;row-gap:0;text-align:center;max-width:671px;}@media (max-width: 1079px){.css-je19u9{max-width:400px;}.css-je19u9>span{white-space:pre;}}@media (max-width: 799px){.css-je19u9{max-width:400px;}.css-je19u9>span{white-space:pre;}} decide what to .css-1kiodld{max-height:56px;display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}@media (max-width: 1079px){.css-1kiodld{display:none;}} build next, decide what to build next, log in or sign up.

Get started for free

Undergraduate Research Experiences for STEM Students: Successes, Challenges, and Opportunities (2017)

Chapter: 9 conclusions and recommendations, 9 conclusions and recommendations.

Practitioners designing or improving undergraduate research experiences (UREs) can build on the experiences of colleagues and learn from the increasingly robust literature about UREs and the considerable body of evidence about how students learn. The questions practitioners ask themselves during the design process should include questions about the goals of the campus, program, faculty, and students. Other factors to consider when designing a URE include the issues raised in the conceptual framework for learning and instruction, the available resources, how the program or experience will be evaluated or studied, and how to design the program from the outset to incorporate these considerations, as well as how to build in opportunities to improve the experience over time in light of new evidence. (Some of these topics are addressed in Chapter 8 .)

Colleges and universities that offer or wish to offer UREs to their students should undertake baseline evaluations of their current offerings and create plans to develop a culture of improvement in which faculty are supported in their efforts to continuously refine UREs based on the evidence currently available and evidence that they and others generate in the future. While much of the evidence to date is descriptive, it forms a body of knowledge that can be used to identify research questions about UREs, both those designed around the apprenticeship model and those designed using the more recent course-based undergraduate research experience (CURE) model. Internships and other avenues by which undergraduates do research provide many of the same sorts of experiences but are not well studied. In any case, it is clear that students value these experiences; that many faculty do as well; and that they contribute to broadening participation in science,

technology, engineering, and mathematics (STEM) education and careers. The findings from the research literature reported in Chapter 4 provide guidance to those designing both opportunities to improve practical and academic skills and opportunities for students to “try out” a professional role of interest.

Little research has been done that provides answers to mechanistic questions about how UREs work. Additional studies are needed to know which features of UREs are most important for positive outcomes with which students and to gain information about other questions of this type. This additional research is needed to better understand and compare different strategies for UREs designed for a diversity of students, mentors, and institutions. Therefore, the committee recommends steps that could increase the quantity and quality of evidence available in the future and makes recommendations for how faculty, departments, and institutions might approach decisions about UREs using currently available information. Multiple detailed recommendations about the kinds of research that might be useful are provided in the research agenda in Chapter 7 .

In addition to the specific research recommended in Chapter 7 , in this chapter the committee provides a series of interrelated conclusions and recommendations related to UREs for the STEM disciplines and intended to highlight the issues of primary importance to administrators, URE program designers, mentors to URE students, funders of UREs, those leading the departments and institutions offering UREs, and those conducting research about UREs. These conclusions and recommendations are based on the expert views of the committee and informed by their review of the available research, the papers commissioned for this report, and input from presenters during committee meetings. Table 9-1 defines categories of these URE “actors,” gives examples of specific roles included in each category, specifies key URE actions for which that category is responsible, and lists the conclusions and recommendations the committee views as most relevant to that actor category.

RESEARCH ON URES

Conclusion 1: The current and emerging landscape of what constitutes UREs is diverse and complex. Students can engage in STEM-based undergraduate research in many different ways, across a variety of settings, and along a continuum that extends and expands upon learning opportunities in other educational settings. The following characteristics define UREs. Due to the variation in the types of UREs, not all experiences include all of the following characteristics in the same way; experiences vary in how much a particular characteristic is emphasized.

TABLE 9-1 Audiences for Committee’s Conclusions and Recommendations

• They engage students in research practices including the ability to argue from evidence.
• They aim to generate novel information with an emphasis on discovery and innovation or to determine whether recent preliminary results can be replicated.
• They focus on significant, relevant problems of interest to STEM researchers and, in some cases, a broader community (e.g., civic engagement).
• They emphasize and expect collaboration and teamwork.
• They involve iterative refinement of experimental design, experimental questions, or data obtained.
• They allow students to master specific research techniques.
• They help students engage in reflection about the problems being investigated and the work being undertaken to address those problems.
• They require communication of results, either through publication or presentations in various STEM venues.
• They are structured and guided by a mentor, with students assuming increasing ownership of some aspects of the project over time.

UREs are generally designed to add value to STEM offerings by promoting an understanding of the ways that knowledge is generated in STEM fields and to extend student learning beyond what happens in the small group work of an inquiry-based course. UREs add value by enabling students to understand and contribute to the research questions that are driving the field for one or more STEM topics or to grapple with design challenges of interest to professionals. They help students understand what it means to be a STEM researcher in a way that would be difficult to convey in a lecture course or even in an inquiry-based learning setting. As participants in a URE, students can learn by engaging in planning, experimentation, evaluation, interpretation, and communication of data and other results in light of what is already known about the question of interest. They can pose relevant questions that can be solved only through investigative or design efforts—individually or in teams—and attempt to answer these questions despite the challenges, setbacks, and ambiguity of the process and the results obtained.

The diversity of UREs reflects the reality that different STEM disciplines operate from varying traditions, expectations, and constraints (e.g., lab safety issues) in providing opportunities for undergraduates to engage in research. In addition, individual institutions and departments have cultures that promote research participation to various degrees and at different stages in students’ academic careers. Some programs emphasize design and problem solving in addition to discovery. UREs in different disciplines can

take many forms (e.g., apprentice-style, course-based, internships, project-based), but the definitional characteristics described above are similar across different STEM fields.

Furthermore, students in today’s university landscape may have opportunities to engage with many different types of UREs throughout their education, including involvement in a formal program (which could include mentoring, tutoring, research, and seminars about research), an apprentice-style URE under the guidance of an individual or team of faculty members, an internship, or enrolling in one or more CUREs or in a consortium- or project-based program.

Conclusion 2: Research on the efficacy of UREs is still in the early stages of development compared with other interventions to improve undergraduate STEM education.

• The types of UREs are diverse, and their goals are even more diverse. Questions and methodologies used to investigate the roles and effectiveness of UREs in achieving those goals are similarly diverse.
• Most of the studies of UREs to date are descriptive case studies or use correlational designs. Many of these studies report positive outcomes from engagement in a URE.
• Only a small number of studies have employed research designs that can support inferences about causation. Most of these studies find evidence for a causal relationship between URE participation and subsequent persistence in STEM. More studies are needed to provide evidence that participation in UREs is a causal factor in a range of desired student outcomes.

Taking the entire body of evidence into account, the committee concludes that the published peer-reviewed literature to date suggests that participation in a URE is beneficial for students .

As discussed in the report’s Introduction (see Chapter 1 ) and in the research agenda (see Chapter 7 ), the committee considered descriptive, causal, and mechanistic questions in our reading of the literature on UREs. Scientific approaches to answering descriptive, causal, and mechanistic questions require deciding what to look for, determining how to examine it, and knowing appropriate ways to score or quantify the effect.

Descriptive questions ask what is happening without making claims as to why it is happening—that is, without making claims as to whether the research experience caused these changes. A descriptive statement about UREs only claims that certain changes occurred during or after the time the students were engaged in undergraduate research. Descriptive studies

cannot determine whether any benefits observed were caused by participation in the URE.

Causal questions seek to discover whether a specific intervention leads to a specific outcome, other things being equal. To address such questions, causal evidence can be generated from a comparison of carefully selected groups that do and do not experience UREs. The groups can be made roughly equivalent by random assignment (ensuring that URE and non-URE groups are the same on average as the sample size increases) or by controlling for an exhaustive set of characteristics and experiences that might render the groups different prior to the URE. Other quasi-experimental strategies can also be used. Simply comparing students who enroll in a URE with students who do not is not adequate for determining causality because there may be selection bias. For example, students already interested in STEM are more likely to seek out such opportunities and more likely to be selected for such programs. Instead the investigator would have to compare future enrollment patterns (or other measures) between closely matched students, some of whom enrolled in a URE and some of whom did not. Controlling for selection bias to enable an inference about causation can pose significant challenges.

Questions of mechanism or of process also can be explored to understand why a causal intervention leads to the observed effect. Perhaps the URE enhances a student’s confidence in her ability to succeed in her chosen field or deepens her commitment to the field by exposing her to the joy of discovery. Through these pathways that act on the participant’s purposive behavior, the URE enhances the likelihood that she persists in STEM. The question for the researcher then becomes what research design would provide support for this hypothesis of mechanism over other candidate explanations for why the URE is a causal factor in STEM persistence.

The committee has examined the literature and finds a rich descriptive foundation for testable hypotheses about the effects of UREs on student outcomes. These studies are encouraging; a few of them have generated evidence that a URE can be a positive causal factor in the progression and persistence of STEM students. The weight of the evidence has been descriptive; it relies primarily on self-reports of short-term gains by students who chose to participate in UREs and does not include direct measures of changes in the students’ knowledge, skills, or other measures of success across comparable groups of students who did and did not participate in UREs.

While acknowledging the scarcity of strong causal evidence on the benefits of UREs, the committee takes seriously the weight of the descriptive evidence. Many of the published studies of UREs show that students who participate report a range of benefits, such as increased understanding of the research process, encouragement to persist in STEM, and support that helps them sustain their identity as researchers and continue with their

plans to enroll in a graduate program in STEM (see Chapter 4 ). These are effective starting points for causal studies.

Conclusion 3: Studies focused on students from historically underrepresented groups indicate that participation in UREs improves their persistence in STEM and helps to validate their disciplinary identity.

Various UREs have been specifically designed to increase the number of historically underrepresented students who go on to become STEM majors and ultimately STEM professionals. While many UREs offer one or more supplemental opportunities to support students’ academic or social success, such as mentoring, tutoring, summer bridge programs, career or graduate school workshops, and research-oriented seminars, those designed for underrepresented students appear to emphasize such features as integral and integrated components of the program. In particular, studies of undergraduate research programs targeting underrepresented minority students have begun to document positive outcomes such as degree completion and persistence in interest in STEM careers ( Byars-Winston et al., 2015 ; Chemers et al., 2011 ; Jones et al., 2010 ; Nagda et al., 1998 ; Schultz et al., 2011 ). Most of these studies collected data on apprentice-style UREs, in which the undergraduate becomes a functioning member of a research group along with the graduate students, postdoctoral fellows, and mentor.

Recommendation 1: Researchers with expertise in education research should conduct well-designed studies in collaboration with URE program directors to improve the evidence base about the processes and effects of UREs. This research should address how the various components of UREs may benefit students. It should also include additional causal evidence for the individual and additive effects of outcomes from student participation in different types of UREs. Not all UREs need be designed to undertake this type of research, but it would be very useful to have some UREs that are designed to facilitate these efforts to improve the evidence base .

As the focus on UREs has grown, so have questions about their implementation. Many articles have been published describing specific UREs (see Chapter 2 ). Large amounts of research have also been undertaken to explore more generally how students learn, and the resulting body of evidence has led to the development and adoption of “active learning” strategies and experiences. If a student in a URE has an opportunity to, for example, analyze new data or to reformulate a hypothesis in light of the student’s analysis, this activity fits into the category that is described as active learning. Surveys of student participants and unpublished evaluations pro-

vide additional information about UREs but do not establish causation or determine the mechanism(s). Consequently, little is currently known about the mechanisms of precisely how UREs work and which aspects of UREs are most powerful. Important components that have been reported include student ownership of the URE project, time to tackle a question iteratively, and opportunities to report and defend one’s conclusions ( Hanauer and Dolan, 2014 ; Thiry et al., 2011 ).

There are many unanswered questions and opportunities for further research into the role and mechanism of UREs. Attention to research design as UREs are planned is important; more carefully designed studies are needed to understand the ways that UREs influence a student’s education and to evaluate the outcomes that have been reported for URE participants. Appropriate studies, which include matched samples or similar controls, would facilitate research on the ways that UREs benefit students, enabling both education researchers and implementers of UREs to determine optimal features for program design and giving the community a more robust understanding of how UREs work.

See the research agenda ( Chapter 7 ) for specific recommendations about research topics and approaches.

Recommendation 2: Funders should provide appropriate resources to support the design, implementation, and analysis of some URE programs that are specifically designed to enable detailed research establishing the effects on participant outcomes and on other variables of interest such as the consequences for mentors or institutions.

Not all UREs need to be the subject of extensive study. In many cases, a straightforward evaluation is adequate to determine whether the URE is meeting its goals. However, to achieve more widespread improvement in both the types and quality of the UREs offered in the future, additional evidence about the possible causal effects and mechanisms of action of UREs needs to be systematically collected and disseminated. This includes a better understanding of the implementation differences for a variety of institutions (e.g., community colleges, primarily undergraduate institutions, research universities) to ensure that the desired outcomes can translate across settings. Increasing the evidence about precisely how UREs work and which aspects of UREs are most powerful will require careful attention to study design during planning for the UREs.

Not all UREs need to be designed to achieve this goal; many can provide opportunities to students by relying on pre-existing knowledge and iterative improvement as that knowledge base grows. However, for the knowledge base to grow, funders must provide resources for some URE designers and social science researchers to undertake thoughtful and well-planned studies

on causal and mechanistic issues. This will maximize the chances for the creation and dissemination of information that can lead to the development of sustainable and effective UREs. These studies can result from a partnership formed as the URE is designed and funded, or evaluators and social scientists could identify promising and/or effective existing programs and then raise funds on their own to support the study of those programs to answer the questions of interest. In deciding upon the UREs that are chosen for these extensive studies, it will be important to consider whether, collectively, they are representative of UREs in general. For example, large and small UREs at large and small schools targeted at both introductory and advanced students and topics should be studied.

CONSTRUCTION OF URES

Conclusion 4: The committee was unable to find evidence that URE designers are taking full advantage of the information available in the education literature on strategies for designing, implementing, and evaluating learning experiences. STEM faculty members do not generally receive training in interpreting or conducting education research. Partnerships between those with expertise in education research and those with expertise in implementing UREs are one way to strengthen the application of evidence on what works in planning and implementing UREs.

As discussed in Chapters 3 and 4 , there is an extensive body of literature on pedagogy and how people learn; helping STEM faculty to access the existing literature and incorporate those concepts as they design UREs could improve student experiences. New studies that specifically focus on UREs may provide more targeted information that could be used to design, implement, sustain, or scale up UREs and facilitate iterative improvements. Information about the features of UREs that elicit particular outcomes or best serve certain populations of students should be considered when implementing a new instantiation of an existing model of a URE or improving upon an existing URE model.

Conclusion 5: Evaluations of UREs are often conducted to inform program providers and funders; however, they may not be accessible to others. While these evaluations are not designed to be research studies and often have small sample sizes, they may contain information that could be useful to those initiating new URE programs and those refining UREs. Increasing access to these evaluations and to the accumulated experience of the program providers may enable URE designers and implementers to build upon knowledge gained from earlier UREs.

As discussed in Chapter 1 , the committee searched for evaluations of URE programs in several different ways but was not able to locate many published evaluations to study. Although some evaluations were found in the literature, the committee could not determine a way to systematically examine the program evaluations that have been prepared. The National Science Foundation and other funders generally require grant recipients to submit evaluation data, but that information is not currently aggregated and shared publicly, even for programs that are using a common evaluation tool. 1

Therefore, while program evaluation likely serves a useful role in providing descriptive data about a program for the institutions and funders supporting the program, much of the summative evaluation work that has been done to date adds relatively little to the broader knowledge base and overall conversations around undergraduate research. Some of the challenges of evaluation include budget and sample size constraints.

Similarly, it is difficult for designers of UREs to benefit systematically from the work of others who have designed and run UREs in the past because of the lack of an easy and consistent mechanism for collecting, analyzing, and sharing data. If these evaluations were more accessible they might be beneficial to others designing and evaluating UREs by helping them to gather ideas and inspiration from the experiences of others. A few such stories are provided in this report, and others can be found among the many resources offered by the Council on Undergraduate Research 2 and on other websites such as CUREnet. 3

Recommendation 3: Designers of UREs should base their design decisions on sound evidence. Consultations with education and social science researchers may be helpful as designers analyze the literature and make decisions on the creation or improvement of UREs. Professional development materials should be created and made available to faculty. Educational and disciplinary societies should consider how they can provide resources and connections to those working on UREs.

Faculty and other organizers of UREs can use the expanding body of scholarship as they design or improve the programs and experiences offered to their students. URE designers will need to make decisions about how to adapt approaches reported in the literature to make the programs they develop more suitable to their own expertise, student population(s), and available resources. Disciplinary societies and other national groups, such as those focused on improving pedagogy, can play important roles in

___________________

1 Personal knowledge of Janet Branchaw, member of the Committee on Strengthening Research Experiences for Undergraduate STEM Students.

2 See www.cur.org [November 2016].

3 See ( curenet.cns.utexas.edu ) [November 2016].

bringing these issues to the forefront through events at their national and regional meetings and through publications in their journals and newsletters. They can develop repositories for various kinds of resources appropriate for their members who are designing and implementing UREs. The ability to travel to conferences and to access and discuss resources created by other individuals and groups is a crucial aspect of support (see Recommendations 7 and 8 for further discussion).

See Chapter 8 for specific questions to consider when one is designing or implementing UREs.

CURRENT OFFERINGS

Conclusion 6: Data at the institutional, state, or national levels on the number and type of UREs offered, or who participates in UREs overall or at specific types of institutions, have not been collected systematically. Although the committee found that some individual institutions track at least some of this type of information, we were unable to determine how common it is to do so or what specific information is most often gathered.

There is no one central database or repository that catalogs UREs at institutions of higher education, the nature of the research experiences they provide, or the relevant demographics (student, departmental, and institutional). The lack of comprehensive data makes it difficult to know how many students participate in UREs; where UREs are offered; and if there are gaps in access to UREs across different institutional types, disciplines, or groups of students. One of the challenges of describing the undergraduate research landscape is that students do not have to be enrolled in a formal program to have a research experience. Informal experiences, for example a work-study job, are typically not well documented. Another challenge is that some students participate in CUREs or other research experiences (such as internships) that are not necessarily labeled as such. Institutional administrators may be unaware of CUREs that are already part of their curriculum. (For example, establishment of CUREs may be under the purview of a faculty curriculum committee and may not be recognized as a distinct program.) Student participation in UREs may occur at their home institution or elsewhere during the summer. Therefore, it is very difficult for a science department, and likely any other STEM department, to know what percentage of their graduating majors have had a research experience, let alone to gather such information on students who left the major. 4

4 This point was made by Marco Molinaro, University of California, Davis, in a presentation to the Committee on Strengthening Research Experience for Undergraduate STEM Students, September 16, 2015.

Conclusion 7: While data are lacking on the precise number of students engaged in UREs, there is some evidence of a recent growth in course-based undergraduate research experiences (CUREs), which engage a cohort of students in a research project as part of a formal academic experience.

There has been an increase in the number of grants and the dollar amount spent on CUREs over the past decade (see Chapter 3 ). CUREs can be particularly useful in scaling UREs to reach a much larger population of students ( Bangera and Brownell, 2014 ). By using a familiar mechanism—enrollment in a course—a CURE can provide a more comfortable route for students unfamiliar with research to gain their first experience. CUREs also can provide such experiences to students with diverse backgrounds, especially if an institution or department mandates participation sometime during a student’s matriculation. Establishing CUREs may be more cost-effective at schools with little on-site research activity. However, designing a CURE is a new and time-consuming challenge for many faculty members. Connecting to nationally organized research networks can provide faculty with helpful resources for the development of a CURE based around their own research or a local community need, or these networks can link interested faculty to an ongoing collaborative project. Collaborative projects can provide shared curriculum, faculty professional development and community, and other advantages when starting or expanding a URE program. See the discussion in the report from a convocation on Integrating Discovery-based Research into the Undergraduate Curriculum ( National Academies of Sciences, Engineering, and Medicine, 2015 ).

Recommendation 4: Institutions should collect data on student participation in UREs to inform their planning and to look for opportunities to improve quality and access.

Better tracking of student participation could lead to better assessment of outcomes and improved quality of experience. Such metrics could be useful for both prospective students and campus planners. An integrated institutional system for research opportunities could facilitate the creation of tiered research experiences that allow students to progress in skills and responsibility and create support structures for students, providing, for example, seminars in communications, safety, and ethics for undergraduate researchers. Institutions could also use these data to measure the impact of UREs on student outcomes, such as student success rates in introductory courses, retention in STEM degree programs, and completion of STEM degrees.

