The effects of adding social-emotional learning to a comprehensive education intervention in El Salvador on teacher well-being: a mixed methods evaluation

ABSTRACT

This article employs a mixed-method approach to assess the effects of El Salvadorian Integrated Systems of Full-Time Inclusive Schools (SI-EITP), which offers in-service teacher professional development (TPD) combined with a socioemotional learning intervention, on teacher well-being. Findings from the cluster-randomized controlled trial with no baseline show null results for most teacher well-being outcomes analyzed. Interviews suggest limited effects on teachers’ mindfulness and emotion regulation and indicate that teachers’ participation in TPD was potentially increasing their stress levels and emotion exhaustion. We recommend that TPD program designs consider how the delivery mode and intensity may influence teachers’ stress.

KEYWORDS:

• Teacher professional development
• teacher well-being
• mindfulness
• stress
• emotion regulation

Acknowledgments

We thank Annette Brown, Carina Omoeva and Wael Mousse for providing comments and feedback and Tanya Smith-Sreen for valuable analytical support.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Ahuachapán, Sonsonate, La Libertad, San Salvador, La Paz, Usulután, San Miguel, La Unión

2 A previous study by Wolf et al. (2015a) has examined the impacts of a teacher professional development intervention in the Democratic Republic of the Congo, on teacher professional well-being. However, the intervention assessed did not include a targeted and explicit component that aimed to improve teacher well-being.

3 See Appendix 1, Table A1 for details.

4 Given that this study assesses program effect on ten different outcomes of interest, we correct for multiple hypothesis testing using the Bonferroni adjustment method. The method adjusts the p values to reflect the multiple-inference problem by controlling for familywise error rate (FWER), which is the probability that at least one of the J true hypotheses in the family is rejected (Anderson, 2008; Fink et al., 2014).

5 Descriptive statistics for the treatment and control groups at endline are shown in Table A2 in Appendix 1, and descriptive statistics at baseline on key outcomes of interest for the treatment group only (as baseline data was not collected from the control group) are shown in Table A3 in Appendix 1.

6 Figure A2 in Appendix 3 shows distributions for key outcomes for the treatment and control groups at endline. Overall, the distribution of the two groups is very similar, with a few exceptions.

7 All surveys were anonymized and issues with linking teachers’ unique ID between baseline and endline prevented us from including the full sample in the analysis. Even though we only identify 169 secondary education teachers with the same unique ID for baseline and endline surveys, program monitoring data shows 322 secondary education teachers who participated on the first and the last workshop (290 with non-missing values for all outcomes).

8 Formally, we estimate the following model: Y_it = ß₀ + ß₁Endline + θ_i + µ, where Y_it denotes the outcomes of interest for teacher i at time t, ß₀ is the constant, Endline equals zero if t if the baseline pre-treatment period and one if t is the endline post-treatment period; θ represents the teacher-specific time invariant effect, which considers all characteristics of each teacher in the sample that are time invariant, both observable and unobservable, and µ is the idiosyncratic error term clustered at the school level.

Additional information

Funding

This work was supported by the Ward Cates Emerging Scientific Leader Award.

Notes on contributors

Fernanda Soares

Fernanda Soares holds a PhD in Policy Analysis from the Maastricht School of Business and Economics and is a Research Technical Advisor at the University of Notre Dame’s Global Center for the Development of the Whole Child. Her research interests comprise education economics, professional learning communities, social-emotional learning, teachers’ well-being and teachers’ professional development.

Nina Cunha

Nina Cunha holds a PhD in Economics of Education from Stanford University and is a Senior Research Associate at the Research and Evaluation Unit within Family Health International 360. Her research interests are in the economics of education, quantitative methods, and causal inference, with a particular interest in teacher quality and parental engagement.