PONE-D-24-14163A Quantitative Geospatial Analysis of the Risk that Boko Haram Will Target a SchoolPLOS ONE

Dear Dr. Subrahmanian,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

 Please accept our apologies for the delay in issuing an editorial decision. Unfortunately, we have to search for a new Academic Editor several times since you submitted your manuscript. 

The manuscript has now been evaluated by six reviewers, and their comments are available below.This is a large number of reviewers. However, we have had difficulty securing reviewers with the relevant expertise to assess the topic and methods of your study. Although all the reviewers have brought their own perspectives to bear, and all offer constructive criticism, please pay particular attention to the comments made by reviewers 2 and 5. Could you please revise the manuscript to carefully address the concerns raised?

Please submit your revised manuscript by Feb 02 2025 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

• A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Steve Zimmerman, PhD

Senior Editor, PLOS One

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please note that PLOS ONE has specific guidelines on code sharing for submissions in which author-generated code underpins the findings in the manuscript. In these cases, all author-generated code must be made available without restrictions upon publication of the work. Please review our guidelines at https://journals.plos.org/plosone/s/materials-and-software-sharing#loc-sharing-code and ensure that your code is shared in a way that follows best practice and facilitates reproducibility and reuse.

3. We note that Figures 1,2 and 5 in your submission contain [map/satellite] images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

1. You may seek permission from the original copyright holder of Figures 1,2 and 5 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an ""Other"" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

2. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: Partly

Reviewer #6: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: Yes

Reviewer #6: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: Yes

Reviewer #6: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: No

Reviewer #6: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The article is interesting and could be used for public awareness. The article needs further improvement. My comment are;

1. Citation is missing in 'A similar claim was repeated by Nigeria’s Information Minister in October 2019()'.

2. Give a predictive AI based solution for the problem mentioning the period it might take for a stable situation

Reviewer #2: In the manuscript, utilizing a geospatially tagged data set, the coauthors identified factors affecting attacks on schools, implemented machine learning models to quantify the likelihood that a school will be at risk of a Boko Haram attack and finally concluded with a policy recommendation. The work presented is dedicated to an interesting and important topic. I have multiple concerns are as follows:

1. Order of Figure 3a vs. 3b is wrong. And in the caption “2023 numbers only run …” 2023 seems redundant?

2. In section Socio-Economic Data, there is no clear definition of “risk score”. How it derived from the data? Whether it’s valid or not?

3. Figure 6 and 8 are switched? Figure 6, 7, 8 are with low resolution.

4. In the sentence below Table 1, Table 1B summarized the non-urban schools or urban schools?

5. In the machine learning based analysis section, It seems like the models were only fitted in training set but not validation set nor testing set. It may lead to overfit problem. And prediction accuracy in training set is not valid for prediction in future events.

6. In line 446, there are two “that”.

7. In Figure 9, results for k=1 were shown. How connect the results with results in Table 3 where models for k=1 provided worst performance? Brief discussion is welcome.

8. For section “Multivariate Machine Learning Inspired Statistical Inferences of Boko Haram School Attacks”, using odds ratio assessing the hypothesis is not justifiable enough. Other assessment such as sensitivity, specificity, PPV, NPV, accuracy, error rate, …, could be provided in addition to OR.

9. Figure 1 & 2 displays the risk of school attacks. How those risk estimated? Which method was used?

Reviewer #3: The research article presents a quantitative geospatial analysis of the risk factors associated with Boko Haram's targeting of schools in Nigeria over nearly 14 years, utilizing statistical inference and machine learning techniques. It identifies three main vulnerability factors: proximity to Boko Haram activity, weak security presence, and socioeconomic conditions. The study finds that both wealthy and poor areas are targeted for different reasons—wealthy areas for ransom and poor areas for capturing sex slaves and child soldiers. Advanced machine learning models, particularly the AdaBoost classifier, achieved an F1 score of 0.85 in predicting attacks within 2km of schools, with key predictive features including distance to security installations and previous attacks. The findings underscore the critical role of security presence in deterring attacks and highlight the complex interplay between socioeconomic factors and risk. Overall, this research provides valuable insights for policymakers and security forces, suggesting that enhanced protection strategies could be developed based on the identified risk factors and predictive models. However, there are some concerns, listed below:

1. The authors should try to incorporate time series analysis or recurrent neural networks to better capture the temporal dynamics of conflict evolution over time.