While individual institutions may choose to collect additional information depending on their goals and resources, relevant student demographics

and the following design elements would provide baseline data. At a minimum, such data should include

• Type of URE;
• Each student’s discipline;
• Duration of the experience;
• Hours spent per week;
• When the student began the URE (e.g., first year, capstone);
• Compensation status (e.g., paid, unpaid, credit); and
• Location and format (e.g., on home campus, on another campus, internship, co-op).

National aggregation of some of the student participation variables collected by various campuses might be considered by funders. The existing Integrated Postsecondary Education Data System database, organized by the National Center for Education Statistics at the U.S. Department of Education, may be a suitable repository for certain aspects of this information.

Recommendation 5: Administrators and faculty at all types of colleges and universities should continually and holistically evaluate the range of UREs that they offer. As part of this process, institutions should:

• Consider how best to leverage available resources (including off-campus experiences available to students and current or potential networks or partnerships that the institution may form) when offering UREs so that they align with their institution’s mission and priorities;
• Consider whether current UREs are both accessible and welcoming to students from various subpopulations across campus (e.g., historically underrepresented students, first generation college students, those with disabilities, non-STEM majors, prospective kindergarten-through-12th-grade teachers); and
• Gather and analyze data on the types of UREs offered and the students who participate, making this information widely available to the campus community and using it to make evidence-based decisions about improving opportunities for URE participation. This may entail devising or implementing systems for tracking relevant data (see Conclusion 4 ).

Resources available for starting, maintaining, and expanding UREs vary from campus to campus. At some campuses, UREs are a central focus and many resources are devoted to them. At other institutions—for example, many community colleges—UREs are seen as extra, and new resources may be required to ensure availability of courses and facilities. Resource-

constrained institutions may need to focus more on ensuring that students are aware of potential UREs that already exist on campus and elsewhere in near proximity to campus. All institutional discussions about UREs must consider both the financial resources and physical resources (e.g., laboratories, field stations, engineering design studios) required, while remembering that faculty time is a crucial resource. The incentives and disincentives for faculty to spend time on UREs are significant. Those institutions with an explicit mission to promote undergraduate research may provide more recognition and rewards to departments and faculty than those with another focus. The culture of the institution with respect to innovation in pedagogy and support for faculty development also can have a major influence on the extent to which UREs are introduced or improved.

Access to UREs may vary across campus and by department, and participation in UREs may vary across student groups. It is important for campuses to consider the factors that may facilitate or discourage students from participation in UREs. Inconsistent procedures or a faculty preference for students with high grades or previous research experience may limit options for some student populations.

UREs often grow based on the initiative of individual faculty members and other personnel, and an institution may not have complete or even rudimentary knowledge of all of the opportunities available or whether there are gaps or inconsistencies in its offerings. A uniform method for tracking the UREs available on a given campus would be useful to students and would provide a starting point for analyzing the options. Tracking might consist of notations in course listings and, where feasible, on student transcripts. Analysis might consider the types of UREs offered, the resources available to each type of URE, and variations within or between various disciplines and programs. Attention to whether all students or groups of students have appropriate access to UREs would foster consideration of how to best allocate resources and programming on individual campuses, in order to focus resources and opportunities where they are most needed.

Conclusion 8: The quality of mentoring can make a substantial difference in a student’s experiences with research. However, professional development in how to be a good mentor is not available to many faculty or other prospective mentors (e.g., graduate students, postdoctoral fellows).

Engagement in quality mentored research experiences has been linked to self-reported gains in research skills and productivity as well as retention in STEM (see Chapter 5 ). Quality mentoring in UREs has been shown

to increase persistence in STEM for historically underrepresented students ( Hernandez et al., 2016 ). In addition, poor mentoring during UREs has been shown to decrease retention of students ( Hernandez et al., 2016 ).

More general research on good mentoring in the STEM environment has been positively associated with self-reported gains in identity as a STEM researcher, a sense of belonging, and confidence to function as a STEM researcher ( Byars-Winston et al., 2015 ; Chemers et al., 2011 ; Pfund et al., 2016 ; Thiry et al., 2011 ). The frequency and quality of mentee-mentor interactions has been associated with students’ reports of persistence in STEM, with mentoring directly or indirectly improving both grades and persistence in college. For students from historically underrepresented ethnic/racial groups, quality mentoring has been associated with self-reported enhanced recruitment into graduate school and research-related career pathways ( Byars-Winston et al., 2015 ). Therefore, it is important to ensure that faculty and mentors receive the proper development of mentoring skills.

Recommendation 6: Administrators and faculty at colleges and universities should ensure that all who mentor undergraduates in research experiences (this includes faculty, instructors, postdoctoral fellows, graduate students, and undergraduates serving as peer mentors) have access to appropriate professional development opportunities to help them grow and succeed in this role.

Although many organizations recognize effective mentors (e.g., the National Science Foundation’s Presidential Awards for Excellence in Science, Mathematics, and Engineering Mentoring), there currently are no standard criteria for selecting, evaluating, or recognizing mentors specifically for UREs. In addition, there are no requirements that mentors meet some minimum level of competency before engaging in mentoring or participate in professional development to obtain a baseline of knowledge and skills in mentoring, including cultural competence in mentoring diverse groups of students. Traditionally, the only experience required for being a mentor is having been mentored, regardless of whether the experience was negative or positive ( Handelsman et al., 2005 ; Pfund et al., 2015 ). Explicit consideration of how the relationships are formed, supported, and evaluated can improve mentor-mentee relationships. To ensure that the mentors associated with a URE are prepared appropriately, thereby increasing the chances of a positive experience for both mentors and mentees, all prospective mentors should prepare for their role. Available resources include the Entering Mentoring course (see Pfund et al., 2015 ) and the book Successful STEM Mentoring Initiative for Underrepresented Students ( Packard, 2016 ).

A person who is an ineffective mentor for one student might be inspiring for another, and the setting in which the mentoring takes place (e.g., a CURE or apprentice-style URE, a laboratory or field-research environment) may also influence mentor effectiveness. Thus, there should be some mechanism for monitoring such relationships during the URE, or there should be opportunity for a student who is unhappy with the relationship to seek other mentors. Indeed, cultivating a team of mentors with different experiences and expertise may be the best strategy for any student. A parallel volume to the Entering Mentoring curriculum mentioned above, Entering Research Facilitator’s Manual ( Branchaw et al., 2010 ), is designed to help students with their research mentor-mentee relationships and to coach them on building teams of mentors to guide them. As mentioned in Chapter 5 , the Entering Research curriculum also contains information designed to support a group of students as they go through their first apprentice-style research experience, each working in separate research groups and also meeting together as a cohort focused on learning about research.

PRIORITIES FOR THE FUTURE

Conclusion 9: The unique assets, resources, priorities, and constraints of the department and institution, in addition to those of individual mentors, impact the goals and structures of UREs. Schools across the country are showing considerable creativity in using unique resources, repurposing current assets, and leveraging student enthusiasm to increase research opportunities for their students.

Given current calls for UREs and the growing conversation about their benefits, an increasing number of two- and four-year colleges and universities are increasing their efforts to support undergraduate research. Departments, institutions, and individual faculty members influence the precise nature of UREs in multiple ways and at multiple levels. The physical resources available, including laboratories, field stations, and engineering design studios and testing facilities, make a difference, as does the ability to access resources in the surrounding community (including other parts of the campus). Institutions with an explicit mission to promote undergraduate research may provide more time, resources (e.g., financial, support personnel, space, equipment), and recognition and rewards to departments and faculty in support of UREs than do institutions without that mission. The culture of the institution with respect to innovation in pedagogy and support for faculty development also affects the extent to which UREs are introduced or improved.

Development of UREs requires significant time and effort. Whether or not faculty attempt to implement UREs can depend on whether departmental

or institutional reward and recognition systems compensate for or even recognize the time required to initiate and implement them. The availability of national consortia can help to alleviate many of the time and logistical problems but not those obstacles associated with recognition and resources.

It will be harder for faculty to find the time to develop UREs at institutions where they are required to teach many courses per semester, although in some circumstances faculty can teach CUREs that also advance their own research ( Shortlidge et al., 2016 ). Faculty at community colleges generally have the heaviest teaching expectations, little or no expectations or incentives to maintain a research program, limited access to lab or design space or to scientific and engineering journals, and few resources to undertake any kind of a research program. These constraints may limit the extent to which UREs can be offered to the approximately 40 percent of U.S. undergraduates who are enrolled in the nation’s community colleges (which collectively also serve the highest percentage of the nation’s underrepresented students). 5

Recommendation 7: Administrators and faculty at all types of colleges and universities should work together within and, where feasible, across institutions to create a culture that supports the development of evidence-based, iterative, and continuous refinement of UREs, in an effort to improve student learning outcomes and overall academic success. This should include the development, evaluation, and revision of policies and practices designed to create a culture supportive of the participation of faculty and other mentors in effective UREs. Policies should consider pedagogy, professional development, cross-cultural awareness, hiring practices, compensation, promotion (incentives, rewards), and the tenure process.

Colleges and universities that would like to expand or improve the UREs offered to their students should consider the campus culture and climate and the incentives that affect faculty choices. Those campuses that cultivate an environment supportive of the iterative and continuous refinement of UREs and that offer incentives for evaluation and evidence-based improvement of UREs seem more likely to sustain successful programs. Faculty and others who develop and implement UREs need support to be able to evaluate their courses or programs and to analyze evidence to make decisions about URE design. This kind of support may be fostered by expanding the mission of on-campus centers for learning and teaching to focus more on UREs or by providing incentives for URE developers from the natural sciences and engineering to collaborate with colleagues in the social sciences or colleges of education with expertise in designing studies

5 See http://nces.ed.gov/programs/coe/indicator_cha.asp [November 2016].

involving human subjects. Supporting closer communication between URE developers and the members of the campus Institutional Review Board may help projects to move forward more seamlessly. Interdepartmental and intercampus connections (especially those between two- and four-year institutions) can be valuable for linking faculty with the appropriate resources, colleagues, and diverse student populations. Faculty who have been active in professional development on how students learn in the classroom may have valuable experiences and expertise to share.

The refinement or expansion of UREs should build on evidence from data on student participation, pedagogy, and outcomes, which are integral components of the original design. As UREs are validated and refined, institutions should make efforts to facilitate connections among different departments and disciplines, including the creation of multidisciplinary UREs. Student engagement in learning in general, and with UREs more specifically, depends largely on the culture of the department and the institution and on whether students see their surroundings as inclusive and energetic places to learn and thrive. A study that examined the relationship between campus missions and the five benchmarks for effective educational practice (measured by the National Survey of Student Engagement) showed that different programs, policies, and approaches may work better, depending on the institution’s mission ( Kezar and Kinzie, 2006 ).

The Council on Undergraduate Research (2012) document Characteristics of Excellence in Undergraduate Research outlines several best practices for UREs based on the apprenticeship model (see Chapter 8 ). That document is not the result of a detailed analysis of the evidence but is based on the extensive experiences and expertise of the council’s members. It suggests that undergraduate research should be a normal part of the undergraduate experience regardless of the type of institution. It also identifies changes necessary to include UREs as part of the curriculum and culture changes necessary to support curricular reform, co-curricular activities, and modifications to the incentives and rewards for faculty to engage with undergraduate research. In addition, professional development opportunities specifically designed to help improve the pedagogical and mentoring skills of instructional staff in using evidence-based practices can be important for a supportive learning culture.

Recommendation 8: Administrators and faculty at all types of colleges and universities should work to develop strong and sustainable partnerships within and between institutions and with educational and professional societies for the purpose of sharing resources to facilitate the creation of sustainable URE programs.

Networks of faculty, institutions, regionally and nationally coordinated URE initiatives, professional societies, and funders should be strengthened

to facilitate the exchange of evidence and experience related to UREs. These networks could build on the existing work of professional societies that assist faculty with pedagogy. They can help provide a venue for considering the policy context and larger implications of increasing the number, size, and scope of UREs. Such networks also can provide a more robust infrastructure, to improve the sustainability and expansion of URE opportunities. The sharing of human, financial, scientific, and technical resources can strengthen the broad implementation of effective, high-quality, and more cost-efficient UREs. It may be especially important for community colleges and minority-serving institutions to engage in partnerships in order to expand the opportunities for undergraduates (both transfer and technical students) to participate in diverse UREs (see discussion in National Academies of Sciences, Engineering, and Medicine, 2015 , and Elgin et al., 2016 ). Consortia can facilitate the sharing of resources across disciplines and departments within the same institution or at different institutions, organizations, and agencies. Consortia that employ research methodologies in common can share curriculum, research data collected, and common assessment tools, lessening the time burden for individual faculty and providing a large pool of students from which to assess the efficacy of individual programs.

Changes in the funding climate can have substantial impacts on the types of programs that exist, iterative refinement of programs, and whether and how programs might be expanded to broaden participation by more undergraduates. For those institutions that have not yet established URE programs or are at the beginning phases of establishing one, mechanisms for achieving success and sustainability may include increased institutional ownership of programs of undergraduate research, development of a broad range of programs of different types and funding structures, formation of undergraduate research offices or repurposing some of the responsibilities and activities of those which already exist, and engagement in community promotion and dissemination of student accomplishments (e.g., student symposia, support for undergraduate student travel to give presentations at professional meetings).

Over time, institutions must develop robust plans for ensuring the long-term sustained funding of high-quality UREs. Those plans should include assuming that more fiscal responsibility for sustaining such efforts will be borne by the home institution as external support for such efforts decreases and ultimately ends. Building UREs into the curriculum and structure of a department’s courses and other programs, and thus its funding model, can help with sustainability. Partnerships with nonprofit organizations and industry, as well as seeking funding from diverse agencies, can also facilitate programmatic sustainability, especially if the UREs they fund can also support the mission and programs of the funders (e.g., through research internships or through CUREs that focus on community-

based research questions and challenges). Partnerships among institutions also may have greater potential to study and evaluate student outcomes from URE participation across broader demographic groups and to reduce overall costs through the sharing of administrative or other resources (such as libraries, microscopes, etc.).

Bangera, G., and Brownell, S.E. (2014). Course-based undergraduate research experiences can make scientific research more inclusive. CBE–Life Sciences Education , 13 (4), 602-606.

Branchaw, J.L., Pfund, C., and Rediske, R. (2010) Entering Research Facilitator’s Manual: Workshops for Students Beginning Research in Science . New York: Freeman & Company.

Byars-Winston, A.M., Branchaw, J., Pfund, C., Leverett, P., and Newton, J. (2015). Culturally diverse undergraduate researchers’ academic outcomes and perceptions of their research mentoring relationships. International Journal of Science Education , 37 (15), 2,533-2,554.

Chemers, M.M., Zurbriggen, E.L., Syed, M., Goza, B.K., and Bearman, S. (2011). The role of efficacy and identity in science career commitment among underrepresented minority students. Journal of Social Issues , 67 (3), 469-491.

Council on Undergraduate Research. (2012). Characteristics of Excellence in Undergraduate Research . Washington, DC: Council on Undergraduate Research.

Elgin, S.C.R., Bangera, G., Decatur, S.M., Dolan, E.L., Guertin, L., Newstetter, W.C., San Juan, E.F., Smith, M.A., Weaver, G.C., Wessler, S.R., Brenner, K.A., and Labov, J.B. 2016. Insights from a convocation: Integrating discovery-based research into the undergraduate curriculum. CBE–Life Sciences Education, 15 , 1-7.

Hanauer, D., and Dolan, E. (2014) The Project Ownership Survey: Measuring differences in scientific inquiry experiences, CBE–Life Sciences Education , 13 , 149-158.

Handelsman, J., Pfund, C., Lauffer, S.M., and Pribbenow, C.M. (2005). Entering Mentoring . Madison, WI: The Wisconsin Program for Scientific Teaching.

Hernandez, P.R., Estrada, M., Woodcock, A., and Schultz, P.W. (2016). Protégé perceptions of high mentorship quality depend on shared values more than on demographic match. Journal of Experimental Education. Available: http://www.tandfonline.com/doi/full/10.1080/00220973.2016.1246405 [November 2016].

Jones, P., Selby, D., and Sterling, S.R. (2010). Sustainability Education: Perspectives and Practice Across Higher Education . New York: Earthscan.

Kezar, A.J., and Kinzie, J. (2006). Examining the ways institutions create student engagement: The role of mission. Journal of College Student Development , 47 (2), 149-172.

National Academies of Sciences, Engineering, and Medicine. (2015). Integrating Discovery-Based Research into the Undergraduate Curriculum: Report of a Convocation . Washington, DC: National Academies Press.

Nagda, B.A., Gregerman, S.R., Jonides, J., von Hippel, W., and Lerner, J.S. (1998). Undergraduate student-faculty research partnerships affect student retention. Review of Higher Education, 22 , 55-72. Available: http://scholar.harvard.edu/files/jenniferlerner/files/nagda_1998_paper.pdf [February 2017].

Packard, P. (2016). Successful STEM Mentoring Initiatives for Underrepresented Students: A Research-Based Guide for Faculty and Administrators . Sterling, VA: Stylus.

Pfund, C., Branchaw, J.L., and Handelsman, J. (2015). Entering Mentoring: A Seminar to Train a New Generation of Scientists (2nd ed). New York: Macmillan Learning.

Pfund, C., Byars-Winston, A., Branchaw, J.L., Hurtado, S., and Eagan, M.K. (2016). Defining attributes and metrics of effective research mentoring relationships. AIDS and Behavior, 20 , 238-248.

Schultz, P.W., Hernandez, P.R., Woodcock, A., Estrada, M., Chance, R.C., Aguilar, M., and Serpe, R.T. (2011). Patching the pipeline reducing educational disparities in the sciences through minority training programs. Educational Evaluation and Policy Analysis , 33 (1), 95-114.

Shortlidge, E.E., Bangera, G., and Brownell, S.E. (2016). Faculty perspectives on developing and teaching course-based undergraduate research experiences. BioScience, 66 (1), 54-62.

Thiry, H., Laursen, S.L., and Hunter, A.B. (2011). What experiences help students become scientists? A comparative study of research and other sources of personal and professional gains for STEM undergraduates. Journal of Higher Education, 82 (4), 358-389.

This page intentionally left blank.

Undergraduate research has a rich history, and many practicing researchers point to undergraduate research experiences (UREs) as crucial to their own career success. There are many ongoing efforts to improve undergraduate science, technology, engineering, and mathematics (STEM) education that focus on increasing the active engagement of students and decreasing traditional lecture-based teaching, and UREs have been proposed as a solution to these efforts and may be a key strategy for broadening participation in STEM. In light of the proposals questions have been asked about what is known about student participation in UREs, best practices in UREs design, and evidence of beneficial outcomes from UREs.

Undergraduate Research Experiences for STEM Students provides a comprehensive overview of and insights about the current and rapidly evolving types of UREs, in an effort to improve understanding of the complexity of UREs in terms of their content, their surrounding context, the diversity of the student participants, and the opportunities for learning provided by a research experience. This study analyzes UREs by considering them as part of a learning system that is shaped by forces related to national policy, institutional leadership, and departmental culture, as well as by the interactions among faculty, other mentors, and students. The report provides a set of questions to be considered by those implementing UREs as well as an agenda for future research that can help answer questions about how UREs work and which aspects of the experiences are most powerful.

READ FREE ONLINE

Welcome to OpenBook!

You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

Do you want to take a quick tour of the OpenBook's features?

Show this book's table of contents , where you can jump to any chapter by name.

...or use these buttons to go back to the previous chapter or skip to the next one.

Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

Switch between the Original Pages , where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

To search the entire text of this book, type in your search term here and press Enter .

Share a link to this book page on your preferred social network or via email.

View our suggested citation for this chapter.

Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

Get Email Updates

Do you enjoy reading reports from the Academies online for free ? Sign up for email notifications and we'll let you know about new publications in your areas of interest when they're released.

• How it works

How To Write Recommendations In A Research Study

Published by Alvin Nicolas at July 12th, 2024 , Revised On July 12, 2024

The ultimate goal of any research process is not just to gather knowledge, but to use that knowledge to make a positive impact. This is where recommendations come in.  A well-written recommendations section in your research study translates your findings into actionable steps and guides future research on the topic. 

This blog is your ultimate guide to understanding how to write recommendations in a research study. But before that, let’s see what is recommendation in research. 

