From Dropout to Classroom: The Role of Mexico’s PROGRESA Education Grants in Reenrollment
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThank you for your article. I believe that the submitted article has a number of merits, but it should be revised to highlight these merits. Here are my suggestions: simplifying and synthesizing the introduction and background
-introducing more recent data and results presented in other similar projects and initiatives in Latin America. Their presentation is mandatory precisely to see the sustainability of similar Prograssa initiatives.
-the Study part should be improved by clearly introducing the objectives of the study;
-the empirical model presented has value, but it needs to be explained in more detail regarding the objectives of the study and the research questions
-at point 5.2. the conclusion is not very clear (beyond the statistics presented)
-many of the aspects from 6 would have been better introduced in the research part for a better substantiation and an evolutionary perspective of the phenomenon.
-From the perspective of 6, the conclusions, I consider that some of the conclusions are too broad assumptions. I advise you to limit yourself to the context presented, to correlating the objectives with the results.
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThank you for the opportunity to review the paper titled “From Dropout to Classroom: The Role of Mexico’s PROGRESA Education Grants in Reenrollment.” This paper examines the impact of Mexico’s PROGRESA conditional cash transfer program on school enrollment and re-enrollment, with a particular focus on children who had dropped out prior to program implementation. Using PROGRESA’s experimental panel data, the author estimates a correlated random effects (CRE) probit model to distinguish enrollment responses between dropouts and non-dropouts, complemented by a cross-country dynamic panel analysis linking female dropout rates to economic growth. The paper’s central contribution lies in highlighting gendered and context-dependent effects of CCTs on re-enrollment decisions, with policy implications for reducing educational inequality.
The topic is important and policy-relevant, and the data source is strong. However, the manuscript would benefit from clearer focus, tighter framing, and consideration of alternative causal strategies that better exploit the program’s rollout structure.
Major Issues and Areas for Improvement
- The introduction devotes substantial space to the general relationship between human capital and economic growth, while the core focus of the paper—PROGRESA and school dropout/reenrollment—does not clearly emerge until around page 5. As written, the introduction reads more like a broad survey of growth theory than a targeted motivation for this specific study.
- The research question should be stated explicitly within the first 1–2 paragraphs.
- The introduction should be substantially shortened and reorganized, with background on human capital and growth streamlined and moved to later sections if needed.
- The current structure gives the impression that the paper’s true introduction begins several pages into the manuscript.
A more concise and sharply focused introduction would significantly improve readability and clarity.
- While the CRE probit model is carefully implemented, the paper does not sufficiently justify why a difference-in-differences (DID) or staggered adoption DID framework is not used, especially given PROGRESA’s phased rollout across communities. The program’s staggered implementation across locations provides a strong setup for modern DID methods. Recent advances (e.g., Callaway & Sant’Anna; Wooldridge’s DID framework) are now standard in program evaluation and could strengthen causal interpretation. Even if the authors ultimately prefer the CRE approach, the paper should either implement a staggered DID analysis as a robustness check or clearly explain why such methods are infeasible or inappropriate in this setting. Addressing this point is important, as many readers will expect DID-style evidence given the policy context.
- The cross-country growth analysis, while interesting, is not tightly integrated with the micro-level PROGRESA results. The macro analysis is correlational and relies on very different data and identification assumptions. The paper would benefit from a clearer explanation of how this extension should be interpreted and why it materially strengthens the contribution, rather than appearing as a separate exercise. Alternatively, the authors might consider shortening or reframing this section to avoid overstating the connection between the two analyses.
- A conceptual issue concerns the definition of the outcome variable. Although the paper is framed around dropout and dropout reduction, the probit model uses current enrollment as the dependent variable. As a result, the analysis estimates the effect of PROGRESA on enrollment (and reenrollment for prior dropouts), rather than directly modeling dropout behavior. This distinction matters for interpretation, as enrollment and dropout are not symmetric processes. The manuscript would benefit from either (i) reframing its contribution explicitly in terms of enrollment and reenrollment outcomes, (ii) adding an analysis that directly models dropout or exit decisions if feasible, or (iii) more clearly justifying enrollment as a proxy for dropout while acknowledging the associated limitations. Clarifying this point would improve conceptual coherence and strengthen the policy interpretation of the results.
- The finding that boys, particularly dropout boys in households with young children, do not benefit, and may even reduce enrollment, is intriguing but underexplained. This result deserves either deeper discussion, additional robustness checks, or a more cautious interpretation.
The paper addresses an important question and has several strong elements, particularly its focus on reenrollment and gender heterogeneity using high-quality data. However, substantive revisions are needed before the manuscript is suitable for publication.
With these revisions, the paper has the potential to make a meaningful contribution to the economics of education and policy evaluation literature.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThank you for your work in providing a more qualitative article.
Unlike previous studies focusing on average treatment effects, this paper distinguishes between enrollment (for those already in school) and reenrollment (for dropouts), providing a more nuanced understanding of human capital dynamics. By linking micro-level causal evidence from CCT programs to macro-level economic growth indicators, the study provides a compelling argument for targeted educational investments in girls.
The manuscript requires intensive proofreading. There are numerous instances of "track changes" or overlapping edits left in the final text (e.g., "Results show Results indicate" or "yield benefits extending beyond individual welfare"). These must be removed to meet professional publication standards.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsI appreciate the substantial revisions made in response to the previous review. The manuscript has improved meaningfully in terms of focus, framing, and conceptual alignment between the research question and the empirical analysis. In particular, the clarification of enrollment versus reenrollment outcomes, the reorganization of the introduction, and the reframing of the macro analysis as complementary are all appropriate and effective.
At this stage, I have no major remaining concerns regarding the empirical strategy or the data. The use of the correlated random effects (CRE) probit model is clearly motivated given the paper’s focus on reenrollment among prior dropouts, and the discussion of alternative approaches is sufficient for the scope and audience of this journal.
That said, I would encourage one final, minor clarification prior to publication. The paper would benefit from a more explicit and prominent statement of its core contribution. While the contribution can be inferred from the analysis, it is not articulated as clearly or succinctly as it could be. In particular, it would help readers if the authors emphasize—in one short paragraph, ideally in the Introduction or Conclusion—that the central contribution lies in shifting the focus from average enrollment effects to reenrollment decisions among children who had already dropped out, a margin not captured by standard enrollment analyses.
If prior studies of this program have examined average enrollment effects, briefly acknowledging this literature would further clarify how the present paper differs. If such analyses are limited or absent, noting this explicitly would also strengthen the positioning of the contribution. In either case, this paper should remain focused on reenrollment, as expanding the analysis is not necessary for its clarity or impact.
This is a matter of emphasis and positioning rather than substance. Addressing it would further improve the manuscript’s clarity and accessibility for a broad education and policy audience.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
