2003年-世界发展银行全球_Supply-Side_School_Improvement_and_the_Learning_Achievement_of_the_Poorest_Children_in_Indigenous_and_Rural_Schools___The_Case_of_PARE_15页_295kb

Summary of the Document

Core Content

This document presents an analysis of the Programa para Abatir el Rezago Educativo (PARE), a supply-side school improvement program in Mexico, focusing on its impact on the learning achievement of the poorest children in indigenous and rural schools. It also discusses the program's effects on urban schools, highlighting the mixed outcomes observed.

The study examines whether targeted educational interventions can significantly improve learning outcomes for disadvantaged students, particularly in low-income regions. It challenges the earlier assumption that direct educational interventions have limited effects on reducing learning inequality due to socioeconomic factors, by showing that well-designed and implemented programs can lead to substantial improvements.

The paper outlines the structure of the PARE program, including its implementation period (1992–1997), objectives, and components. It highlights the focus on improving the availability and quality of educational services and strengthening institutional capacity in the four states with the highest poverty rates.

Main Points

• PARE Objectives:

  • Improve the quality and efficiency of primary education in poor, rural, and indigenous schools.
  • Reduce repetition and dropout rates.
  • Raise cognitive achievement levels.
  • Strengthen management and implementation of the education system.

• Program Structure:

  • The project included two main components:
    1. Educational Services Improvements (72% of cost): Focused on providing books, didactic materials, teacher training, infrastructure, and distance education technologies.
    2. Institutional Strengthening (28% of cost): Aimed at improving the efficiency of the education sector through monitoring and evaluation systems.

• Implementation Context:

  • The program was implemented in four states (Oaxaca, Guerrero, Chiapas, Hidalgo), which had high poverty rates and low educational indicators.
  • In these states, the majority of public schools were under federal control, with some states also having a coexisting state system.
  • The program was specifically targeted towards rural and indigenous areas due to the severity of poverty and lack of resources.

• Database and Methodology:

  • The study uses data collected from both experimental and control groups.
  • Standardized tests in Spanish and mathematics were administered to students in fourth, fifth, and sixth grades.
  • A two-stage least squares regression model was used to estimate the relationship between learning achievement and the monetary value of PARE assistance.

• Key Findings:

  • Indigenous Schools: PARE significantly improved the learning achievement of less poor students, with a 42.3% increase in Spanish test scores. The poorest students saw a smaller improvement.
  • Rural Schools: PARE improved learning outcomes for both the poorest and less poor students, with the poorest benefiting slightly less.
  • Urban Schools: PARE had a negative effect on learning achievement, with a larger negative impact on the poorest students.
  • The learning achievement elasticity with respect to the cost of intervention was found to be 0.33, indicating that the program's impact was not as strong as previously thought due to implementation issues.
  • Staff quality (school directors, teachers, and supervisors) was a significant determinant of student learning achievement.
  • Parental participation had a negative correlation, possibly due to simultaneity bias, where parents become more involved when children are struggling.

• Hypothesis and Model:

  • The main hypothesis is that targeting school improvement programs on poor areas does not necessarily lead to significant improvements for the poorest children.
  • The model used in the study is:
    $$
    \begin{array}{l} E S P A N O L 6 (i) = a + b X (i) + c P A R E (i) * P O B (i) + d P A R E (i) (1 - P O B (i)) \ + e E S P A N O L 4 (i) + U (i) \ \end{array}
    $$
    Where:
    • $ PARE(i) = 1 $ if the student is in an experimental school, 0 otherwise.
    • $ POB(i) = 1 $ if the student is from the poorest socioeconomic group, 0 otherwise.
    • The hypothesis is that $ c < d $, meaning that the poorest children benefit less from PARE than less poor children.

Key Information

• The PARE program was implemented in 1992–1997 in four Mexican states with high poverty rates.
• The program aimed to improve primary education quality and reduce learning inequality.
• Indigenous and rural students showed positive learning effects, but urban students experienced a negative impact.
• The cost-effectiveness of PARE was noted, particularly for indigenous and rural students.
• Implementation issues (such as budget delays and poor delivery) may have affected the program's effectiveness, especially in urban areas.
• The elasticity of learning achievement with respect to PARE intervention was found to be 0.33, suggesting that improvements are modest relative to costs.
• Staff quality and socioeconomic background were found to be important factors influencing student learning outcomes.

Conclusion

• Supply-side interventions like PARE can have substantial effects on the learning achievement of children in indigenous and rural schools in poor areas.
• However, greater attention is needed for the poorest children within these groups.
• The urban sample suggests that PARE may have had negative effects, reinforcing earlier findings about the limited effectiveness of the program in urban settings.
• The negative correlation in urban schools remains a puzzle, possibly due to disruptive implementation or other unaccounted factors.
• Further research is needed to fully understand the implications of these findings and improve the effectiveness of similar programs.