What Is Recommendation In Research 

In a research study, the recommendation section refers to a suggested course of action based on the findings of your research . It acts as a bridge between the knowledge you gained and its practical implications. 

Recommendations take your research results and propose concrete steps on how to use them to address a problem or improve a situation. Moreover, you can suggest new avenues and guide future research in building upon your work. This will improve the credibility of your research. For studies that include real-world implications, recommendations are a great way to provide evidence-based suggestions for policymakers or practitioners to consider. 

Difference Between Research Recommendations and Implication

Research recommendations and implications often confuse researchers. They cannot easily differentiate between the two. Here is how they are different. 

Where To Add Recommendations 

Recommendations are mostly part of your conclusion and discussion sections. If you are writing a practical dissertation , you can include a separate section for your recommendations. 

Types of Research Recommendations

There are different forms of recommendations in research. Some of them include the following. 

How To Construct The Recommendations Section

There are different ways in which different scholars write the recommendations section. A general observation is a research question → conclusion → recommendation.

The following example will help you understand this better.

Research Question

How can the education of mothers impact the social skills of kindergarten children?

The role of mothers is a significant contributor towards the social skills of children. From an early age, kids tend to observe how their mother interacts with others and follow in her footsteps initially. Therefore, mothers should be educated and interact with good demeanour if they want their children to have excellent social skills.

Recommendation

The study revealed that a mother’s education plays an important role in building the social skills of children on kindergarten level. Future research could explore how the same continues in junior school level children.

How To Write Recommendations In Research

Now that you are familiar with the definition and types, here is a step-by-step guide on how to write a recommendation in research.

Step 1: Revisit Your Research Goals

Before doing anything else, you have to remind yourself of the objectives that you set out to achieve in your research. It allows you to match your recommendations directly to your research questions and see if you made any contribution to your goals.

Step 2: Analyse Your Findings

You have to examine your data and identify your key results. This analysis forms the foundation for your recommendations. Look for patterns and unexpected findings that might suggest new areas for other researchers to explore.

Step 3: Consider The Research Methods

Ask these questions from yourself: were the research methods effective? Is there any other way that would have been better to perform this research, or were there any limitations associated with the research methods?

Step 4: Prioritise Recommendations

You might have a lot of recommendations in mind, but all are not equal. You have to consider the impact and feasibility of each suggestion. Prioritise these recommendations, while remaining realistic about implementation.

Step 5: Write Actionable Statements

Do not be vague when crafting statements. Instead, you have to use clear and concise language that outlines specific actions. For example, if you want to say “improve education practices,” you could write “implement a teacher training program” for better clarity.

Step 6: Provide Evidence

You cannot just make suggestions out of thin air, and have to ground them in the evidence you have gathered through your research. Moreover, cite relevant data or findings from your study or previous literature to support your recommendations.

Step 7: Address Challenges

There are always some limitations related to the research at hand. As a researcher, it is your duty to highlight and address any challenges faced or what might occur in the future.

Tips For Writing The Perfect Recommendation In Research

Use these tips to write the perfect recommendation in your research.

• Be Concise – Write recommendations in a clear and concise language. Use one sentence statements to look more professional.
• Be Logical & Coherent – You can use lists and headings according to the requirements of your university.
• Tailor According To Your Readers – You have to aim your recommendations to a specific audience and colleagues in the field of study.
• Provide Specific Suggestions – Offer specific measures and solutions to the issues, and focus on actionable suggestions.
• Match Recommendations To Your Conclusion – You have to align your recommendations with your conclusion.
• Consider Limitations – Use critical thinking to see how limitations may impact the feasibility of your solutions.
• End With A Summary – You have to add a small conclusion to highlight suggestions and their impact.

Example Of Recommendation In Research

Context of the study:

This research studies how effective e-learning platforms are for adult language learners compared to traditional classroom instruction. The findings suggest that e-learning platforms can be just as effective as traditional classrooms in improving language proficiency.

Research Recommendation Sample

Language educators can incorporate e-learning tools into existing curriculums to provide learners with more flexibility. Additionally, they can develop training programs for educators on how to integrate e-learning platforms into their teaching practices.

E-learning platform developers should focus on e-learning platforms that are interactive and cater to different learning styles. They can also invest in features that promote learner autonomy and self-directed learning.

Future researchers can further explore the long-term effects of e-learning on language acquisition to provide insights into whether e-learning can support sustained language development.

Frequently Asked Questions

How to write recommendations in a research paper.

• Revisit your research goals
• Analyse your findings 
• Consider the research methods 
• Prioritise recommendations 
• Write actionable statements 
• Provide evidence 
• Address challenges

How to present recommendations in research?

• Be concise 
• Write logical and coherent 
• Match recommendations to conclusion 
• Ensure your recommendations are achievable

What to write in recommendation in research?

Your recommendation has to be concrete and specific and support the research with a clear rationale. Moreover, it should be connected directly to your research. Your recommendations, however, should not undermine your own work or use self-criticism. 

You May Also Like

Writing a dissertation can be tough if this is the first time you are doing it. You need to look into relevant literature, analyze past researches, conduct surveys, interviews etc.

Table of contents is an essential part of dissertation paper. Here is all you need to know about how to create the best table of contents for dissertation.

Your dissertation introduction chapter provides detailed information on the research problem, significance of research, and research aim & objectives.

USEFUL LINKS

LEARNING RESOURCES

COMPANY DETAILS

• How It Works

Have a language expert improve your writing

Run a free plagiarism check in 10 minutes, generate accurate citations for free.

• Knowledge Base
• Dissertation

How to Write a Thesis or Dissertation Conclusion

Published on September 6, 2022 by Tegan George and Shona McCombes. Revised on November 20, 2023.

The conclusion is the very last part of your thesis or dissertation . It should be concise and engaging, leaving your reader with a clear understanding of your main findings, as well as the answer to your research question .

In it, you should:

• Clearly state the answer to your main research question
• Summarize and reflect on your research process
• Make recommendations for future work on your thesis or dissertation topic
• Show what new knowledge you have contributed to your field
• Wrap up your thesis or dissertation

Instantly correct all language mistakes in your text

Upload your document to correct all your mistakes in minutes

Table of contents

Discussion vs. conclusion, how long should your conclusion be, step 1: answer your research question, step 2: summarize and reflect on your research, step 3: make future recommendations, step 4: emphasize your contributions to your field, step 5: wrap up your thesis or dissertation, full conclusion example, conclusion checklist, other interesting articles, frequently asked questions about conclusion sections.

While your conclusion contains similar elements to your discussion section , they are not the same thing.

Your conclusion should be shorter and more general than your discussion. Instead of repeating literature from your literature review , discussing specific research results , or interpreting your data in detail, concentrate on making broad statements that sum up the most important insights of your research.

As a rule of thumb, your conclusion should not introduce new data, interpretations, or arguments.

Prevent plagiarism. Run a free check.

Depending on whether you are writing a thesis or dissertation, your length will vary. Generally, a conclusion should make up around 5–7% of your overall word count.

An empirical scientific study will often have a short conclusion, concisely stating the main findings and recommendations for future research. A humanities dissertation topic or systematic review , on the other hand, might require more space to conclude its analysis, tying all the previous sections together in an overall argument.

Your conclusion should begin with the main question that your thesis or dissertation aimed to address. This is your final chance to show that you’ve done what you set out to do, so make sure to formulate a clear, concise answer.

• Don’t repeat a list of all the results that you already discussed
• Do synthesize them into a final takeaway that the reader will remember.

An empirical thesis or dissertation conclusion may begin like this:

A case study –based thesis or dissertation conclusion may begin like this:

In the second example, the research aim is not directly restated, but rather added implicitly to the statement. To avoid repeating yourself, it is helpful to reformulate your aims and questions into an overall statement of what you did and how you did it.

Your conclusion is an opportunity to remind your reader why you took the approach you did, what you expected to find, and how well the results matched your expectations.

To avoid repetition , consider writing more reflectively here, rather than just writing a summary of each preceding section. Consider mentioning the effectiveness of your methodology , or perhaps any new questions or unexpected insights that arose in the process.

You can also mention any limitations of your research, but only if you haven’t already included these in the discussion. Don’t dwell on them at length, though—focus on the positives of your work.

• While x limits the generalizability of the results, this approach provides new insight into y .
• This research clearly illustrates x , but it also raises the question of y .

Here's why students love Scribbr's proofreading services

Discover proofreading & editing

You may already have made a few recommendations for future research in your discussion section, but the conclusion is a good place to elaborate and look ahead, considering the implications of your findings in both theoretical and practical terms.

• Based on these conclusions, practitioners should consider …
• To better understand the implications of these results, future studies could address …
• Further research is needed to determine the causes of/effects of/relationship between …

When making recommendations for further research, be sure not to undermine your own work. Relatedly, while future studies might confirm, build on, or enrich your conclusions, they shouldn’t be required for your argument to feel complete. Your work should stand alone on its own merits.

Just as you should avoid too much self-criticism, you should also avoid exaggerating the applicability of your research. If you’re making recommendations for policy, business, or other practical implementations, it’s generally best to frame them as “shoulds” rather than “musts.” All in all, the purpose of academic research is to inform, explain, and explore—not to demand.

Make sure your reader is left with a strong impression of what your research has contributed to the state of your field.

Some strategies to achieve this include:

• Returning to your problem statement to explain how your research helps solve the problem
• Referring back to the literature review and showing how you have addressed a gap in knowledge
• Discussing how your findings confirm or challenge an existing theory or assumption

Again, avoid simply repeating what you’ve already covered in the discussion in your conclusion. Instead, pick out the most important points and sum them up succinctly, situating your project in a broader context.

The end is near! Once you’ve finished writing your conclusion, it’s time to wrap up your thesis or dissertation with a few final steps:

• It’s a good idea to write your abstract next, while the research is still fresh in your mind.
• Next, make sure your reference list is complete and correctly formatted. To speed up the process, you can use our free APA citation generator .
• Once you’ve added any appendices , you can create a table of contents and title page .
• Finally, read through the whole document again to make sure your thesis is clearly written and free from language errors. You can proofread it yourself , ask a friend, or consider Scribbr’s proofreading and editing service .

Here is an example of how you can write your conclusion section. Notice how it includes everything mentioned above:

V. Conclusion

The current research aimed to identify acoustic speech characteristics which mark the beginning of an exacerbation in COPD patients.

The central questions for this research were as follows: 1. Which acoustic measures extracted from read speech differ between COPD speakers in stable condition and healthy speakers? 2. In what ways does the speech of COPD patients during an exacerbation differ from speech of COPD patients during stable periods?

All recordings were aligned using a script. Subsequently, they were manually annotated to indicate respiratory actions such as inhaling and exhaling. The recordings of 9 stable COPD patients reading aloud were then compared with the recordings of 5 healthy control subjects reading aloud. The results showed a significant effect of condition on the number of in- and exhalations per syllable, the number of non-linguistic in- and exhalations per syllable, and the ratio of voiced and silence intervals. The number of in- and exhalations per syllable and the number of non-linguistic in- and exhalations per syllable were higher for COPD patients than for healthy controls, which confirmed both hypotheses.

However, the higher ratio of voiced and silence intervals for COPD patients compared to healthy controls was not in line with the hypotheses. This unpredicted result might have been caused by the different reading materials or recording procedures for both groups, or by a difference in reading skills. Moreover, there was a trend regarding the effect of condition on the number of syllables per breath group. The number of syllables per breath group was higher for healthy controls than for COPD patients, which was in line with the hypothesis. There was no effect of condition on pitch, intensity, center of gravity, pitch variability, speaking rate, or articulation rate.

This research has shown that the speech of COPD patients in exacerbation differs from the speech of COPD patients in stable condition. This might have potential for the detection of exacerbations. However, sustained vowels rarely occur in spontaneous speech. Therefore, the last two outcome measures might have greater potential for the detection of beginning exacerbations, but further research on the different outcome measures and their potential for the detection of exacerbations is needed due to the limitations of the current study.

Checklist: Conclusion

I have clearly and concisely answered the main research question .

I have summarized my overall argument or key takeaways.

I have mentioned any important limitations of the research.

I have given relevant recommendations .

I have clearly explained what my research has contributed to my field.

I have  not introduced any new data or arguments.

You've written a great conclusion! Use the other checklists to further improve your dissertation.

If you want to know more about AI for academic writing, AI tools, or research bias, make sure to check out some of our other articles with explanations and examples or go directly to our tools!

Research bias

• Survivorship bias
• Self-serving bias
• Availability heuristic
• Halo effect
• Hindsight bias
• Deep learning
• Generative AI
• Machine learning
• Reinforcement learning
• Supervised vs. unsupervised learning

 (AI) Tools

• Grammar Checker
• Paraphrasing Tool
• Text Summarizer
• AI Detector
• Plagiarism Checker
• Citation Generator

In a thesis or dissertation, the discussion is an in-depth exploration of the results, going into detail about the meaning of your findings and citing relevant sources to put them in context.

The conclusion is more shorter and more general: it concisely answers your main research question and makes recommendations based on your overall findings.

While it may be tempting to present new arguments or evidence in your thesis or disseration conclusion , especially if you have a particularly striking argument you’d like to finish your analysis with, you shouldn’t. Theses and dissertations follow a more formal structure than this.

All your findings and arguments should be presented in the body of the text (more specifically in the discussion section and results section .) The conclusion is meant to summarize and reflect on the evidence and arguments you have already presented, not introduce new ones.

For a stronger dissertation conclusion , avoid including:

• Important evidence or analysis that wasn’t mentioned in the discussion section and results section
• Generic concluding phrases (e.g. “In conclusion …”)
• Weak statements that undermine your argument (e.g., “There are good points on both sides of this issue.”)

Your conclusion should leave the reader with a strong, decisive impression of your work.

The conclusion of your thesis or dissertation shouldn’t take up more than 5–7% of your overall word count.

The conclusion of your thesis or dissertation should include the following:

• A restatement of your research question
• A summary of your key arguments and/or results
• A short discussion of the implications of your research

Cite this Scribbr article

If you want to cite this source, you can copy and paste the citation or click the “Cite this Scribbr article” button to automatically add the citation to our free Citation Generator.

George, T. & McCombes, S. (2023, November 20). How to Write a Thesis or Dissertation Conclusion. Scribbr. Retrieved August 29, 2024, from https://www.scribbr.com/dissertation/write-conclusion/

Is this article helpful?

Tegan George

Other students also liked, how to write a discussion section | tips & examples, how to write an abstract | steps & examples, how to write a thesis or dissertation introduction, what is your plagiarism score.

Warning: The NCBI web site requires JavaScript to function. more...

An official website of the United States government

The .gov means it's official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• Browse Titles

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

O’Hara R, Johnson M, Hirst E, et al. A qualitative study of decision-making and safety in ambulance service transitions. Southampton (UK): NIHR Journals Library; 2014 Dec. (Health Services and Delivery Research, No. 2.56.)

A qualitative study of decision-making and safety in ambulance service transitions.

Chapter 8 conclusions and recommendations.

The aim of this study was to explore the range and nature of influences on safety in decision-making by ambulance service staff (paramedics). A qualitative approach was adopted using a range of complementary methods. The study has provided insights on the types of decisions that staff engage in on a day-to-day basis. It has also identified a range of system risk factors influencing decisions about patient care. Although this was a relatively small-scale exploratory study, confidence in the generalisability of the headline findings is enhanced by the high level of consistency in the findings, obtained using multiple methods, and the notable consensus among participants.

The seven predominant system influences identified should not be considered discrete but as overlapping and complementary issues. They also embody a range of subthemes that represent topics for future research and/or intervention.

The apparently high level of consistency across the participating trusts suggests that the issues identified may be generic and relevant to other ambulance service trusts.

In view of the remit of this study, aspects relating to system weaknesses and potential threats to patient safety dominate in the account of findings. However, it should be noted that respondent accounts also provided examples of systems that were said to be working well, for example specific care management pathways, local roles and ways of working and technological initiatives such as IBIS and the ePRF.

• Implications for health care

The NHS system within which the ambulance service operates is characterised in our study as fragmented and inconsistent. For ambulance service staff the extent of variation across the geographical areas in which they work is problematic in terms of knowing what services are available and being able to access them. The lack of standardisation in practice guidelines, pathways and protocols across services and between areas makes it particularly challenging for staff to keep up to date with requirements in different parts of their own trust locations and when crossing trust boundaries. Although a degree of consistency across the network is likely to improve the situation, it is also desirable to have sufficient flexibility to accommodate the needs of specific local populations. There was some concern over the potential for further fragmentation with the increased number of CCGs.

Ambulance services are increasingly under pressure to focus on reducing conveyance rates to A&E; this arguably intensifies the need to ensure that crews are appropriately skilled to be able to make effective decisions over the need to convey or not to convey if associated risks to patients are to be minimised. Our findings highlight the challenges of developing staff and ensuring that their skills are utilised where they are most needed within the context of organisational resource constraints and operational demands. Decisions over non-conveyance to A&E are moderated by the availability of alternative care pathways and providers. There were widespread claims of local variability in this respect. Staff training and development, and access to alternatives to A&E, were identified as priorities for attention by workshop attendees.

One of the difficulties for ambulance services is that they operate as a 24/7 service within a wider urgent and emergency care network that, beyond A&E, operates a more restricted working day. The study findings identify this as problematic for two reasons. First, it fuels demand for ambulance service care as a route to timely treatment, when alternatives may involve delay. Second, it contributes to inappropriate conveyance to A&E because more appropriate options are unavailable or limited during out-of-hours periods. Ultimately, this restricts the scope for ensuring that patients are getting the right level of care at the right time and place. Study participants identified some patient populations as particularly poorly served in terms of alternatives to A&E (e.g. those with mental health issues, those at the end of life, older patients and those with chronic conditions).

The effectiveness of the paramedic role in facilitating access to appropriate care pathways hinges on relationships with other care providers (e.g. primary care, acute care, mental health care, community health care). An important element relates to the cultural profile of paramedics in the NHS, specifically, the extent to which other health professionals and care providers consider the clinical judgements/decisions made by paramedics as credible and actionable. Staff identified this as a barrier to access where the ambulance service is still viewed primarily as a transport service. Consideration could be given to ways of improving effective teamworking and communication across service and professional boundaries.

Although paramedics acknowledged the difficulties of telephone triage, they also identified how the limitations of this system impact on them. Over-triage at the initial call-handling stage places considerable demands on both staff and vehicle resources. A related concern is the limited information conveyed to crews following triage. Initial triage was suggested as an area that warrants attention to improve resource allocation.

The findings highlight the challenges faced by front-line ambulance service staff. It was apparent that the extent and nature of the demand for ambulance conveyance represents a notable source of strain and tension for individuals and at an organisational level. For example, there were widespread claims that meeting operational demands for ambulance services limits the time available for training and professional development, with this potentially representing a risk for patients and for staff. Staff perceptions of risk relating to patient safety extend to issues of secondary risk management, that is, personal and institutional liabilities, in particular risks associated with loss of professional registration. The belief that they are more likely to be blamed than supported by their organisation in the event of an incident was cited by staff as a source of additional anxiety when making more complex decisions. This perceived vulnerability can provoke excessively risk-averse decisions. These issues merit further attention to examine the workforce implication of service delivery changes, including how to ensure that staff are appropriately equipped and supported to deal effectively with the demands of their role.

Paramedics identified a degree of progress in relation to the profile of patient safety within their organisations but the apparent desire within trusts to prioritise safety improvement was felt to be constrained by service demands and available resources. Attempts to prioritise patient safety appear to focus on ensuring that formal systems are in place (e.g. reporting and communication). Concerns were expressed over how well these systems function to support improvement, for example how incident reports are responded to and whether lessons learned are communicated to ambulance staff within and between trusts. Consideration could be given to identifying ways of supporting ambulance service trusts to develop the safety culture within their organisation.

Service users attributed the increased demand for ambulance services to difficulties in identifying and accessing alternatives. They were receptive to non-conveyance options but felt that lack of awareness of staff roles and skills may cause concern when patients expect conveyance to A&E.