2. The authors should remove the words first ever from abstract and conclusion.

3. Clarify the rationale behind choosing specific machine learning algorithms.

4. Provide more details on data preprocessing and feature selection methods.

5. Explain how the 2km threshold for predictions was determined.

6. All the tables should include axis labels.

7. It would be helpful with provide visualizations (e.g., ROC curves, confusion matrices) to illustrate model performance.

8. Address potential biases in the dataset and discuss their implications

9. Provide a separate methods section on statistical methods and analysis

Reviewer #4: More details about the data sources, particularly the socioeconomic data and the locations of security installations, would improve the manuscript. Reproducibility depends on transparency on data reliability and potential biases in data sources.

Describe any imputation methods that were employed and include a detail on how any missing data was handled. It is also necessary to address the handling of outliers, especially in socioeconomic and activity factors.

The manuscript would benefit from a brief description of how insights from the AdaBoost model and decision trees could be applied in a practical setting, even if the research highlights the significance of features in prediction.

By employing quantitative techniques to examine the risk factors associated with Boko Haram's attacks on schools, this paper tackles a pressing and important subject. It makes a significant contribution to the fields of geospatial analysis and counterterrorism studies.

Despite their strength, machine learning models have drawbacks. Talk about possible drawbacks such as the model's tendency to overfit historical data, particularly in light of Boko Haram's changing strategies, and how it might apply to attacks in the future or in other regions.

Since predictive models can impact real-world security decisions, address any ethical considerations related to this research, such as the potential consequences of model inaccuracies.

Reviewer #5: The paper A Quantitative Geospatial Analysis of the Risk that Boko Haram Will Target a School submitted to PLOSOne makes a number of interesting contributions that empirically, theoretically and practically relevant in the study of conflict and international development. The paper develops a one-of-a-kind dataset on Boko Haram (BH) attacks with substantial contextual information. It is a step up from widely-used ACLED. It then develops a set of spatial measures to examine hypotheses about mechanisms that may be involved in BH attacks. There is reasonable observational evidence that the level of general BH activity in the area and proximity to security (e.g., police or military facilities) are associated with the likelihood of attack on a school. If the patterns of association have a causal underpinning, then the authors point to useful policy recommendations.

Overall, the paper may be acceptable for publication in PLOSOne with substantial revisions. I offer the following suggestions in the hope that they help the authors revise the paper for resubmission.

The general points to consider are:

1. The paper could use a careful editing for language.

2. The story gets lost in a relatively large number of tests and separate analyses that are not fully described. The authors might consider slimming down the manuscript, concentrate on solid narrative development, and move “nice to have but not essential” testing to an integrated Supplementary file.

3. The main measure is a bit confusing. I have a hard time visualizing what it means to measure school attacks within 10 km against BH activity within 50 km. What if there is more than one school within that 10 km radius and one of them is attacked? I think a visualization of the method would go a long way to better supporting the findings.

The remainder of my comments are in order of presentation and mix both larger issues with some smaller ones.

1. Pg. 2, lines 16-17, is a very powerful statement. It would punch even harder if it ended with a something like "...all of this impact arises from a group with an estimated N number of members/fighters." Bring home the point that asymmetry can still produce huge harm.

2. Pg. 2, lines 43-53. This is small, but I would hang the activity hypothesis first on "routine activities theory" with a secondary pointer to "journey-to-crime". The basic idea is that areas where BH are operating have established some type of (operational) routine. Targets that fall near or within that routine are easier to accommodate than ones that are outside of the routine.

@article{cohen1979rat,

author = {Cohen, L. E. and Felson, M.},

title = {Social change and crime rate trends: A routine activity approach},

journal = {Am Sociol Rev},

volume = {44},

pages = {588-608},

DOI = {10.2307/2094589},

year = {1979},

type = {Journal Article}

}

3. Pg. 3, lines 61-62. If you want an explanation for "why" this might be, you could argue that the police stations serve as a deterrent where BH fighters perceive a greater likelihood of getting caught if operating near those spaces.

