Differentiated effects of community schooling on cognitive and social-emotional learning outcomes

Abstract

Community schools are becoming increasingly popular. They aim to enhance children’s educational performance by offering extended educational and social services. As community schools mainly focus on disadvantaged children, this study evaluates the effects of community schooling on the educational outcomes of these pupils. We focus on care pupils and pupils whose parents have a low educational attainment. We hypothesize that community schools are particularly beneficial to care pupils, namely, pupils with additional educational needs, and for pupils with parents with lower levels of educational attainment. Our analyses show that both pupil groups benefit in terms of reduced underachievement. The duration of community school attendance as such does not affect cognitive outcomes, but it proves beneficial to the educational achievement of both groups when the community school subsidy is taken into account as well. We show that different community school activities have partly contrary effects on care pupils and non-care pupils.

Keywords:

• community school effectiveness
• educational progress
• cognitive outcomes
• social-emotional outcomes
• disadvantaged children

Notes

1. Dutch: Brede scholen.

2. In England, community schools denote local government control over the school’s student admissions policy, staff employment, and ownership of the grounds and buildings (UK Government, 2014).

3. Risbo is a Dutch Research Training Consultancy, affiliated with the University of Rotterdam.

4. Dutch: CohortOnderzoek OnderwijsLoopbanen van 5 tot 18 jaar [Cohort Research Educational Careers, 5 to 18 years].

5. A descriptive table on the comparability between community and regular school students can be obtained from the researchers. t tests on all variables confirm that, after the matching, no significant differences between the community school pupils (CS) and the regular school pupils (RS) in terms of observable characteristics remain. None of the differences passes the threshold of 95% of significance. As the samples consist of groups of over 300 observations, we are confident that the lack of statistical significance is not likely to derive from power issues (Type II errors), because with this amount of observations differences of the order of less than 20% of a standard deviation would be detected at a conventional level of power (.80).

6. For non-care pupils in community schools, it is 1.741 (SD 2.534).

7. For community schools pupils with parents with higher levels of educational attainment, it is 1.963 (SD 2.670).

8. For non-care pupils in community schools, it is 2.978 (SD 3.423).

9. For community school pupils with relatively higher levels of educational attainment, it is 3.075 (SD 3.420).

10. For non-care pupils in community schools, it is €120.43 (SD 50.10).

11. For community school pupils with relatively higher levels of educational attainment, it is €122.46 (SD 47.19).

12. For non-care pupils in community schools, it is €154.66 (SD 54.87).

13. For community school pupils with levels of educational attainment, it is €148.039 (SD 59.84).

14. A specification with a dummy indicator confirms the results in Table 10. In a split sample analysis, the community school duration indicator proves not to be significant for any of the outcomes.

15. A specification with a dummy indicator confirms the results in Table 10. In a split sample analysis, an increased number of years in a community school significantly reduces underachievement.

16. The results of estimating a dummy specification confirm the findings; the only exception is that the interaction is negative and significant for study skills. The split sample analysis confirms the results in Table 10.

17. The dummy specification confirms the findings. The results from the split sample analysis show a negative effect on underachievement.

18. A split sample analysis generally supports the findings in Table 11: A negative effect throughout of subsidy on the outcomes, while community school duration is positively significant for mathematics, world orientation, and study skills.

19. A split sample analysis broadly confirms the results in Table 11: Longer exposure to community school education relates to reduced underachievement. Differing from the results in Table 11, subsidy itself is not significant for underachievement but positively significant for perseverance. The results also suggest that community school exposure duration negatively affects perseverance.

20. These findings are partly confirmed by the split sample analysis: a negative effect of subsidy throughout, while the effect of community school attendance duration is throughout positively significant.

21. A split sample analysis confirms the findings in Table 11.

Additional information

Funding

This work was supported by Platform31.

Notes on contributors

Marieke Heers

Marieke Heers is affiliated with the Top Institute for Evidence Based Education Research (TIER) at Maastricht University, The Netherlands. She holds a German-French BA (European Business Studies) from the University of Paderborn and the University of Le Mans and a MSc (Business Administration, Strategic Human Resource Management, cum laude) degree from Radboud University Nijmegen. After graduating in 2009, Marieke joined the Top Institute for Evidence Based Education Research at Maastricht University as a PhD candidate. During her doctoral studies, she visited the Teachers College at Columbia University in New York City. She currently works as a postdoctoral researcher at the Institut National d’Etudes Démographiques in Paris.

Joris Ghysels

Joris Ghysels is an Assistant Professor at TIER. He studied Economics and Sociology at the University of Antwerp. He worked in Labour and Social Security research at the Herman Deleeck Centre for Social Policy (CSB) (Antwerp) and the Nitlapán Institute (Managua). In 2003, he obtained his PhD in Economics at the University of Antwerp with a dissertation entitled The Work and Care Game: an Empirical Analysis of Household Decision Making in the European Union. Between 2003 and 2011, he was coordinator of the research team on family and care at the CSB. He has been teaching sociology, statistics, and labor economics to regular students and in adult education programs. His current research interests relate to the field of educational economics, program evaluation and experimental methods, public economics, and models of human behavior.

Wim Groot

Wim Groot is full professor for Evidence Based Education and full professor for Health Economics, both at Maastricht University. He was a visiting fellow at Stanford University, University of California at Berkely and Cornell University in the United States, Monash University in Australia, and the European University Institute in Italy. In 2004, he received an honory doctorate degree of the National University of “Kyiv-Mohyla Academy” in Kiev, Ukraine.

Henriëtte Maassen van den Brink

Henriëtte Maassen van den Brink is full professor in Economics (Education and Labour economics) at the faculty of Economics and Econometrics at the University of Amsterdam. She is also full professor in Evidence Based Education at the faculty of Humanities and Sciences at Maastricht University. She was visiting professor at Stanford University and Cornell University in the United States and the European University Institute in Italy.