• Recommendations for research

The workshop attendees identified a range of areas for attention in relation to intervention and research, which are provided in Chapter 6 (see Suggestions for potential interventions and research ). The following recommendations for research are based on the study findings:

• Limited and variable access to services in the wider health and social care system is a significant barrier to reducing inappropriate conveyance to A&E. More research is needed to identify effective ways of improving the delivery of care across service boundaries, particularly for patients with limited options at present (e.g. those with mental health issues, those at the end of life and older patients). Research should address structural and attitudinal barriers and how these might be overcome.
• Ambulance services are increasingly focused on reducing conveyance to A&E and they need to ensure that there is an appropriately skilled workforce to minimise the potential risk. The evidence points to at least two issues: (1) training and skills and (2) the cultural profile of paramedics in the NHS, that is, whether others view their decisions as credible. Research could explore the impact of enhanced skills on patient care and on staff, for example the impact of increased training in urgent rather than emergency care. This would also need to address potential cultural barriers to the effective use of new skills.
• Research to explore the impact of different aspects of safety culture on ambulance service staff and the delivery of patient care (e.g. incident reporting, communication, teamworking, and training) could include comparisons across different staff groups and the identification of areas for improvement, as well as interventions that could potentially be tested.
• The increased breadth of decision-making by ambulance service crews with advanced skills includes more diagnostics; therefore, there is a need to look at the diagnostic process and potential causes of error in this environment.
• There is a need to explore whether there are efficient and safe ways of improving telephone triage decisions to reduce over-triage, particularly in relation to calls requiring an 8-minute response. This could include examining training and staffing levels, a higher level of clinician involvement or other forms of decision support.
• There is a need to explore public awareness of, attitudes towards, beliefs about and expectations of the ambulance service and the wider urgent and emergency care network and the scope for behaviour change interventions, for example communication of information about access to and use of services; empowering the public through equipping them with the skills to directly access the services that best meet their needs; and informing the public about the self-management of chronic conditions.
• A number of performance measures were identified engendering perverse motivations leading to suboptimal resource utilisation. An ongoing NIHR Programme Grant for Applied Research (RP-PG-0609–10195; ‘Pre-hospital Outcomes for Evidence-Based Evaluation’) aims to develop new ways of measuring ambulance service performance. It is important that evaluations of new performance metrics or other innovations (e.g. Make Ready ambulances, potential telehealth technologies or decision-support tools) address their potential impact on patient safety.

Included under terms of UK Non-commercial Government License .

• Cite this Page O’Hara R, Johnson M, Hirst E, et al. A qualitative study of decision-making and safety in ambulance service transitions. Southampton (UK): NIHR Journals Library; 2014 Dec. (Health Services and Delivery Research, No. 2.56.) Chapter 8, Conclusions and recommendations.
• PDF version of this title (1.5M)

In this Page

Other titles in this collection.

• Health Services and Delivery Research

Recent Activity

• Conclusions and recommendations - A qualitative study of decision-making and saf... Conclusions and recommendations - A qualitative study of decision-making and safety in ambulance service transitions

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

Connect with NLM

National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894

Web Policies FOIA HHS Vulnerability Disclosure

Help Accessibility Careers

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Childhood trauma, PTSD/CPTSD and chronic pain: A systematic review

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Validation, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliations Centre Hospitalier Agen-Nérac, Agen, France, UR 4139 Laboratoire de Psychologie, Université de Bordeaux, Bordeaux, France

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Writing – review & editing

Affiliation UR 4139 Laboratoire de Psychologie, Université de Bordeaux, Bordeaux, France

Roles Methodology, Validation, Writing – review & editing

Roles Conceptualization, Methodology, Supervision, Validation, Writing – review & editing

• Maria Karimov-Zwienenberg, 
• Wilfried Symphor, 
• William Peraud, 
• Greg Décamps

• Published: August 30, 2024
• https://doi.org/10.1371/journal.pone.0309332
• Peer Review
• Reader Comments

Despite the growing body of literature on posttraumatic stress disorder (PTSD) and chronic pain comorbidity, studies taking into account the role of childhood exposure to traumatic and adverse events remains minimal. Additionally, it has been well established that survivors of childhood trauma may develop more complex reactions that extend beyond those observed in PTSD, typically categorized as complex trauma or CPTSD. Given the recent introduction of CPTSD within diagnostic nomenclature, the aim of the present study is to describe associations between childhood trauma in relation to PTSD/CPTSD and pain outcomes in adults with chronic pain.

Following PRSIMA guidelines, a systematic review was performed using the databases Pubmed, PsychInfo, Psychology and Behavioral Sciences Collection, and Web of Science. Articles in English or French that reported on childhood trauma, PTSD/CPTSD and pain outcomes in individuals with chronic pain were included. Titles and abstracts were screened by two authors independently and full texts were consequently evaluated and assessed on methodological quality using JBI checklist tools. Study design and sample characteristics, childhood trauma, PTSD/CPTSD, pain outcomes as well as author’s recommendations for scientific research and clinical practice were extracted for analyses.

Of the initial 295 search records, 13 studies were included in this review. Only four studies explicitly assessed links between trauma factors and pain symptoms in individuals with chronic pain. Findings highlight the long-term and complex impact of cumulative childhood maltreatment (e.g., abuse and neglect) on both PTSD/CPTSD and chronic pain outcomes in adulthood.

This review contributes to current conceptual models of PTSD and chronic pain comorbidity, while adding to the role of childhood trauma and CPTSD. The need for clinical and translational pain research is emphasized to further support specialized PTSD/CPTSD treatment as well as trauma-informed pain management in routine care.

Citation: Karimov-Zwienenberg M, Symphor W, Peraud W, Décamps G (2024) Childhood trauma, PTSD/CPTSD and chronic pain: A systematic review. PLoS ONE 19(8): e0309332. https://doi.org/10.1371/journal.pone.0309332

Editor: Inga Schalinski, Universitat der Bundeswehr München: Universitat der Bundeswehr Munchen, GERMANY

Received: March 21, 2024; Accepted: August 9, 2024; Published: August 30, 2024

Copyright: © 2024 Karimov-Zwienenberg et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript.

Funding: The author(s) received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Over the past two decades, the comorbidity between chronic pain (i.e., persistent pain >3 months) and post-traumatic stress disorder (PTSD) has been well established [ 1 – 3 ]. PTSD is a psychiatric diagnosis based on the presence of a set of specific symptoms (e.g., flashbacks, hypervigilance, avoidance) that might occur after experiencing or witnessing a life-threatening event such as a disaster or assault. A recent meta-analysis including 21 studies reported higher PTSD prevalence up to 57% in individuals with chronic pain compared to 2–9% in the general population [ 4 ]. In the context of pain management, this alarming comorbidity represents many challenges as it has been associated with higher levels of pain severity [ 5 ], pain disability [ 6 ], and opioid use [ 7 ]. Furthermore, individuals with chronic pain and comorbid PTSD typically report increased levels of PTSD severity, emotional distress and psychiatric comorbidity than controls [ 8 – 10 ].

Several conceptual frameworks have been proposed, such as shared vulnerability and mutual maintenance models suggesting the interplay of neurobiological, emotional and cognitive factors involved in comorbidity [ 2 , 11 , 12 ]. Despite different hypotheses of causality and interaction, the particular nature of the relationship between chronic pain and PTSD remains uncertain. Depending on the studied population or condition, pain could both contribute to and maintain PTSD. Similarly, PTSD has been considered an important risk factor in the development of chronic pain when compared to controls [ 13 ].

Studies agree however that a history of adverse childhood events may be associated with both PTSD and chronic pain in adulthood [ 14 – 16 ]. Childhood adversity typically includes experiences of abuse, neglect as well as exposure to household dysfunction, parental psychopathology and early parental loss [ 17 ]. There is cumulative systematic and meta-analytical evidence demonstrating increased risk of chronic pain and pain-related disability in individuals reporting single or cumulative exposure to adverse childhood events, in particular maltreatment (e.g., childhood abuse, neglect) [ 15 , 18 , 19 ]. Although psychological distress has been identified as a key aspect to this phenomenon, few studies examined the role of PTSD in this context, indicating a gap in clinical and translational pain research, particularly in regard to trauma-informed pain management [ 20 ] as well as psychological treatment for comorbid trauma and chronic pain [ 21 ].

Additionally, it has been well established that survivors of childhood adversity may develop more complex and multifaceted reactions that extend beyond those observed in PTSD. These reactions have been commonly categorized as complex trauma or complex PTSD (CPTSD) [ 22 , 23 ]. CPTSD describes the widespread and long-lasting consequences following exposure to ongoing and often inescapable interpersonal traumatic stress that occurs within the context of a significant relationship (e.g., childhood abuse, intimate personal violence) [ 22 , 24 ]. Disparate adaptations to interpersonal trauma were initially conceptualized as an associated feature of PTSD by Disorders of Extreme Stress Not Otherwise Specified (DESNOS) [ 25 ]. However, due to the lack of sufficient evidence to support its inclusion as a unique diagnostic entity, DESNOS was eventually dropped from the fifth version of the Diagnostic and Statistical Manual (DSM) [ 26 ]. More recently, the World Health Organization published the 11 th version of the International Classification of Diseases (ICD-11) [ 27 ] introducing CPTSD for the first time into diagnostic nomenclature. Alongside the crucial presence of PTSD symptoms, the current model shares many similarities with DESNOS, including affect dysregulation, negative self-concept and interpersonal difficulties which are typically referred to as disturbances in self-organization (i.e., DSO symptoms) [ 28 , 29 ]. Additionally, consistent with recent data [ 30 , 31 ] and earlier conceptual research [ 29 , 32 ], current ICD-11 guidelines expanded trauma exposure definition for PTSD and CPTSD by taking into account different types of interpersonal trauma, including childhood neglect and emotional abuse, in addition to DSM criterion A events. In the context of chronic pain, there is some preliminary evidence suggesting worsened pain outcomes in survivors of childhood abuse with CPTSD as opposed to PTSD symptoms alone [ 33 ]. As PTSD and CPTSD are currently considered related disorders, it seems of timely interest to address how these relate to pain chronicity in order to promote effective treatment options and pain management for individuals with comorbid PTSD/CPTSD and chronic pain.

Despite the growing body of research on the trauma-chronic pain relationship, evidence in relation to PTSD/CPTSD following childhood exposure to traumatic or adverse events remains scarce. The aim of this study is to conduct a systematic review exploring existing data on the described links, while taking into account authors’ recommendations for future research and clinical practice. For the purpose of this review, in line with previous conceptual research and current ICD-11 PTSD/CPTSD guidelines, the term childhood trauma is used to address the exposure of traumatic or adverse events before the age of 18 years.

Specifically, this review seeks to describe in individuals with chronic pain:

• Childhood trauma
• Posttraumatic stress symptomatology, including PTSD and CPTSD symptoms.
• Relationship between childhood trauma and posttraumatic stress symptomatology, including PTSD and CPTSD.
• Relationship between trauma factors and chronic pain symptoms
• Scientific research
• Clinical practice

Search strategy

Before conducting this systematic review, a search in the Prospero database showed that, to our knowledge, no literature review is currently in progress on this subject ( https://www.crd.york.ac.uk/PROSPERO/ accessed on July 2023).

To conduct the present systematic review, we followed the guidelines described by the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) [ 34 ]. A search was performed from 1st of August 2023 using the following databases: Pubmed (Medline); PsychInfo (EBSCO host ), Psychology & Behavioral Sciences Collection (EBSCO host ), and Web of Science (Web of Knowledge). Search strategy terms are presented in Table 1 .

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0309332.t001

Inclusion and exclusion criteria

As per guidance, PICOTS framework [ 35 ] was used to structure the review process by defining selection criteria as follows: [ 1 ] Population, [ 2 ] Intervention, [ 3 ] Comparison, [ 4 ] Outcome, [ 5 ] Time and [ 6 ] Setting. Predefined inclusion and exclusion criteria are presented in Table 2 .

https://doi.org/10.1371/journal.pone.0309332.t002

Study selection and data extraction

Studies were selected independently by two authors (MKZ and WS) by screening titles and abstracts in systematic review. The selected studies were then subject to full text screening by applying the selection criteria. Reasons were documented during the process. In case of disagreement, discrepancies were adjudicated by a third author (WP) until a consensus was reached among the three authors. Once study eligibility was confirmed, data was extracted between September 2023 and December 2023 by one author (MKZ) which was then verified by a second author (WS). The following items were identified for data collection: authors, year, country, study design, study sample, chronic pain condition, chronic pain symptomatology, childhood trauma exposure, PTSD/CPTSD, interaction data between trauma factors and chronic pain symptomatology, and finally, author’s recommendations for scientific research and clinical practice.

Critical appraisal of study quality

The methodological quality of each included study was independently assessed by two researchers (MKZ and WS) using the corresponding design-specific critical appraisal checklist tools provided by the Joanna Briggs Institute (JBI) [ 36 ]. The following JBI critical appraisal checklist tools were used for this review: case control studies, analytical cross-sectional studies, quasi-experimental, as well as cohort studies. Each component was rated as “Yes”, “No”, “Unclear, or “Not Applicable”. If needed, discrepancies were discussed between reviewers or by consulting a third author (WP) until consensus was reached. Based on previous systematic reviews [ 37 , 38 ], studies with a JBI score higher than 70% were considered as high quality, those with scores between 50% and 70% as moderate quality, and those with a score less than 50% as low quality.

Study design and participants characteristics

The initial search returned 297 records, of which 36 were retained for full-text analysis. Finally, 13 articles [ 39 – 50 ] were included in this systematic review without disagreement (i.e., inter-judge agreement = 100%). Fig 1 presents a flow-diagram of the research article selection process.

https://doi.org/10.1371/journal.pone.0309332.g001

The 13 studies included in this review were published between 2005 and 2023 and conducted in Europe (Italy, n = 1; Belgium, n = 1; Spain, n = 2; Germany, n = 1), Turkey n = 1; Israel ( n = 3), and the US ( n = 4). Four were case-control studies, 7 cross-sectional studies, 1 quasi-experimental study and 1 cohort study. There was some variety in sample sizes across the studies, ranging from 70 to 295 participants, recruited both from clinical ( n = 9) and community settings ( n = 4). All study populations compromised exclusively ( n = 7) or predominantly female participants (>64%). Finally, Fibromyalgia (FM) was found to be the most studied pain condition ( n = 9), followed by unspecified chronic pain ( n = 3), and Interstitial cystitis/bladder pain syndrome (IC/BPS) ( n = 1). In terms of missing data, it was found that the majority of the included studies did not address all outcomes of interest to this review. Unreported information on outcomes was identified as “Not Reported” (N/R). Study findings are listed in Tables 3 and 4 .

https://doi.org/10.1371/journal.pone.0309332.t003

https://doi.org/10.1371/journal.pone.0309332.t004

Quality of the included studies

The results of the quality assessment are summarized in Table 5 . Quality appraisal using the JBI checklist tools indicated overall moderate to high quality studies. Nine studies scored above 70% [ 39 , 40 , 42 – 44 , 46 – 49 ], three studies scored between 50% and 70% [ 41 , 45 , 51 ], and the remaining one study [ 50 ] scored 13%. The main limitations of the single low-quality study were lack of objective and valid methods of assessment regarding chronic pain outcomes.

https://doi.org/10.1371/journal.pone.0309332.t005

Study objective 1: Descriptive data of trauma factors in individuals with chronic pain

A. childhood trauma..

All but one study [ 47 ] included in this review reported childhood trauma in terms of maltreatment, demonstrating higher prevalence [ 39 , 41 , 42 ] and severity [ 45 , 48 ] for emotional abuse and neglect compared to other forms of childhood maltreatment in individuals reporting chronic pain. In addition, a cohort study [ 48 ] demonstrated significative interrelations between all types of abuse and neglect, except for sexual abuse and neglect in a clinical sample of FM patients. Ciccone et al. [ 40 ] found no differences in childhood physical or sexual abuse between women reporting FM and healthy controls.

When compared with other medical conditions, studies found higher childhood maltreatment rates and severity in individuals with chronic pain, in particular with regards to neglect [ 39 , 41 , 45 ], sexual abuse [ 39 , 41 ], and emotional abuse [ 41 , 45 ].

Only two studies assessed childhood trauma exposure based on PTSD qualifying stressors following DSM criteria [ 42 , 47 ]. For example, Gardoki-Souto et al. [ 42 ] found that most prevalent traumatic events were reported during childhood compared to adulthood. Physical, sexual, and emotional abuse were identified as most commonly reported traumatic events during childhood. McKernan et al. [ 47 ] demonstrated differences in gender, with higher rates of childhood neglect observed in women, while men seemed to report more general disaster/trauma [ 47 ].

Finally, Hart-Johnson & Green [ 43 ] identified confounding effects of race and sex showing higher physical abuse under the age of 14 in male participants with chronic pain as opposed to women reporting chronic pain, with highest rates of abuse reported in black male participants and lowest in white female participants. Sexual penetration during childhood was found to be most prevalent among black female participants when compared with male or white female participants.

b. Posttraumatic stress symptomatology: PTSD/CPTSD.

In this review, the majority of the included studies described PTSD prevalence exclusively for predominantly female FM study samples with rates ranging from 10.7% to 37% [ 39 – 41 , 44 , 45 , 48 , 51 ]. One study [ 42 ] reported PTSD prevalence up to 71% following exposure to cumulative trauma as categorized by age. Results showed that most prevalent traumatic events occurred during childhood but continued into adulthood in the form of both different and recurrent types of events favoring a process of continuous re-traumatization. The lifelong impact of childhood trauma was further emphasized by high levels of current perceived distress in relation to past experiences of early life adversity.

When compared to controls, multiple studies showed higher PTSD prevalence and severity in individuals with chronic pain versus other medical conditions, including rheumatoid arthritis (RA) [ 45 ], functional disorders and achalasia [ 41 ], as well as healthy individuals [ 40 ]. Only one study [ 40 ] investigated PTSD symptom clusters, and found significatively higher rates for Intrusion and Arousal clusters, but not Avoidance when comparing a community sample of women with FM to healthy controls. Groups did not differ in childhood exposure to physical and/or sexual abuse.

Two studies included CPTSD measures in addition to PTSD [ 49 , 50 ] providing evidence for CPTSD and chronic pain comorbidity following childhood sexual abuse. For example, Peles et al. [ 49 ] demonstrated CPTSD prevalence rates between 19.1% and 60% in female survivors of childhood sexual abuse receiving methadone maintenance treatment versus those without a history of opioid addiction. Chronic pain comorbidity rates differed between CPTSD versus non CPTSD patients (100% vs 50%) without a history of addiction. Tsur [ 50 ] investigated PTSD/CPTSD in association with trauma-related pain symptoms and found higher levels of CPTSD symptoms (i.e., PTSD + DSO) linked to higher rates of pain flashbacks (23.1%), which is considered a posttraumatic stress response centralizing around pain, compared to women reporting non-pain flashbacks (36.3%) and no flashback symptoms (40.6%). In both studies, chronic pain was a self-reported outcome based on the presence of persistent pain lasting for more than six months. None of the included studies in this review reported on CPTSD in clinically diagnosed chronic pain patients or those receiving care for pain management.

Study objective 2: Interaction data between trauma factors and pain symptoms in individuals with chronic pain

A. childhood trauma, ptsd/cptsd in individuals with chronic pain..

Except for two studies [ 49 , 50 ], all included studies assessed childhood trauma in relation to PTSD as opposed to CPTSD. Several studies found that more severe childhood trauma, in particular maltreatment, was associated with PTSD in individuals with chronic pain [ 42 , 48 ] when compared to those without PTSD and healthy controls [ 46 ]. For example, in a community sample, higher rates of childhood trauma exposure, including sexual abuse, were found in participants with IC/BPS and comorbid PTSD as opposed to those without PTSD, represented by medium to large effect sizes. No differences were found regarding adult trauma exposure, including physical and sexual abuse, between these groups [ 47 ]. As for evidence on CPTSD outcomes, Tsur [ 50 ] associated higher childhood sexual abuse severity with increased experiences of pain flashbacks as well as CPTSD symptoms compared to controls (i.e., non-pain flashbacks, no flashbacks).

b. Trauma factors and pain symptoms in individuals with chronic pain.

Four studies included in this review explicitly investigated the association between childhood trauma, PTSD/CPTSD, and pain symptoms in individuals with chronic pain [ 41 , 47 , 49 , 50 ]. For example, Coppens et al. [ 41 ] assessed childhood maltreatment in relation to perceived pain experiences and found an indirect effect of childhood abuse and neglect on both quantitative and qualitative pain reports through PTSD severity, representing medium effect sizes. No relationship between childhood maltreatment severity and pain reports was revealed, nor a moderator effect of PTSD, suggesting a mediation effect. Other studies included in this review found direct effects of childhood trauma, in particular neglect and emotional abuse, on pain outcomes, including pain-related health impact and disability [ 39 , 42 ].