@article{RN5221,

author = {Nagin, Daniel S.},

title = {Deterrence in the Twenty-First Century},

journal = {Crime and Justice},

volume = {42},

pages = {199-263},

DOI = {10.1086/670398},

year = {2013},

type = {Journal Article}

}

@article{RN2912,

author = {Loughran, Thomas A. and Paternoster, Raymond and Piquero, Alex R. and Pogarsky, Greg},

title = {On Ambiguity in Perceptions of Risk: Implications for Criminal Decision Making and Deterrence},

journal = {Criminology},

volume = {49},

number = {4},

pages = {1029-1061},

DOI = {10.1111/j.1745-9125.2011.00251.x},

year = {2011},

type = {Journal Article}

}

4. Pg. 3, lines 65-66. This probably directly relates to so-called social disorganization theory in criminology. A key citation is (though there are thousands of others too):

@article{RN3950,

author = {Bursik Jr., Robert J.},

title = {Social Disorganization and Theories of Crime and Delinquency: Problems and Prospects},

journal = {Criminology},

volume = {26},

number = {4},

pages = {519-552},

DOI = {10.1111/j.1745-9125.1988.tb00854.x},

year = {1988},

type = {Journal Article}

}

5. Pg. 3, line 68. Hypothesis 4 is not particularly clear to me. I think the authors are looking forward to their results.

6. Pg. 6, lines 201-2002. There are a few years with no recorded attacks. Is this because there were no attacks or because of other potential issues with data acquisition?

7. Pg. 6, lines 210-211. When are schools on break? “While non-school attacks are more or less uniformly distributed over the 12 months of the year, school attacks occur less frequently in January, August, and December.”

8. Pp. 6-8. Overall, a tabular summary of data might be more efficient use of space.

9. Pg. 8, lines 276-277. The core measure used in the paper needs better explanation. It is not clear if this is a measure that the authors have developed themselves or if it has a source in the literature. There are a couple of ways to understand the statement “a school attacked happened within k ={1,2,3,5,10} km”. This could mean that there are, say, 10 students from school s. They weren't kidnapped directly from the school s but at a location k-kilometers away from the location of s. Or, there is a school s_i and a school s_j. They are 4.3km apart. If schooI s_i is attacked then school s_j also is considered attacked within the 5km distance band of school s_i. What is the correct interpretation?

10. Pg. 8, lines 300-301 and Figure 6. The functional form of the points as you move from the k=1 to the k = 10 panel is very regular; basically moving from exponential-like to linear. This suggest to me that there is something fundamental about the geometry of the measurement units that is driving the pattern. It would be very useful to simulate a null model here to know how this measure behaves when there are no correlations in the data. For example, one could assume that all BH events are all 2D Poisson, attacked and non-attacked schools and security installations are also Poisson distributions of points. Then investigate how the measures change as discs of radius k are compared.

11. Pg. 13, lines 463-466. Finding 5. It seems unlikely for such extreme events, but is it possible that there is reporting bias that is correlated with proximity to police stations? That is, attacks farther from formal law enforcement rely on alternative means of solution (such as family, clan groups, local traditional leaders) and therefore are never reported to a formal source (news or police).

12. Pg. 14, lines 494-509 (and Table 4 on pg. 13). It is hard to intuit the rank order arrangement of distance to nth-closest security installation. It feels like there is some interaction between the closeness of security installations and the area size implied by k, but there’s no rhyme nor reason to it. Sure, it is ok to basically say this variable has predictive power, but it would also be nice to have a better understanding why in a behavioral (and geographic) sense certain orderings matter as one scale and not another. For example, why would the 3rd, 4th and 5th closest be important at k = 5, but not 1 and 2. Some explanation seems needed.

13. Pp. 13-14. The statistical testing associated with the decision tree analyses could use better explanation.

Reviewer #6: Overall, this is a very will written manuscript, doing a great job at spatially analyzing Boko Haram's attacks on Schools. There are only two concerns that I have in improving the manuscript.

First, the argument for hypothesis #3 is quite weak and the author(s) need to bolster the rationale for it.

The other addition that I would suggest is for the author(s) explicitly state what the unit of analysis is (i.e. the areal unit). Is it the school? The ward?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Linchen He

Reviewer #3: No

Reviewer #4: No

Reviewer #5: No

Reviewer #6: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

A Quantitative Geospatial Analysis of the Risk that Boko Haram Will Target a School

PONE-D-24-14163R1

Dear Dr. Subrahmanian,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. If you have any questions relating to publication charges, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Jessica Leight, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

Reviewer #5: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Partly

Reviewer #5: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

Reviewer #5: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

Reviewer #5: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

Reviewer #5: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: (No Response)

Reviewer #5: The authors have addressed all of my concerns. I believe this paper is an important contribution to the literature.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

Reviewer #5: No

**********