McKernan et al. [ 47 ] investigated the role of criterion A trauma on the relationship between chronic pain phenotypes and PTSD. In a convenience sample of participants with IC/BPS and comorbid PTSD, higher rates of current pain and clinically relevant central sensitization (CS) were observed in individuals as opposed to those without PTSD, represented by medium to large effect sizes. When comparing IC/BPS subgroups based on CS levels, all patients with PTSD corresponded to criteria of the widespread IC/BPS phenotype, associated with higher rates of polysymptomatic complaints, psychosocial distress and pain levels. While IC/BPS participants with CS reported higher rates of childhood trauma as well as lifetime physical and sexual abuse, PTSD was shown to be uniquely related over and above trauma exposure to widespread pain phenotype of IC/BPS.

Another study, using quasi-experimental design assessed analgesic responses in FM patients with and without PTSD based on stress-induced changes in pain and intolerance thresholds during a Social Stress Test task [ 46 ]. Results revealed lower basal pressure pain and intolerance thresholds during recovery when compared to healthy controls, indicating hyper sensitivity at basal function in FM patients, regardless the presence of PTSD. In response to acute stress, however, FM patients showed differences in hypo reactivity during the task, such as a lack of hyperalgesic response in FM with PTSD during and after exposure as opposed to a delay of a hyperalgesic response in FM patients without PTSD. Higher childhood trauma severity was found in FM patients with PTSD than those without PTSD. Groups did not vary in pain intensity or chronicity levels of FM symptoms.

Regarding CPTSD, two studies investigated associations with chronic pain comorbidity in female survivors of childhood sexual abuse. For example, a cross-sectional study conducted in a clinical sample, demonstrated positive correlations between chronic pain symptoms (e.g., pain severity, number of painful body regions), sexual abuse-related PTSD and CPTSD severity in adulthood. Age of onset of first experience of sexual abuse was negatively associated with pain duration [ 49 ]. Another study provided evidence for understanding the link between childhood sexual abuse, CPTSD and chronic pain by highlighting the role of somatic pain-related manifestations of PTSD/CPTSD, in particular pain flashbacks. Further, results identified peritraumatic pain during childhood sexual abuse as a risk factor for chronic pain in adulthood [ 50 ]. Overall, both studies including CPTSD measurement highlighted high prevalence of chronic pain in survivors of childhood sexual abuse associated with higher psychiatric comorbidity, namely CPTSD.

Finally, two studies demonstrated transcultural validity for associations between childhood trauma, PTSD and chronic pain symptoms drawing from evidence obtained in clinical settings across Europe, North America, and the Middle-East [ 44 , 45 ]. A study conducted in a community sample elucidated differences in chronic pain experiences in relation to abuse history based on sex differences [ 43 ]. Particularly, molestation was associated with higher affective pain, but only in men with chronic pain when compared with female participants. Similarly, childhood molestation predicted pain-related PTSD only in men, when controlling for race, sex and education. Female survivors of childhood sexual abuse were equally likely to have pain-related PTSD as women without a history of abuse.

Study objective 3: Author’s recommendations for future research and clinical practice

A. scientific research..

In the study of etiology and pathophysiology of chronic pain, comorbid mental disorders and psychological distress should be considered [ 44 ]. Additional research is also needed identifying mediating or moderating factors on the childhood trauma–HPA axis dysregulation relationship in chronic pain, using psychophysiological measures [ 48 , 51 ]. Suggested characteristics of childhood trauma typically include developmental timing and subtypes, while calling for empirical attention to childhood neglect [ 45 ], as well as subsequent experiences of violence or abuse, and ongoing interpersonal relations later in life [ 48 ]. Concurrently, more attention should be addressed to pain-specific posttraumatic stress symptoms (e.g., pain flashbacks, avoidance of trauma-related pain sensations), as well as somatic manifestations of CPTSD in relation to chronic pain [ 50 ]. Future research should assess trauma focused-interventions in FM in order to further clarify trauma-based etiology of FM in comparison to other functional somatic syndromes, medically unexplained symptoms, somatic symptoms, and related psychopathology [ 42 ]. Some findings included in this review also warrant further investigation on whether some psychological states of detachment (e.g., dissociation) might explain hypo reactivity in FM patients as a coping strategy. When addressing trauma in the context of chronic pain, differences in patients based on the presence of PTSD should be considered in future research by using a differential profile approach [ 46 ]. Finally, in the study of abuse and trauma in relation to chronic pain, more research should include men [ 43 ].

b. Clinical practice.

The majority of the included studies recommend systematic screening for trauma factors such as childhood trauma and PTSD/CPTSD [ 41 , 42 ], regardless of race, age or gender [ 43 ]. Specific training might be needed to reduce identified barriers (e.g., lack of time, discomfort with subject, or lack of familiarity with the role of abuse) to appropriate and effective screening methods [ 43 ]. Screening procedures should also include detection for potential comorbid mental disorders in relation to abuse, such as somatoform dissociation disorder and alexithymia, using appropriate tools [ 44 , 51 ]. Trauma-focused therapies may include Eye Movement Desensitization and Reprocessing (EMDR) [ 42 ], as well as intervention techniques based on Eccleston’s model of tripartite system of threat protection in order to support FM patients with and without PTSD to engage in more adaptive stress responses [ 46 ]. As PTSD appears to be associated with the “widespread” pain phenotype, multimodal treatment should be considered for these patients [ 47 ]. Trauma-informed care is recommended in a more general way, emphasizing patient-care provider trust and rapport, reducing anxiety and increasing patient control and safety during appointments and medical examination procedures [ 47 ]. Finally, clinicians treating survivors of abuse should specifically inquire about chronic pain complaints, in order to facilitate tailored adequate approaches in comprehensive treatment [ 49 ].

Despite the growing evidence on the trauma-pain relationship, literature examining the association between childhood trauma and PTSD in relation to pain outcomes remains limited. This review further adds on existing systematic data by including evidence on CPTSD in individuals with chronic pain. In total, 13 studies were included in this systematic review. Study highlights have been summarized into the following sections in order to guide future research as well as recommended evidence-based clinical practice and policy in routine pain management.

Childhood trauma: Neglect and emotional abuse in individuals with chronic pain

Different aspects of childhood trauma have been previously identified as risk factors for chronic pain conditions, such as nature of trauma [ 15 , 52 ], and cumulative experiences of maltreatment to [ 19 , 53 , 54 ]. In addition to existing systematic and metanalytical data, studies included in this review particularly emphasize the long-term consequences of emotional abuse and neglect as opposed to physical and sexual abuse. Consistent with DSM A-criterion type of traumatic events, other reviews typically focused on the impact of abuse specific childhood trauma (e.g., physical abuse, sexual abuse) [ 10 , 52 , 55 , 56 ]. There is some research, however, indicating an independent relationship between PTSD symptoms and chronic pain outcomes following the presence of criterion A trauma history [ 57 ]. Moreover, present findings provide evidence for the expanded definition of trauma exposure by current PTSD/CPTSD ICD-11 guidelines, in particular with respect to the inclusion of childhood neglect and emotional abuse, in addition to DSM A criterion events. Despite suggested relevance to chronic pain etiology and PTSD/CPTSD comorbidity, research clarifying the differential impact of neglect and emotional abuse alongside events of childhood physical and sexual abuse remains minimal and warrants further investigation whether and to what extent these forms of trauma are associated with unique healthcare needs in chronic pain management.

The long-term impact of childhood trauma: Evidence for differential patterns in PTSD/CPTSD and pain modulation processes

In total, only four studies included in this review explicitly investigated relationships between childhood trauma, PTSD/CPTSD and pain outcomes in individuals with chronic pain. The present findings are in accordance with other research demonstrating the negative impact of PTSD on pain outcomes when linked to childhood maltreatment compared to lower levels of pain typically experienced by individuals who have been diagnosed with PTSD alone [ 54 , 58 ]. The long-term impact of cumulative childhood trauma was further recognized by an indirect dose-response relationship associated with increased risk of re-traumatization, higher levels of PTSD and perceived distress when compared to adulthood trauma. Similar to results of a recent systematic review [ 1 ], certain chronic pain phenotypes (e.g., “widespread pain”) were identified as risk factors for described links.

This review also included evidence on biomarkers involved in pain modulation processes (e.g., cortisol secretion, pressure pain thresholds). In addition to existing systematic data [ 59 ], study findings support inhibitory capacity of adaptive allodynic responses in chronic pain patients with a history of childhood trauma by adding information to the role of PTSD. In this connection, differential neurophysiological patterns in chronic pain patients with PTSD compared to those without PTSD were associated with two main psychological/behavioral responses, namely hyperarousal and dissociation [ 46 , 48 ]. This hypothesis is in line with previous studies, suggesting a unique paradoxical pain profile in individuals with chronic pain and PTSD, characterized by both pain-related hypo- and hyperresponsivity when compared to controls [ 8 , 60 ]. Other research has emphasized the role of childhood versus adulthood trauma exposure in advancing current understanding of differential PTSD-related conditions (e.g., dissociation, depression) and physical health symptoms, including pain [ 61 ].

It is important to note, however, that results associating childhood trauma, PTSD, and pain are typically obtained in the absence of any CPTSD assessment. Only one study included in this review examined differential role of CPTSD symptoms in relation to childhood trauma, while identifying pain-related somatic manifestations (e.g., pain flashbacks) both as maintaining and worsening factors of chronic pain outcomes. These results are consistent with some preliminary research demonstrating associations between CPTSD symptoms (i.e., DSO symptoms) and higher rates of somatization [ 62 ] as well as abusive pain personification in individuals with childhood trauma compared to those with PTSD [ 33 ]. Despite important implications for empirical and clinical efforts as argued by a recent review [ 63 ], our understanding of trauma-related bodily experiences remains an underdeveloped realm of translational pain research. In particular, findings in this review corroborate the current lack of validated and standardized assessment for pain-related trauma factors (e.g., peri and posttraumatic pain) which was identified as a major barrier to more robust methodological evidence. The need for future research adopting a differential analytical approach (e.g., cluster analysis), has also been issued to verify theorized relationships in order to extend current conceptual models of comorbidity and pain phenotypes by considering the unique features of CPTSD alongside PTSD symptoms.

Trauma–pain comorbidity: Intersectional disparities

Although transcultural validity of trauma factors in chronic pain outcomes was consistently reported in this review [ 44 , 45 ], the majority of the included study samples represented predominantly Caucasian and female individuals suffering from FM. Only one study provided some insight into intersectional disparities regarding childhood abuse in adults with chronic pain [ 43 ]. Findings corroborate the lack of available evidence identified by a recent review [ 64 ], emphasizing the critical need for more inclusive research to ensure that underrepresented groups receive equitable benefit from chronic pain research in terms of health and social policy. The same applies to trauma factors that remain oftentimes under-recognized, under-treated, or inadequately treated among marginalized groups [ 65 ]. More research is needed to explore the interplay of social factors (e.g., socioeconomic status, gender, race) and health disparities, while building on evidence for a more precise understanding of trauma-pain comorbidity and management within social context.

Trauma focused treatment versus trauma-informed care

Considering the widespread prevalence of childhood trauma and both its long-term and complex impact on posttraumatic symptoms and pain related outcomes later in life, recommendations for clinical practice included in this review address the need for systematic screening of trauma factors in individuals seeking care for chronic pain. Consequently, psychotherapeutic strategies should target PTSD/CPTSD to relief illness burden, helping individuals with chronic pain to engage in more adaptive stress responses and promote general functioning [ 41 , 42 , 46 ]. Despite extensive literature on psychological treatment for PTSD, there is currently no “gold standard” for CPTSD screening or intervention methods. Furthermore, numerous limitations have been associated with first-line, evidence-based treatments for PTSD, including early dropout and worsening of symptoms in survivors of interpersonal trauma [ 66 – 68 ]. In this regard, Trauma Center Trauma Sensitive Yoga (TCTSY) [ 69 ]’, an evidence-based protocol for complex trauma or treatment-resistant PTSD, appears to be a particularly promising therapeutic strategy, drawing specific focus to interoception (i.e., awareness of bodily sensations) and empowerment processes. While there is cumulative qualitative and quantitative evidence demonstrating protocol efficacity compared to conventional psychotherapy modalities [ 70 – 72 ], the use of TCTSY in individuals with chronic pain has not yet been investigated. In addition to trauma specialized treatment, and in line with a recent topical review [ 73 ], the present findings further support the importance of a systems approach to trauma care in pain management and rehabilitation services. Future research is needed to investigate comprehensive models of trauma-informed care based on principals such as safety, collaboration and choice within routine practice as a means to improve patient adherence, pain outcomes and prevent re-traumatization.

Methodological considerations

This systematic review was conducted following recommended guidelines for search strategy as well as quality assessment allowing for a more rigorous process regarding methodological appraisal. Some limitations, however, should be taken into consideration in analyzing key findings. The search was not limited to study design, year of publication or methodological quality. Further, inclusion criteria for chronic pain and trauma factors were generally defined such as to provide a broad overview of the current state of art, limiting therefore conclusive or generalizing evidence regarding subtypes of trauma in relation to specific pain syndromes or phenotypes. Despite the inclusive approach to this review, only a short list of mostly moderate to high quality evidence, was identified, highlighting the preliminary nature of research in this area. Overall, the selected studies used appropriate and validated measurement for childhood trauma, PTSD/CPTSD and chronic pain which included a variety of self-reported as well as physician-based assessment. However, the heterogeneity of tools included in this review, in particular for PTSD/CPTSD, warrants vigilance to generalization of findings. Further, the majority of selected studies used the Childhood Trauma Questionnaire [CTQ; 74 ] as primary measurement for childhood trauma. While this is a validated and widely utilized instrument in the study of childhood trauma history, it provides assessment limited only to childhood maltreatment (i.e., abuse and neglect). Only two studies in this review assessed childhood trauma exposure based on PTSD qualifying stressors following DSM diagnostic criteria. This review recognizes the instability around diagnostic consensus of PTSD/CPTSD proposed by distinct classification models over the past two decades. For example, based on earlier diagnostic and clinical literature [ 22 , 25 ], somatization was typically considered a core feature of DSM DESNOS, but does not appear in the current WHO ICD-11 model of CPTSD. Finally, to the best of our knowledge, there is currently no randomized controlled, longitudinal or case study evidence investigating intervention modalities for PTSD/CPTSD and chronic pain comorbidity in individuals with a history of childhood trauma.

The findings of this systematic review highlight the importance of taking into account childhood trauma, in particular neglect and emotional abuse, in the study of PTSD/CPTSD and chronic pain comorbidity in adults. The long-term impact of childhood trauma was further emphasized by an indirect dose-response relationship associated with increased risk of re-traumatization, higher levels of PTSD and perceived distress later in life when compared to adulthood trauma. This review also included evidence on specific neurophysiological patterns in chronic pain patients with PTSD suggesting differential pain modulation processes following trauma, in particular childhood maltreatment. Only a few selected studies reported on CPTSD and chronic pain comorbidity, providing preliminary evidence on the role of trauma-related physical pain (e.g., pain flashbacks). The need for future research adopting a differential approach has been issued in order to extend current models of comorbidity in relation to pain phenotypes, while also accounting for intersectional disparities. Considering the widespread prevalence of childhood trauma and its long-term and complex impact on both PTSD/CPTSD and pain chronicity later in life, recommendations for clinical practice draw attention to the need for PTSD/CPTSD specialized treatment as well as trauma-informed pain management in routine care.

Supporting information

S1 file. prisma checklist 2020..

https://doi.org/10.1371/journal.pone.0309332.s001

S2 File. List of identified studies in the literature search.

https://doi.org/10.1371/journal.pone.0309332.s002

• View Article
• PubMed/NCBI
• Google Scholar

Assessment of Machine Learning Algorithms for Land Cover Classification in a Complex Mountainous Landscape

• Open access
• Published: 30 August 2024
• Volume 8 , article number  34 , ( 2024 )

Cite this article

You have full access to this open access article

• Gomal Amin   ORCID: orcid.org/0000-0003-1700-115X 1 , 2 ,
• Iqra Imtiaz 1 ,
• Ehsan Haroon 1 ,
• Najum us Saqib 3 ,
• Muhammad Imran Shahzad   ORCID: orcid.org/0000-0003-3795-9832 1 &
• Majid Nazeer   ORCID: orcid.org/0000-0002-7631-1599 2 , 4  

Mapping land cover (LC) in mountainous regions, such as the Gilgit-Baltistan (GB) area of Pakistan, presents significant challenges due to complex terrain, limited data availability, and accessibility constraints. This study addresses these challenges by developing a robust, data-driven approach to classify LC using high-resolution Sentinel-2 (S-2) satellite imagery from 2019 within Google Earth Engine (GEE). The research evaluated the performance of various machine learning (ML) algorithms, including classification and regression tree (CART), maximum entropy (gmoMaxEnt), minimum distance (minDistance), support vector machine (SVM), and random forest (RF), without extensive hyperparameter tuning. Additionally, ten different scenarios based on various band combinations of S-2 data were used as input for running the ML models. The LC classification was performed using 2759 sample points, with 70% for training and 30% for validation. The results indicate that the RF algorithm outperformed all other classifiers under scenario S1 (using 10 bands), achieving an overall accuracy (OA) of 0.79 and a kappa coefficient of 0.76. The final RF-based LC mapping shows the following percentage distribution: barren land (46.7%), snow cover (22.9%), glacier (7.9%), grasses (7.2%), water (4.7%), wetland (2.9%), built-up (2.7%), agriculture (1.9%), and forest (1.2%). It is suggested that the best identified RF classifier within the GEE environment should be used for advanced multi-source data image classification with hyperparameter tuning to increase OA. Additionally, it is suggested to build the capacity of various stakeholders in GB for better monitoring of LC changes and resource management using geospatial big data.

Similar content being viewed by others

Evaluating the impact of classification algorithms and spatial resolution on the accuracy of land cover mapping in a mountain environment in pakistan.

Spatial Resolution Impacts on Land Cover Mapping Accuracy

Land use and land cover classification using machine learning algorithms in google earth engine

Explore related subjects.

• Artificial Intelligence

Avoid common mistakes on your manuscript.

Introduction

Land cover (LC) can be significantly altered by factors such as excessive agricultural development, rapid population growth, and the overexploitation of natural resources, leading to landscape degradation (Beuchle et al. 2015 ). The mountain environments worldwide, including the Hindu Kush, Karakoram, and Himalayan (HKH) range, are highly susceptible to various anthropogenic and natural hazards, such as climate change, tourism, urbanization, population growth, and economic development (us Saqib et al. 2019 ; Saini & Singh 2024 ). These mountain regions are prone to natural disasters like landslides, flash floods, earthquakes, and glacier lake outburst floods (GLOFs) (Bacha et al. 2018 ; Jamil et al. 2019 ; Xu et al. 2009 ). Understanding the LC patterns in such areas can help in identifying locations that are at greater risk of such kind of natural disasters and can aid in developing strategies to mitigate them (Saini & Singh 2024 ).

Apart from these natural disasters, the rapid population growth, excessive agricultural expansion, and uncontrolled urbanization in mountainous area have led to the overexploitation of natural resources, landscape degradation, and land deterioration. Understanding the LC features and implementing better management strategies for natural resources are crucial for sustainable development in the environmental, social, and economic sectors (Dang and Kawasaki 2017 ). Any adverse impact on the fragile mountainous environment can pose severe challenges for the large human population residing in these areas. Different studies have emphasised the importance of LC analysis for effective management of natural resources in the fragile mountainous environments (Satti et al. 2023 ). To examine such changes in LC with exceptionally high accuracy, remote sensing (RS) images classified with machine learning (ML) algorithms are considered as standard tools that are persistently used all around the world (Gargiulo et al. 2020 ; Jia et al. 2023 ).

A large number of ML-based LC classification algorithms have been explored over the past decade to produce accurate, up-to-date, and long-term LC maps (Zhang & Zhang 2020 ; Wang et al. 2023 ). For instance, artificial neural network (ANN) (Yuan et al. 2009 ), random forest (RF) (Gislason et al. 2006 ; Wang et al. 2023 ), classification and regression tree (CART) (Shao & Lunetta 2012 ), and support vector machine (SVM) (He et al. 2005 ) have demonstrated superior performance in mapping different LC types compared to traditional classifiers (Belgiu and Drăgu 2016 ). The RF classifier is particularly popular in the RS community (Xiong et al. 2017 ) due to its high accuracy, achieved by constructing multiple decision trees (DTs). For example, RF algorithm was used in Western Himalayas to classify vegetation types with an overall accuracy (OA) of 80%, considering topographic and climate variables for improved accuracy (Singh et al. 2023 ). Similarly, a study conducted by Zurqani ( 2024 ) for forest canopy cover using RF achieved an OA ranging between 83.31 and 94.35%. Moreover, Mansaray et al. ( 2019 )deployed both SVM and RF classifiers for mapping paddy rice in China using Landsat-8 and Sentinel-2 images for the 2015 and 2016, and the RF classifier demonstrated higher accuracy (95%) compared with the SVM classifier (90.8%). Delalay et al. ( 2019 ) utilized CART, maximum entropy, and RF classifiers within the Google Earth Engine (GEE) environment for LC classification using Sentinel-2 data. The results showed that the RF technique had the highest OA (95%), followed by maximum entropy (93%) and CART (61%) in the mountainous region of Nepal. A recent study (Mahmoodzada et al. 2024 ) utilized the SVM and the multilayer perceptron (MLP) to map snow cover area in Pamir region of Hindukush with kappa coefficient of 0.75 and 0.83, respectively. Shetty et al. ( 2021 ) evaluated the impacts of training sampling design on LC classification results within the GEE, concluded RF outperformed both CART and SVM.

The selection of an appropriate ML classifier for LC mapping is a challenging task due to the large number of available algorithms, their varying computational performance, and the conflicting information about their OA. Additionally, the combination of spectral bands used as input can significantly affect the classification accuracy (Shetty et al. 2021 ; Xiong et al. 2017 ). Various researchers have explored the use of different spectral band combinations from RS data, such as Sentinel-2 (Silveira et al. 2023 ) to improve the LC classification accuracy (Gumma et al. 2020 ; Stromann et al. 2020 ). However, there appears to be a lack of published research evaluating the performance of diverse ML algorithms applied to Sentinel-2 imagery for LC mapping in the HKH region of Pakistan. The majority of existing studies in this geographic context have focused on classifying a limited set of LC types (Khan et al. 2020a , b; Qamer et al. 2016 ; Satti et al. 2023 , 2024 ). These investigations have typically employed single image and conventional classification algorithms for mapping within small sub-regions, often resulting in varying and even contradictory outcomes. For instance, in the current study area Khan et al., ( 2019 ) and Ali et al., ( 2019a , b) performed LC mapping for Gilgit city and Gilgit district in Pakistan, respectively, using Landsat data and maximum likelihood classifier (MLC) to identify five generic LC classes. This makes it challenging to compare the accuracy of the generated LC maps and identify the most reliable approach for this complex mountainous environment (Delalay et al. 2019 ). As of the time of writing this paper, there have been no published studies focusing on detailed LC mapping in the HKH region of Pakistan utilizing high-resolution RS imagery, despite the significant advancements in the field.

This gap in the literature underscores the need for a comprehensive evaluation of the performance of a wider range of ML algorithms, including their computational efficiency and classification accuracy, when applied to high-resolution Sentinel-2 data for LC mapping in the HKH region of Pakistan. Such an assessment would provide valuable insights to support the selection of the most appropriate LC classification approach for this ecologically significant yet geographically challenging area. When mapping LC over a large extent (such as the current study area), researchers had to consider the key challenges regarding the processing of ‘big earth data’ and the availability of images (Satti et al. 2024 ). Previous researchers in the study area were limited in their ability to run ML classification algorithms due to constraints in computing power and storage. However, the utilization of GEE free cloud-based computing platform has enabled scientists to utilize the satellite data for large-scale LC mapping in a more efficient and effective way (Gorelick et al. 2017 ; Zurqani 2024 ).

The objective of this scientific study is to compare and evaluate the performance of various machine learning algorithms available within the Google Earth Engine platform for land cover mapping in the HKH region of Pakistan. This assessment will be conducted without any hyperparameter tuning, using the full range of spectral band combinations from Sentinel-2 imagery with a temporal aggregation method. Moreover, the final product generated from this study will be made freely available for users, facilitating broader access and utilization of the LC mapping results to support land management, spatial planning, and disaster risk management to achieve sustainable development in the region.

Material and Methods

Gilgit-Baltistan (GB) located in the north of Pakistan is characterized by a remote mountainous environment, surrounded by the world’s famous highest mountain ranges, i.e., Hindu Kush, Karakoram, and Himalaya. Administratively, GB is divided into ten districts, namely, Astore (5179 km 2 ), Diamer (6901 km 2 ), Ghanche (8525 km 2 ), Ghizer (12043 km 2 ), Gilgit (4009 km 2 ), Hunza (11343 km 2 ), Nagar (2993 km 2 ), Kharmang (2802 km 2 ), Shigar (8810 km 2 ), and Skardu (7200 km 2 ) (Amin et al. 2021 ) (Fig.  1 ).

Overview of the study area Gilgit-Baltistan (GB) overlaid over an elevation map derived from 30 m Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Global Digital Elevation Model (GDEM). The inset map indicates the location of study area in Pakistan

GB can be broadly classified into five distinct ecological zones: dry alpine zones and permanent snowfields, alpine meadows and alpine scrub, sub-alpine scrub, dry temperate coniferous forest, and dry temperate evergreen oak scrub. Moreover, GB has an exceptionally complex mountain system, with approximately 90% of its area covered by rugged mountain ranges and glaciers, while the remaining area consists of arable land (Hussain & Bangash 2017 ).

The climate of Gilgit-Baltistan is influenced by both the monsoon season, which contributes up to 80% of the region’s summer precipitation, and westerly cyclones, which account for approximately 66% of the high-elevation snowfall. However, the steep topography of the Karakoram range diminishes the influence of these wind systems as one moves towards the northern parts of the region (Bolch et al. 2012 ; Rankl et al. 2014 ). Generally, the weather conditions are severe with cold winters and extremely hot summers. The region receives precipitation of approximately 200–2000 mm per year, varying in different elevation zones, whereas temperature ranges between 10 °C (in winters) and 40 °C (in summer) depending on the valley’s elevation range (Gilani et al. 2020 ; Nawaz et al. 2019 ).

Training Sample Selection

Nine LC classes (Table  1 ) were defined based on a detailed literature review of the study area (Ali et al. 2019a , b; Gumma et al. 2020 ; Khan et al. 2020a , b; Qamer et al. 2016 ; Rahim et al. 2018 ). Training and validation sample points for each LC class were selected through simple random sampling using high-resolution Google Earth imagery and the authors’ personal experience of the study area. During the preparation and verification of the sample points, the principles of ‘consistency’ and ‘reliability’ were carefully maintained (Hill et al. 2008 ). This involved minimizing the inclusion of mixed pixels by avoiding sample collection from the edges of LC class boundaries and fragmented landscapes (Hu & Hu 2019 ; Phan et al. 2020 ).

Auxiliary data available in the GEE, such as the Advanced Land Observing Satellite (ALOS) Digital Surface Model (DSM) (Tadono et al. 2014 ), Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) (Farr et al. 2007 ), and Defence Meteorological Satellite Program (DMSP) Operational Line-Scan System (OLS) night time imagery, were used to support the selection of training samples and improve the overall classification accuracy. Additionally, DEM-derived products, such as slope, aspect, and elevation, were utilized as supplementary data for visual interpretation during the training sample selection process.

In total, 2759 sample points were randomly divided, with 70% used for training the classification models and 30% reserved for validation and accuracy assessment. The distribution of sample points across the nine LC classes was as follows: water (341), forest (304), grasses (306), wetland (364), agriculture (389), barren land (467), built-up (197), glacier (176), and snow (215). The same training and validation datasets were used to evaluate the performance of each classification model.

Data Acquisition and Processing

The European Space Agency (ESA) and the European Commission (EC) launched the Sentinel-2 (S-2) satellites in 2015 and 2017 under the Copernicus program. The Sentinel-2A and 2B satellites have a revisit interval of 10 days individually and a combined revisit time of 5 days. The improved spatial and spectral resolution of the S-2 imagery has opened up new possibilities for environmental studies and monitoring (ESA 2015 ; Gargiulo et al. 2020 ). The twin S-2 satellites provide global coverage with 13 spectral bands, ranging from the visible to the short-wave infrared (SWIR) wavelengths, offering different spatial and spectral resolutions (Table  2 ). In the GEE catalogue, the Sentinel-2 Multispectral Instrument (MSI) data is available as Level-2A Surface Reflectance (SR) product, processed by running the sen2cor algorithm (COPERNICUS 2017 ). The Sentinel-2 Level-2A SR data within the GEE can be accessed using the code snippet ‘ee.ImageCollection(“COPERNICUS/S2_SR”)’, which provides 12 UINT16 spectral bands representing surface reflectance scaled by a factor of 10,000.

Figure  2 provides an overview of the image processing steps employed to map the LC of the study area. The methodology implemented in this study involved temporal aggregation of all available images within the study area to generate a composite image free from cloud cover. To mitigate the influence of monsoon rains and fresh snow on the classification outcome, a temporal frame ranging from 1st May to 30th September 2019 (152 days) and cloud coverage threshold of less than 20% was selected. This resulted in an image collection of 192 scenes, distributed across ten Sentinel-2 tiles. Subsequently, a 10-m spatial resolution cloud free image composite was generated by calculating the median pixel value from the entire image collection.

Flowchart of land cover mapping using Google Earth Engine. The developed methodology included the data acquisition and processing, training sample selection, and accuracy assessment for GEE in-build machine learning classifiers

The median function was utilized as the temporal aggregation method to minimize the impact of missing or gapped cells in the Sentinel-2 imagery. This approach is commonly employed in the literature to reduce noise, particularly along scene borders (Carrasco et al. 2019 ; Rudiyanto et al. 2019 ). For each of the 10 spectral bands considered, any pixel values identified as blanks were replaced by the median value of that band across all images in the acquisition period. This ensures that a pixel identified as blank remains as such only if all images taken during the study period have a blank value for that location. However, in this study, no blank or void pixels were identified after the temporal aggregation process was applied.

For Sentinel-2 image classification, ten scenarios with different band combinations were used (Table  3 ) based on comprehensive literature review (Adepoju & Adelabu 2020 ; Alifu et al. 2020 ; Li et al. 2020 ; Xiong et al. 2017 ). These studies have suggested specific band combinations that have shown promising results in LC mapping applications. By evaluating these band combinations against ML classifiers in the GEE platform, we aimed to identify the best performing classifier without engaging in any hyperparameter tuning.

Classification

The ML classifiers included in the current study were classification and regression tree (CART), maximum entropy, minimum distance, support vector machine (SVM), and random forest (RF). All of these classification models were implemented without any hyperparameter tuning, using the default parameter values to ensure a consistent comparison across all models (Table S1 ), as suggested by Maxwell et al. ( 2018 ).

CART Classifier

The CART algorithm proposed by Breiman et al. ( 1984 ) is a decision tree construction method that works on the principle of a dichotomous recursive segmentation system. The CART algorithm utilizes the Gini coefficient as the criterion for identifying the ideal test variances and segmentation thresholds to create a binary tree-based decision tree (DT) for classification. The CART algorithm operates by recursively splitting the training data at each decision node, known as the greedy splitting approach, to increase the homogeneity of the data in the resulting nodes based on a statistical test such as the Gini index. The Gini coefficient is defined as follows in Eqs. ( 1 , 2 and 3 ):

where P ( j | h ) is a randomly selected sample from a training set or relative frequency of category, n j ( h ) corresponds to the number of samples in category j when the value of test variable in the training set is h node, where n ( h ) is the number of samples in the training dataset with the test variable value of h , and j denotes the category number.

To be precise, the CART algorithm takes the training dataset and partitions it into smaller subsets recursively. This partitioning process continues until the smaller cells are grouped based on the same class label, with the maximum accuracy of prediction validated by the pruning value (Hayes et al. 2015 ; Mondal et al. 2019 ). The CART algorithm does not require parameters and has the advantages of fast operating speed and easy manipulation (Shao & Lunetta 2012 ); different studies have successfully deployed CART for classification of satellite imagery with promising accuracy (Hu et al. 2018 ; Johansen et al. 2015 ).

Maximum Entropy Classifier

Maximum entropy (MaxEnt) or gmoMaxEnt classifier in GEE is based on the maximum entropy principle to select the data with maximum entropy from all training sets (Mcdonald et al. 2009 ). It works best in a condition where the prior distribution and conditional dependency are unknown making it difficult to perform prediction with any assumption. Therefore, the gmoMaxEnt classifier utilizes a machine learning approach to perform spatial predictions using incomplete or limited training data (Moreno et al. 2011 ).

Minimum Distance Classifier

The minimum distance classifier uses spectral characteristics of the training samples which have been selected as representatives of the different feature classes. The Euclidean distance between the selected pixel values and the mean values of each class is calculated. Later, the candidate or selected pixel is assigned to the class with which it has the shortest Euclidean distance (Hu & Hu 2019 ).

SVM Classifier

SVM was described by Cortes and Vapnik ( 1995 ) which is commonly used in a range of RS applications (Rudiyanto et al. 2019 ; Stromann et al. 2020 ). SVM is a supervised machine learning technique that aims to find an ideal hyperplane that discriminates different classes from their decision boundary. During classification, SVM classifiers use an iterative process to allocate candidate pixels to classes by maximizing class separability from the training set and labels each pixel according to their nearest class in feature space (Boser et al. 1992 ; Tsai et al. 2018 ). The selection of the support vectors mainly depends on the choice of cost parameter C, kernel functions, and Gamma. The most used kernel functions include linear, polynomial, and radial basis function (RBF). A detailed description of the SVM classifier can be found in Melgani and Bruzzone ( 2004 ). The mathematical equations of linear, polynomial, radial basis and sigmoid kernel functions are listed below as Eqs. ( 4 – 7 ).

where k is kernel, j is the feature, \({x}_{i}\) are input data points, and \({y}_{i}\) are the corresponding output data points. In polynomial kernel, d shows the degree of polynomial, whereas \(r\) in the polynomial and sigmoid function is considered as bias term. \(r\) is the gamma term that presents in all types of function except linear which describes the impact of the training range. The model will be constrained and not be able to handle the complexity of data if the value of gamma is too small and contrariwise.

RF Classifier

Random forest (RF) is a nonparametric supervised ML algorithm (Lee et al. 2018 ). The RF classifier is a nonlinear, relatively fast classifier that acts robustly to noisy training data. The RF algorithm was developed as an extension of the CART decision tree method and generates multiple classification trees to improve the overall prediction performance of the model. It operates by using a number of decision trees (DTs), where each tree is created from an independently constructed random sample of the training data to assign classification labels to each class (McCord et al. 2017 ). The RF algorithm applies a bagging technique, randomly selecting a subset of features from the input observations for each decision tree to be grown (Belgiu and Drăgu 2016 ).

As mentioned in Ahmed et al. ( 2019 ), for each tree ‘d’ from total ‘D’ number of trees, select any random data from training dataset and create the random forest tree T d by randomly select ‘i’ point. This process must be recursive until the minimum node size is achieved. Combining the outputs of all trees as we get Eq. ( 8 ):

For the new prediction at any point i , the regression will be Eq. ( 9 ):

Methods for Accuracy Assessment

The accuracy of the LC classification was measured using various derivates of the confusion matrix, such as overall accuracy (OA), kappa coefficient, producer’s accuracy (PA), and user’s accuracy (UA) for each class. The OA represents the percentage of pixels that were correctly labelled by the classifier, while the kappa coefficient is a measure of the overall agreement between the classification results and the reference data, accounting for chance agreement. The OA estimation and the kappa coefficient were used to compare the classification accuracy of each machine learning algorithm across the different band combination scenarios, ultimately supporting the selection of the best-performing classifier. Additionally, the Pearson correlation (PC) coefficient was calculated between the OA and kappa values. The PC coefficient measures the statistical relationship or association between these two continuous variables, providing information about the magnitude and direction of the correlation, which can range from − 1 to + 1 (Benesty et al. 2009 ).

To further evaluate the performance of the LC classification models, we employed standard metrics including precision, recall, and F1-score (Saini & Singh 2024 ). These metrics were computed for each LC class, providing a comprehensive evaluation of the classifier’s ability to accurately identify and differentiate between distinct LC types. Precision quantifies the proportion of correctly classified samples within a specific class, while recall represents the proportion of samples from a given class that were correctly identified. The F1-score, a harmonic mean of precision and recall, offers a balanced measure of the classifier’s overall performance for each class, providing a robust indicator of the model’s effectiveness in classifying different LC types (Rapinel et al. 2023 ).

Comparison of Classification Accuracy of GEE Classifiers

The results show that the various ML classifiers achieved differing levels of OA and kappa index under the different band combination scenarios (Table  4 and Fig.  3 ). The 10-band set (S1, as detailed in Table  3 ) of Sentinel-2 data, covering the visible to SWIR band regions, resulted in OAs ranging between 0.59 and 0.79 and kappa coefficients between 0.54 and 0.76 across the ML algorithms. In the S2 scenario (Table  3 ), which utilized a different band configuration, the CART classifier achieved the highest OA of 0.73 and kappa of 0.69 (Table  4 ). In contrast, the S9 scenario, which used only the combination of Sentinel-2 bands B2-B4, did not perform well, with no classification method exhibiting satisfactory accuracy. The comparative analysis of the ML algorithms and band combinations provides valuable insights for the final selection of the optimal approach for the LC mapping task.

Five selected thematic land cover maps of the Gilgit-Baltistan in HKH region (with highest OA) produced in the GEE for the year 2019 using S-2 imagery, where ( a ) RF, b gmoMaxEnt, c CART, d SVM, and ( e ) minDistance

From the results presented in Table  4 , it is evident that the RF classifier generally had the highest OA and kappa values across all the evaluated scenarios. The SVM classifier had the second-highest OA and kappa values in some scenarios, but its performance was generally lower than that of RF. In contrast, the minDistance classifier exhibited the lowest OA and kappa values across all scenarios and among all the classifiers. The gmoMaxEnt and CART classifiers generally had an intermediate level of performance, with OA and kappa values that were lower than RF but higher than minDistance. To better represent and explain the results, the relatively low-performing scenarios for all classifiers were dropped, and only the top five scenarios were selected for further analysis and discussion (Fig.  3 ). This approach allows for a more focused and informative presentation of the most promising classification outcomes for Gilgit-Baltistan.

A PC coefficient of 0.9 with a p -value of 0.0 was observed between the OA and kappa values across all ML classifiers. This indicates a strong positive and statistically significant relationship between these two-performance metrics. The close correspondence between OA and kappa suggests that the kappa index was closely tied to the overall classification accuracy, such that higher OA values were consistently associated with higher kappa scores, and vice versa.

The analysis of individual LC class accuracies revealed several notable trends. The RF exhibited strong performance, achieving high PA of 0.97 and UA of 0.88 for the water class. In contrast, the SVM and minDistance were unable to match the UA and PA values attained by RF for this class. RF also demonstrated high UA (0.77) and PA (0.75) for the forest cover, as well as for the grasses (PA = 0.71, UA = 0.78). Among the classifiers, CART and minDistance showed the least underestimation for the grasses class, with UA = 0.55 and PA = 0.48. For agriculture class, RF achieved a high PA of 0.89 and UA of 0.78, while minDistance recorded relatively lower UA and PA values (UA = 0.60, PA = 0.65). The RF classifier also performed exceptionally well for the wetland class, with UA of 0.85 and PA of 0.79, whereas minDistance exhibited considerably lower UA (0.53) and PA (0.69) values. Interestingly, SVM attained the highest PA (0.93) but the lowest UA (0.59) among all classifiers for the barren land class. Similarly, for the built-up areas, RF achieved the highest UA (0.89), while minDistance had the lowest UA (0.32). However, SVM had the lowest PA (0.07), whereas the gmoMaxEnt classifier recorded the highest PA (0.68) for the built-up class. In the case of glacier classification, CART achieved the highest PA (0.59), while gmoMaxEnt had the lowest PA (0.14). RF, on the other hand, attained the highest UA (0.67), and minimum distance had the lowest UA (0.39). For the snow class, minDistance obtained the highest UA (0.97) and gmoMaxEnt had the lowest UA (0.75), while RF achieved the highest PA (0.97), and CART had the lowest PA (0.89). These findings suggest that classifiers utilizing a higher number of spectral bands generally exhibit more favourable UA and PA performance across the various LC classes.

Moreover, misclassification of LC classes was observed across all classification results. For instance, many barren land validation points were incorrectly labelled as agriculture, and similar issues occurred for built-up areas being mislabelled as water (Fig.  4 a–e). These misclassification patterns highlight the challenges in accurately discriminating certain LC types, likely due to spectral similarities or mixed pixels. Despite these issues, the comparison of OA and kappa index indicates that the RF classifier performed excellently among the five implemented algorithms, by exhibiting the highest and most consistent classification results.

Confusion matrix for best classification scenario with a RF with scenario S1, b gmoMaxEnt with scenario S1, c CART with scenario S2, d SVM with scenario S1, e minimumDistance with scenario S1. The LC classes are numbered 0 through 8, where 0 is water, 1 is forest, 2 is grasses, 3 is wetland, 4 is agriculture, 5 is barren land, 6 is build-up, 7 is glacier, and 8 is snow. UA represents user’s accuracy, and PA represents the producer’s accuracy

Moreover, we also compared and evaluated the classification performance of ML algorithms across different LC categories using precision, recall, and F1-score metrics (Fig.  5 ). These findings reveal that RF demonstrated superior precision, recall, and F1-scores for water bodies, wetlands, agriculture, barren land, and snow LC type. It achieves precision scores of 0.96, 0.83, 0.85, and 0.91, respectively. SVM and gmoMaxEnt also exhibit competitive precision scores, particularly in the water, wetland, and agriculture categories. However, all algorithms struggle to accurately classify the forest category, with significantly lower precision scores across the board. The minimumDistance algorithm consistently performs the poorest in terms of precision for all LC categories. When considering recall, RF again shows strong performance in correctly identifying LC class of water (0.94), grasses (0.67), and snow (0.98). However, SVM and gmoMaxEnt exhibit relatively lower recall scores, especially for the forest and wetland categories. The minimumDistance algorithm consistently has lower recall scores compared to other algorithms, indicating difficulties in correctly classifying various land LC. Comparing the performance of CART, it also shows mixed results across different LC categories. In terms of precision, CART performs relatively well for water (0.82), agriculture (0.71), and barren land (0.75). However, its precision scores are lower compared with RF in most categories. CART struggles particularly in the forest category, achieving a precision score of 0.57, which is significantly lower compared to other algorithms. In terms of recall, CART performs decently for water (0.86) and barren land (0.68) categories. However, it falls behind RF in terms of recall scores across most LC classes. It demonstrates challenges in correctly identifying instances of forest and wetland, with recall scores of 0.59 and 0.73, respectively.

Precision, recall, and F1-score comparison results for top five performing algorithms classification scenario against each LC class. The LC classes are numbered 0 through 8, where 0 is water, 1 is forest, 2 is grasses, 3 is wetland, 4 is agriculture, 5 is barren land, 6 is build-up, 7 is glacier, and 8 is snow. The vertical colour bar shows the accuracy of classification as low (0) with red and high (1) with green

Additionally, focusing on the F1-score (Fig.  5 c), RF consistently achieves high scores for water, agriculture, barren land, and snow categories, indicating a balanced performance in terms of precision and recall. SVM and gmoMaxEnt also exhibit competitive F1-scores for the water, wetland, and agriculture categories. Like the other metrics, the forest category poses a challenge for all algorithms, resulting in relatively lower F1-scores. The minDistance algorithm consistently performs the worst across all LC categories, while the CART achieves moderate F1-scores scores for water (0.75) and barren land (0.71) categories. However, its overall performance is lower compared to RF and SVM in most LC categories.

Comparing the performance among LC classes (Fig.  5 ), it is evident that the algorithms excel in different categories. RF consistently performs well in water, agriculture, barren land, and snow. SVM and gmoMaxEnt showcase strengths in water and wetland categories. However, all algorithms struggle with accurately classifying the forest category, indicating the inherent complexity of distinguishing forest cover. CART, on the other hand, lags behind RF, SVM, and gmoMaxEnt in terms of F1-scores across most LC categories. It struggles in accurately classifying the forest category, resulting in a relatively lower F1-score of 0.58. Moreover, the F1-scores algorithm consistently performs the poorest among all the algorithms, indicating limitations in accurately classifying LC categories.

These results suggest that RF is the most reliable algorithm across multiple LC categories, demonstrating superior precision, recall, and F1-scores. However, the performance of each algorithm varies depending on the specific LC class. These findings emphasize the importance of selecting appropriate ML models based on the target LC type. Further research and experimentation are required to uncover the underlying factors causing the observed performance variations and to refine the classification accuracy for challenging LC categories, such as forests.

Land Cover Classification of Gilgit-Baltistan

The classified maps (Fig.  3 ) based on ten different scenarios (Table  3 ) and five ML classifiers were visually the same and consistent, displaying all results was not possible due to limitation of space. Based on accuracy assessment (OA and Kappa), RF (scenario S1) was found to be the best ML classifier among all. Therefore, RF LC classification was selected as a final map and used for area distribution of each LC class in the Gilgit-Baltistan (Fig.  6 ).

Final classified maps of Gilgit-Baltistan produced by RF under scenario S1 in GEE Platform for year 2019 using Sentinel-2 Satellite data

The final RF-based LC results (Fig.  6 and Fig.  7 ) depict that barren land (33,201.77 km 2 ) is the largest LC type classified in the study area, followed by snow cover (16,275.59 km 2 ), glacier (5651.42 km 2 ), grasses (5141.72 km 2 ), water (3356.22 km 2 ), wetland (2063.60 km 2 ), build-up (1928.81 km 2 ), agriculture (1316.73 km 2 ), and forest (868.04 km 2 ).

Area estimates of each land cover class resulting from image classification using best performing RF classifier. The label on each bar represents the area of land cover class in km 2

The study area is predominantly classified as barren land, which is distributed generally throughout the region, while the snow and glacier classes were found to be distributed primarily across the northern and eastern sides. The agriculture class has a distinct distribution, concentrated along the river and stream channels, which also contain the human settlements and built-up areas distributed throughout the valleys. Forest and wetland classes were found to be concentrated along the high-altitude ridges and rangelands, mostly between the south-west and north of the study area, while grasses were distributed in the northern, southern, south-eastern, and north-western parts, comprising the alpine pastures and rangelands that are usually covered with snow during the winter months (approximately 6 months). The built-up class in GB is typically distributed in the valleys, surrounded by agricultural and uncultivated lands, with the most concentrated areas observed in the main cities of Astore, Ghizer, Skardu, Chilas, and Gilgit (Ali et al. 2019a , b), providing insights into the landscape characteristics and human–environment interactions within the study area.

The district-wise area estimates of each LC class (Table  5 ) revealed that water is concentrated in Hunza district (1099.85 km 2 ), district Diamer has the largest area of forest (353.66 km 2 ), grasses (1162.57 km 2 ), and wetland (644.36 km 2 ), while the district Astore covers the largest area of agricultural land. Moreover, barren land and build-up were distributed largely in district Ghizer with an area of 7053.31 km 2 and 329.99 km 2 , respectively, while glacier (1751.57 km 2 ) and snow (4114.97 km 2 ) classes contribute to the highest area in district Hunza and Shigar, respectively.

Advantages and Opportunities of GEE Cloud Platform

Mapping LC features over large areas often faces challenges due to the limited availability and inconsistency of cloud-free satellite imagery. In this study, these challenges were addressed by employing temporal aggregation of Sentinel-2 imagery to create a comprehensive LC map for the Gilgit-Baltistan region of HKH. Previous classification efforts in this area had been hindered by the shortage of cloud-free images. The introduction of the web-based GEE cloud platform has significantly addressed the computational limitations that often hinder LC mapping in developing countries (Li et al. 2020 ; Zhou et al. 2020 ).

Few localized studies have been conducted in the study area, applying traditional approaches for LC mapping. However, as already mentioned, these previous efforts have utilized fewer LC classes or focused on easily distinguishable classes like barren land, forests, or water bodies in small sub-basin areas (Khan et al. 2020a , b; Qamer et al. 2016 ). Comparing our results with previous studies is challenging due to differences in methodologies and data sources. Previous studies in the area utilized Landsat-8 (L-8), Landsat-7/5/4, and MODIS products, while our study stands out as the first to utilize Sentinel-2 data. This is significant because Sentinel-2 offers shorter revisit times and higher-resolution imagery, enabling more accurate and practical LC classification in our target area.

The accuracy evaluation in this study demonstrates that the use of the GEE cloud platform enables robust and accurate regional scale LC mapping. GEE offers key advantages, such as the ability to quickly and precisely select sample points using supplementary data like socio-economic information, population, DMSP-OLS, DEM products, and satellite datasets, surpassing the capabilities of traditional tools. Moreover, GEE scripts can be easily enhanced for long-term monitoring of LC changes and driving indicators. Collaboration among stakeholders and agencies can further strengthen the capacity to address challenges like food security and flood mapping using open-source geospatial data. One such example is the SERVIR program, a joint NASA and USAID initiative that helps developing countries utilize Earth observation satellites and geospatial technologies (SERVIR 2005 ). A similar collaborative approach can be adopted in the Gilgit-Baltistan for improved planning and decision-making. The use of the GEE cloud platform has enabled the development of high-resolution, regional-scale LC products, overcoming the computational limitations that had previously hindered such large-scale mapping efforts (Faqe Ibrahim et al. 2023 ). This innovative approach provides a strong foundation for future LC monitoring and change analysis to support sustainable management in the study region.

Strengths and Implication of Our Land Cover Classification

The LC information on the nine classes mapped (Fig.  6 ) in this study is essential for improving our understanding of various earth surface processes in Gilgit-Baltistan. For instance, the delineation of water and glacier bodies can enable disaster monitoring of GLOFs, which are a frequent occurrence in the study area (Jamil et al. 2019 ).

During the current study, ML classifiers under investigation were not exhaustively tuned using their hyperparameters. Instead, the models were left to operate independently using their default input settings to map the land features based on the same training dataset. This approach identified the most efficient ML classifier, which upon further hyperparameter tuning could potentially achieve exceptional results. Choosing a set of optimal hyperparameters for a ML model is time exhausting process and is user-dependent which may affect the classification accuracy. Although default values for these parameters are usually suggested, to ensure that the accurate classification has been produced (Maxwell et al. 2018 ). The optimal parameters of the model vary from area to area depending on the quality of the dataset and the number of sample points, spatial distribution, and derivative features such as texture and spectral indices (Tsai et al. 2018 ). However, the limitation of less quantity of sample points in the current study was likely addressed and improved by the bagging method of RF classifier (Breiman et al. 1984 ) which performed very well in the current study. To overcome such limitations under various scenarios (Table  3 ), RF is considered one of the best ML classifiers and has been tested widely by many researchers (Gargiulo et al. 2020 ; Pradhan et al. 2020 ; Stromann et al. 2020 ). Thus, the current approach provides an opportunity for future studies in the Gilgit-Baltistan to select RF for LC classification, which upon hyperparameter tuning would provide excellent classification accuracy.

However, to estimate the extent of LC classes, various regional and global LC datasets exist which are prepared using different imagery (Landsat-5/7/8, S-2, and MODIS, etc.), algorithms, and with varying resolutions. Accuracy of these products is needed to be improved when applied at local or regional level (Wagle et al. 2020 ) and specially in the mountainous regions. Due to multiple differences such as input data, classifier type, data acquisition time, and spatial resolution, it creates poor agreement among different products when applied at the regional or global level. Thus, producing a reliable local or regional level LC classification products is essential and applicable. Figure comparing the multiple global LC models and current study results is included in the supplementary file (Fig. S1 ), which provide evidence for the stated argument.

During the study, few challenges were encountered, mostly pertaining to uncertainties between build-up, water, agriculture, wetland, and barren land mapping producing misclassification of these LC classes probably due to similar or alike spectral responses (Fig.  8 and 9 ). This problem was more related in villages where the houses are of masonry type with roofs covered with clay and sand (Rafi et al. 2016 ), causing misclassification among barren land and build-up class. Also, it was difficult to distinguish between croplands and seasonal grasslands using the S-2 imagery due to overlapping phenology signatures of agriculture, forest, and grasslands especially in case of their sparse presence (Fig.  8 a). We tackled these difficulties by acquiring high-quality samples from secondary data, as well as using the high-resolution Google Earth data (as reference). However, higher mapping accuracy would likely be achieved with larger and more accurate training datasets along with hyperparameter tuning of classifier (Ka & Sa 2018 ; Tsai et al. 2018 ).

Comparison of accuracy of various LC classes from current study results and with ESA WorldCover product based on Senitnel-2 and RF model at different locations. a , d , g show basemap from Maxar, b , e , h are results from current study, and c , f , i represent the ESA global LC product, whereas ( j ) is orthorectified image acquired using unmanned aerial vehicle (UAV) in year July 2022 for location ( a )

Comparison of Sentinel-2 SR and wavelength (nm) response for each land cover classes at same locations. a Reflectance over median composite image used for land cover analysis. b Reflectance over single image acquired on 5 July 2019. The grey rectangles illustrate the wavelength ranges for each Sentinel-2 band

To establish a rigorous comparison and ensuring scientific validity of results of our LC model and global LC product, we utilized ESA WorldCover product (Chaaban et al. 2022 ; Zanaga et al. 2022 ) which has same spatial resolutions and is also produced using RF model with OA of 74.4% (Fig.  8 ). It is observed that our RF-based LC model achieves superior results at the local scale and can further improve with hyperparameter tuning. For instance, Fig.  8 a–c reveals agricultural fields that are clearly visible and accurately mapped by our model. In contrast, the ESA WorldCover has inaccurately classified these areas as shrubland or grasslands. The reference image (Fig.  8 j) acquired by an unmanned aerial vehicle (UAV) over the same area further corroborates the accurate mapping of agricultural fields by our model. Also, Fig.  8 d–f illustrates an example where a barren land (braided river) was misclassified as water by WorldCover product (Fig.  7 f), while the same extent is accurately mapped by current study model (Fig.  8 e). Additionally, Fig.  8 i highlights areas of misclassification compared to Fig.  8 g–h, where the water class (water stream/river) in Fig.  8 i was misclassified as shrubs, trees, and bare classes. Nevertheless, the ESA world cover product has better resolution as compared with other global products, but producing global scale LC maps still remains a challenge. In such cases, regional-level studies prove to be a better option, offering improved efficiency and OA of LC products. Practitioners at the local and regional level can utilize regional accurate data products for planning and designing new development programs for local communities.

Another aspect of current classification is the use of median operation for temporal aggregation of satellite imagery which might influence the overall classification accuracy. The use of median composite from Sentinel-2 data offers significant advantages LC mapping. To understand this, spectral response of median composite using all images (Fig.  9 a) and a single image at same locations is provided (Fig.  9 b). By combining data from multiple acquisitions, the median composite effectively reduces the impact of atmospheric conditions and minimizes temporal variability, resulting in more accurate and reliable reflectance values. The response of LC classes remained equivalent without showing irregular responses which suggests that the median composite has no erroneous data which might have negatively influenced the overall classification. However, there is an observed overlap in the median composite SWIR region (Band 12) for snow and water (Fig.  9 a), which can be attributed the common physical properties of these two materials, such as their high reflectance in the NIR region and their relatively low absorption in the SWIR region (Shao et al. 2020 ). However, RF is well suited for tackling the challenge of overlapping reflectance values among various LC classes by employing ensemble learning. RF combines multiple decision trees trained on random subsets of the data from all bands, introducing diversity into the model. This allows the algorithm to capture a broader range of spectral patterns and features beyond the overlapping wavelengths, enhancing the discrimination between snow and water. Also, previous studies (Phan et al. 2020 ; Xie et al. 2019 ) have also validated the accuracy of median operation for preparing composite image.

Conclusions

The Sentinel-2 is exceptional among presently operating Earth-observation satellites due to its wide spectral wavelength, its 10 m spatial resolution, and revisit time. This study represents a first assessment and evaluation of five ML algorithms using GEE to classify LC classes using temporally aggregated Sentinel-2 data for the year 2019 (May–September) based on ten scenarios (50 LC products) over a complex mountainous environment. The five tested ML algorithms produced OA ranging between 0.59 and 0.79, without any hyperparameter tuning. Among these classifiers, two of the ML algorithms, RF and gmoMaxEnt performed exceptionally well with scenario S1, while CART (S2 scenario) and SVM (S1 scenario) performed ordinary with a difference of OA of 0.06 and 0.14 as compared with RF classifier (S1 scenario), respectively.

Moreover, the current study has compared GEE’s in-build ML algorithms using default input parameter values to remove biasness among classifiers, providing consistent environment. Doing so, the OA-based evaluation identified RF classifier best suitable for mapping mountainous areas like Gilgit-Baltistan with complex mountain system. Therefore, in the future, the best identified RF classifier with scenario S1 within GEE environment should be used for advance multi-source data image classification with hyperparameter tuning to increase overall OA and better prediction. Also, it is suggested to build the capacity of various stakeholders in Gilgit-Baltistan for better monitoring the LC changes and resource management using big data coupled with the GEE cloud platform. 

Data Availability

The final land cover product and code is available at https://github.com/gomalhunzai/Gilgit-Baltistan-LandCover-GEE . Additional data will be made available on request.

Adepoju KA, Adelabu SA (2020) Improving accuracy evaluation of Landsat-8 OLI using image composite and multisource data with Google Earth Engine. Remote Sens Lett 11(2):107–116. https://doi.org/10.1080/2150704X.2019.1690792

Article   Google Scholar  

Ahmed N, Islam MN, Hasan MF, Motahar T, Sujauddin M (2019) Understanding the political ecology of forced migration and deforestation through a multi-algorithm classification approach: the case of Rohingya displacement in the southeastern border region of Bangladesh. Geol, Ecol, Landscapes 3(4):282–294. https://doi.org/10.1080/24749508.2018.1558025

Ali K, Bajracharya RM, Chapagain NR, Raut N, Kumar B, Begum F, Khan MZ, Ali M, Ahmed A (2019) Analyzing Land Cover Change Using Remote Sensing and GIS : a Case Study of Gilgit 10(1):100–105

Google Scholar  

Ali S, Biermanns P, Haider R, Reicherter K (2019) Landslide susceptibility mapping by using a geographic information system (GIS) along the China-Pakistan Economic Corridor (Karakoram Highway), Pakistan. Nat Hazard 19(5):999–1022. https://doi.org/10.5194/nhess-19-999-2019

Alifu H, Vuillaume JF, Johnson BA, Hirabayashi Y (2020) Machine-learning classification of debris-covered glaciers using a combination of Sentinel-1/-2 (SAR/optical), landsat 8 (thermal) and digital elevation data. Geomorphology 369:107365. https://doi.org/10.1016/j.geomorph.2020.107365

Amin G, Haroon E, Imtiaz I, Saqib NU, Shahzad MI (2021) Ecotourism potential assessment for Gilgit-Baltistan, Pakistan using integration of GIS, remote sensing AHP and crowd-sourced data. Geocarto Int 37(25):1–20. https://doi.org/10.1080/10106049.2021.2005157

Bacha AS, Shafique M, van der Werff H (2018) Landslide inventory and susceptibility modelling using geospatial tools, in Hunza-Nagar valley, northern Pakistan. J Mt Sci 15(6):1354–1370. https://doi.org/10.1007/s11629-017-4697-0

Belgiu M, Drăgu L (2016) Random forest in remote sensing: a review of applications and future directions. ISPRS J Photogramm Remote Sens 114:24–31. https://doi.org/10.1016/j.isprsjprs.2016.01.011

Benesty J, Chen J, Huang Y, Cohen I (2009) Pearson Correlation Coefficient BT - Noise Reduction in Speech Processing In: Cohen Y, Huang J, Chen J, Benesty (eds.); pp. 1–4. Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-00296-0_5

Beuchle R, Grecchi RC, Shimabukuro YE, Seliger R, Eva HD, Sano E, Achard F (2015) Land cover changes in the Brazilian Cerrado and Caatinga biomes from 1990 to 2010 based on a systematic remote sensing sampling approach. Appl Geogr 58:116–127. https://doi.org/10.1016/j.apgeog.2015.01.017

Bolch T, Kulkarni A, Kääb A, Huggel C, Paul F, Cogley JG, Frey H, Kargel JS, Fujita K, Scheel M (2012) The state and fate of Himalayan glaciers. Science 336(6079):310–314

Article   CAS   Google Scholar  

Boser BE, Guyon IM, Vapnik VN (1992) A training algorithm for optimal margin classifiers. In: Proceedings of the fifth annual workshop on Computational learning theory (COLT '92). Association for Computing Machinery, New York, 44–152. https://doi.org/10.1145/130385.130401

Breiman L, Friedman J, Stone CJ, Olshen RA (1984) Classification and regression trees. CRC Press

Carrasco L, O’Neil AW, Daniel Morton R, Rowland CS (2019) Evaluating combinations of temporally aggregated Sentinel-1, Sentinel-2 and Landsat 8 for land cover mapping with Google Earth Engine. Remote Sens 11(3):288. https://doi.org/10.3390/rs11030288

Chaaban F, El Khattabi J, Darwishe H (2022) Accuracy assessment of ESA WorldCover 2020 and ESRI 2020 land cover maps for a region in Syria. J Geovisualization Spat Anal 6(2):31. https://doi.org/10.1007/s41651-022-00126-w

Copernicus (2017) Sentinel-2 MSI: multispectral instrument, level-2A: Earth Engine Data Catalog. Google Earth Engine. https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_S2_SR_HARMONIZED

Cortes C, Vapnik V (1995) Support vector machine. Mach Learn 20(3):273–297

Dang AN, Kawasaki A (2017) Integrating biophysical and socio-economic factors for land-use and land-cover change projection in agricultural economic regions. Ecol Model 344:29–37. https://doi.org/10.1016/j.ecolmodel.2016.11.004

Delalay M, Tiwari V, Ziegler AD, Gopal V, Passy P (2019) Land-use and land-cover classification using Sentinel-2 data and machine-learning algorithms: operational method and its implementation for a mountainous area of Nepal. J Appl Remote Sens 13(01):1. https://doi.org/10.1117/1.jrs.13.014530

ESA (2015) Sentinel-2 user handbook. European Space Agency Standard Document 1:1–64

Faqe Ibrahim GR, Rasul A, Abdullah H (2023) Improving crop classification accuracy with integrated Sentinel-1 and Sentinel-2 data: a case study of barley and wheat. J Geovisualization Spat Anal 7(2):22. https://doi.org/10.1007/s41651-023-00152-2

Farr TG, Rosen PA, Caro E, Crippen R, Duren R, Hensley S, Kobrick M, Paller M, Rodriguez E, Roth L (2007) The shuttle radar topography mission. Rev Geophys, 45(2). https://doi.org/10.1029/2005RG000183

Gargiulo M, Dell’aglio DAG, Iodice A, Riccio D, Ruello G (2020) Integration of Sentinel-1 and Sentinel-2 data for land cover mapping using w-net. Sensors (Switzerland) 20(10):1–16. https://doi.org/10.3390/s20102969

Gilani H, ArifGoheer M, Ahmad H, Hussain K (2020) Under predicted climate change: distribution and ecological niche modelling of six native tree species in Gilgit-Baltistan Pakistan. Ecol Indic 111:106049. https://doi.org/10.1016/j.ecolind.2019.106049

Gislason PO, Benediktsson JA, Sveinsson JR (2006) Random forests for land cover classification. Pattern Recogn Lett 27(4):294–300. https://doi.org/10.1016/j.patrec.2005.08.011

Gorelick N, Hancher M, Dixon M, Ilyushchenko S, Thau D, Moore R (2017) Google Earth Engine: planetary-scale geospatial analysis for everyone. Remote Sens Environ 202:18–27. https://doi.org/10.1016/j.rse.2017.06.031

Gumma MK, Thenkabail PS, Teluguntla PG, Oliphant A, Xiong J, Giri C, Pyla V, Dixit S, Whitbread AM (2020) Agricultural cropland extent and areas of South Asia derived using Landsat satellite 30-m time-series big-data using random forest machine learning algorithms on the Google Earth Engine cloud. Gisci Remote Sens 57(3):302–322. https://doi.org/10.1080/15481603.2019.1690780

Hayes T, Usami S, Jacobucci R, McArdle JJ (2015) Using classification and regression trees (CART) and random forests to analyze attrition: results from two simulations. Psychol Aging 30(4):911–929. https://doi.org/10.1037/pag0000046

He LM, Kong FS, Shen ZQ (2005) Multiclass SVM based land cover classification with multisource data. 2005 Int Conf Mach Learn Cyber, ICMLC 2005:3541–3545. https://doi.org/10.1109/icmlc.2005.1527555

Hill J, Stellmes M, Udelhoven T, Röder A, Sommer S (2008) Mediterranean desertification and land degradation: mapping related land use change syndromes based on satellite observations. Global Planet Change 64(3–4):146–157

Hu Y, Hu Y (2019) Land cover changes and their driving mechanisms in Central Asia from 2001 to 2017 supported by Google Earth Engine. Remote Sens 11(5):564. https://doi.org/10.3390/rs11050554

Hu Y, Dong Y, Batunacun. (2018) An automatic approach for land-change detection and land updates based on integrated NDVI timing analysis and the CVAPS method with GEE support. ISPRS J Photogramm Remote Sens 146(October):347–359. https://doi.org/10.1016/j.isprsjprs.2018.10.008

Hussain A, Bangash R (2017) Impact of climate change on crops’ productivity across selected agro-ecological zones in Pakistan. Pakistan Dev Rev 56(2):163–187. https://doi.org/10.30541/v56i2pp.163-187

Jamil A, Khan AA, Bayram B, Iqbal J, Amin G, Yesiltepe M, & Hussain D (2019) Spatio-temporal glacier change detection using deep learning: a case study of Shishper glacier in Hunza. International Symposium on Applied Geoinformatics , 5. https://www.researchgate.net/publication/337720130

Jia M, Wang Z, Mao D, Ren C, Song K, Zhao C, Wang C, Xiao X, Wang Y (2023) Mapping global distribution of mangrove forests at 10-m resolution. Sci Bullet 68(12):1306–1316

Johansen K, Phinn S, Taylor M (2015) Mapping woody vegetation clearing in Queensland, Australia from Landsat imagery using the Google Earth Engine. Remote Sens Applic: Soc Environ 1:36–49. https://doi.org/10.1016/j.rsase.2015.06.002

Ka A, & Sa A (2018) Improved Landsat-8 Oli and Sentinel-2 MSI classification in mountainous terrain using machine learning on Google Earth Engine. October

Khan AA, ul Hassan SN, Baig S, Khan MZ, Muhammad A (2019) The response of land surface temperature to the changing land-use land-cover in a mountainous landscape under the influence of urbanization: Gilgit city as a case study in the Hindu Kush Himalayan Region of Pakistan. Int J Econ Environ Geol 10(3):40–49

Khan AA, Jamil A, Hussain D, Taj M, Jabeen G, Malik MK (2020) Machine-learning algorithms for mapping debris-covered glaciers: the Hunza basin case study. IEEE Access 8:12725–12734. https://doi.org/10.1109/ACCESS.2020.2965768

Khan A, Said A, Ullah I (2020) Landsat based distribution mapping of high-altitude peatlands in Hindu Kush Himalayas — a case study of Broghil Valley. Pakistan J Mountain Sci 17(1):42–49. https://doi.org/10.1007/s11629-019-5384-0

Lee J, Cardille JA, Coe MT (2018) BULC-U: Sharpening resolution and improving accuracy of land-use/land-cover classifications in Google Earth Engine. Remote Sens 10(9):1–21. https://doi.org/10.3390/rs10091455

Li Q, Qiu C, Ma L, Schmitt M, Zhu XX (2020) Mapping the land cover of africa at 10 m resolution from multi-source remote sensing data with google earth engine. Remote Sens 12(4):1–22. https://doi.org/10.3390/rs12040602

Mahmoodzada AB, Das P, Varade D, Akhtar MA, Shimada S (2024) High-resolution mapping of seasonal snow cover extent in the Pamir Hindu Kush using machine learning-based integration of multi-sensor data. Acta Geophys 72(2):1455–1470. https://doi.org/10.1007/s11600-023-01281-4

Mansaray LR, Wang F, Huang J, Yang L, Kanu AS (2019) Accuracies of support vector machine and random forest in rice mapping with Sentinel-1a, Landsat-8 and Sentinel-2a datasets. Geocarto Int 35(10):1–21. https://doi.org/10.1080/10106049.2019.1568586

Maxwell AE, Warner TA, Fang F, Maxwell AE, Warner TA, Implementation FF, Maxwell AE, Warner TA (2018) Implementation of machine-learning classification in remote sensing : an applied review. Int J Remote Sens 39(9):2784–2817. https://doi.org/10.1080/01431161.2018.1433343

McCord SE, Buenemann M, Karl JW, Browning DM, Hadley BC (2017) Integrating remotely sensed imagery and existing multiscale field data to derive rangeland indicators: application of Bayesian additive regression trees. Rangel Ecol Manage 70(5):644–655. https://doi.org/10.1016/j.rama.2017.02.004

Mcdonald R, Mohri M, Silberman N, Walker D, Mann G (2009). Efficient large-scale distributed training of conditional maximum entropy models. Adv Neural Inform Process Syst, 1231–1239. https://doi.org/10.1007/s13218-010-0039-x

Melgani F, Bruzzone L (2004) Classification of hyperspectral remote sensing. IEEE Trans Geosci Remote Sens 42(8):1778–1790

Mondal P, Liu X, Fatoyinbo TE, Lagomasino D (2019) Evaluating combinations of Sentinel-2 data and machine-learning algorithms for mangrove mapping in West Africa. Remote Sens 11(24):2928. https://doi.org/10.3390/rs11242928

Moreno R, Zamora R, Molina JR, Vasquez A, Herrera MÁ (2011) Predictive modeling of microhabitats for endemic birds in South Chilean temperate forests using maximum entropy (Maxent). Eco Inform 6(6):364–370

Nawaz MA, Khan AA, Khalid U, Buerkert A, Wiehle M (2019) Superfruit in the Niche—underutilized Sea Buckthorn in Gilgit-Baltistan. Pakistan Sustain 11(20):5840

Phan TN, Kuch V, Lehnert LW (2020) Land Cover Classification using Google Earth Engine and random forest classifier—the role of image composition. Remote Sens 12(15):2411. https://doi.org/10.3390/rs12152411

Pradhan B, Al-Najjar HAH, Sameen MI, Tsang I, Alamri AM (2020) Unseen land cover classification fromhigh-resolution orthophotos using integration of zero-shot learning and convolutional neural networks. Remote Sens 12(10):1–26. https://doi.org/10.3390/rs12101676

Qamer FM, Shehzad K, Abbas S, Murthy MSR, Xi C, Gilani H, Bajracharya B (2016) Mapping deforestation and forest degradation patterns in Western Himalaya. Pakistan Remote Sens 8(5):1–17. https://doi.org/10.3390/rs8050385

Rafi MM, Lodi SH, Ahmed M, Kumar A, Verjee F (2016) Development of building inventory for northern Pakistan for seismic risk reduction. Int J Dis Resilience Built Environ 7(5):501–20

Rahim I, Ali SM, Aslam M (2018) GIS based landslide susceptibility mapping with application of analytical hierarchy process in District Ghizer, Gilgit Baltistan Pakistan. J Geosci Environ Protect 06(02):34–49. https://doi.org/10.4236/gep.2018.62003

Rankl M, Kienholz C, Braun M (2014) Glacier changes in the Karakoram region mapped by multimission satellite imagery. Cryosphere 8(3):977–989. https://doi.org/10.5194/tc-8-977-2014

Rapinel S, Panhelleux L, Gayet G, Vanacker R, Lemercier B, Laroche B, Chambaud F, Guelmami A, Hubert-Moy L (2023) National wetland mapping using remote-sensing-derived environmental variables, archive field data, and artificial intelligence. Heliyon 9(2):1–17. https://doi.org/10.1016/j.heliyon.2023.e13482

Rudiyanto MB, Shah RM, CheSoh N, Arif C, IndraSetiawan B (2019) Automated near-real-time mapping and monitoring of rice extent, cropping patterns, and growth stages in Southeast Asia using Sentinel-1 time series on a Google Earth Engine platform. Remote Sens 11(14):1–27. https://doi.org/10.3390/rs11141666

Saini R, Singh S (2024) Land use land cover mapping and snow cover detection in Himalayan region using machine learning and multispectral Sentinel-2 satellite imagery. Int J Inf Technol 16(2):675–686. https://doi.org/10.1007/s41870-023-01673-1

Satti Z, Naveed M, Shafeeque M, Ali S, Abdullaev F, Ashraf TM, Irshad M, Li L (2023) Effects of climate change on vegetation and snow cover area in Gilgit Baltistan using MODIS data. Environ Sci Pollut Res 30(7):19149–19166

Satti Z, Naveed M, Shafeeque M, Li L (2024) Investigating the impact of climate change on trend shifts of vegetation growth in Gilgit Baltistan. Global Planetary Chang 232:104341. https://doi.org/10.1016/j.gloplacha.2023.104341

Servir (2005) Servir global: connecting space to village

Shao Y, Lunetta RS (2012) Comparison of support vector machine, neural network, and CART algorithms for the land-cover classification using limited training data points. ISPRS J Photogramm Remote Sens 70:78–87

Shao D, Xu W, Li H, Wang J, Hao X (2020) Modeling snow surface spectral reflectance in a land surface model targeting satellite remote sensing observations. Remote Sens 12(18):3101. https://doi.org/10.3390/rs12183101

Shetty S, Gupta PK, Belgiu M, Srivastav SK (2021) Assessing the effect of training sampling design on the performance of machine learning classifiers for land cover mapping using multi-temporal remote sensing data and Google Earth Engine. Remote Sens 13(8):1433. https://doi.org/10.3390/rs13081433

Silveira EMO, Radeloff VC, Martinuzzi S, Pastur GJM, Bono J, Politi N, Lizarraga L, Rivera LO, Ciuffoli L, Rosas YM (2023) Nationwide native forest structure maps for Argentina based on forest inventory data, SAR Sentinel-1 and vegetation metrics from Sentinel-2 imagery. Remote Sens Environ 285:113391

Singh R, Mishra AP, Kumar M, & Pande CB (2023) Classification of Vegetation Types in the Mountainous Terrain Using Random Forest Machine Learning Technique. In Climate Change Impacts on Natural Resources, Ecosystems and Agricultural Systems (pp. 615–628). Springer. https://doi.org/10.1007/978-3-031-19059-9_27

Stromann O, Nascetti A, Yousif O, Ban Y (2020) Dimensionality reduction and feature selection for object-based land cover classification based on Sentinel-1 and Sentinel-2 time series using Google Earth Engine. Remote Sens 12(1):76. https://doi.org/10.3390/RS12010076

Tadono T, Ishida H, Oda F, Naito S, Minakawa K, Iwamoto H (2014) Precise global DEM generation by ALOS PRISM. ISPRS Annal Photogramm, Remote Sens Spat Inform Sci 2(4):71

Tsai YH, Stow D, Chen HL, Lewison R, An L, Shi L (2018) Mapping vegetation and land use types in Fanjingshan National Nature Reserve using google earth engine. Remote Sens 10(6):927. https://doi.org/10.3390/rs10060927

us Saqib N, Yaqub A, Amin G, Khan I, Ajab H, Zeb I, Ahmad D (2019) The impact of tourism on local communities and their environment in Gilgit Baltistan, Pakistan: a local community perspective. Environ Socio-Econ Stud 7(3):24–37. https://doi.org/10.2478/environ-2019-0015

Wagle N, Acharya TD, Kolluru V, & Huang H (2020) Applied sciences multi-temporal land cover change mapping using Google Earth Engine and ensemble learning methods https://doi.org/10.3390/app10228083

Wang M, Mao D, Wang Y, Xiao X, Xiang H, Feng K, Luo L, Jia M, Song K, Wang Z (2023) Wetland mapping in East Asia by two-stage object-based random forest and hierarchical decision tree algorithms on Sentinel-1/2 images. Remote Sens Environ 297(August):113793. https://doi.org/10.1016/j.rse.2023.113793

Xie Y, Lark TJ, Brown JF, Gibbs HK (2019) Mapping irrigated cropland extent across the conterminous United States at 30 m resolution using a semi-automatic training approach on Google Earth Engine. ISPRS J Photogramm Remote Sens 155(February):136–149. https://doi.org/10.1016/j.isprsjprs.2019.07.005

Xiong J, Thenkabail PS, Tilton JC, Gumma MK, Teluguntla P, Oliphant A, Congalton RG, Yadav K, Gorelick N (2017) Nominal 30-m cropland extent map of continental Africa by integrating pixel-based and object-based algorithms using Sentinel-2 and Landsat-8 data on Google Earth Engine. Remote Sens 9(10):1–27. https://doi.org/10.3390/rs9101065

Xu J, Grumbine RE, Shrestha A, Eriksson M, Yang X, Wang Y, Wilkes A (2009) The melting Himalayas: cascading effects of climate change on water, biodiversity, and livelihoods. Conserv Biol 23(3):520–530. https://doi.org/10.1111/j.1523-1739.2009.01237.x

Yuan H, Van Der Wiele CF, Khorram S (2009) An automated artificial neural network system for land use/land cover classification from landsat TM imagery. Remote Sens 1(3):243–265. https://doi.org/10.3390/rs1030243

Zanaga D, Van De Kerchove R, Daems D, De Keersmaecker W, Brockmann C, Kirches G, Wevers J, Cartus O, Santoro M, Fritz S, Lesiv M, Herold M, Tsendbazar NE, Xu P, Ramoino F, Arino O, Zanaga D, Van De Kerchove R, Daems D (2022) WorldCover 2020:v100. https://doi.org/10.5281/zenodo.7254221

Zhang D-D, Zhang L (2020) Land cover change in the central region of the Lower Yangtze River based on landsat imagery and the Google Earth Engine: a case study in Nanjing. China Sensors 20(7):2091. https://doi.org/10.3390/s20072091

Zhou B, Okin GS, Zhang J (2020) Leveraging Google Earth Engine (GEE) and machine learning algorithms to incorporate in situ measurement from different times for rangelands monitoring. Remote Sens Environ 236:111521. https://doi.org/10.1016/j.rse.2019.111521

Zurqani HA (2024) High-resolution forest canopy cover estimation in ecodiverse landscape using machine learning and Google Earth Engine: validity and reliability assessment. Remote Sens Applic: Soc Environ 33:101095. https://doi.org/10.1016/j.rsase.2023.101095

Download references

Acknowledgements

The authors thank ESA for providing Copernicus Seninel-2 data and ESA WorldCover project [2020] and GEE team for provision of sample code and cloud computing platform for analysis. The data that support the findings of this study are openly available in Earth Engine Data Catalog. The authors would also like to thank JAXA team for provision of ALOS DSM ( https://ror.org/059yhyy33 ) and NOAA’s National Geophysical Data Center for DMPS/OLS ( https://ror.org/02z5nhe81 ) data and Google Earth team for high-resolution imagery. The author would also like to thank the respective institutions for providing environment and platform for conducting the research.

Open access funding provided by The Hong Kong Polytechnic University This work is supported by The Hong Kong Polytechnic University’s Start-up Fund for RAPs under the Strategic Hiring Scheme [Project ID: P0044784] and The Hong Kong Polytechnic University’s Research Institute for Sustainable Urban Development [Project ID: 1-BBG2].

Author information

Authors and affiliations.

Earth and Atmospheric Remote Sensing Lab (EARL), Department of Meteorology, COMSATS University Islamabad, Islamabad, 45550, Pakistan

Gomal Amin, Iqra Imtiaz, Ehsan Haroon & Muhammad Imran Shahzad

Department of Land Surveying and Geo-Informatics, The Hong Kong Polytechnic University, Hong Kong, SAR, China

Gomal Amin & Majid Nazeer

Department of Environmental Sciences, Quaid-I-Azam University, Islamabad, 15320, Pakistan

Najum us Saqib

Research Institute of Land and Space, The Hong Kong Polytechnic University, Hong Kong, SAR, China

Majid Nazeer

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Gomal Amin .

Ethics declarations

Ethics approval.

The researchers confirm that they are abided to the ethical standards of conducting research.

Informed Consent

We affirm that all research participants gave their full consent.

Conflict of Interest

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (DOCX 616 KB)

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Amin, G., Imtiaz, I., Haroon, E. et al. Assessment of Machine Learning Algorithms for Land Cover Classification in a Complex Mountainous Landscape. J geovis spat anal 8 , 34 (2024). https://doi.org/10.1007/s41651-024-00195-z

Download citation

Accepted : 19 August 2024

Published : 30 August 2024

DOI : https://doi.org/10.1007/s41651-024-00195-z

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Supervised classification
• Sentinel-2 data
• Land cover classification
• Gilgit-Baltistan
• Google Earth Engine
• Find a journal
• Publish with us
• Track your research