Teaching practices and students' learning: an experiment in Chile
Last registered on April 14, 2016


Trial Information
General Information
Teaching practices and students' learning: an experiment in Chile
Initial registration date
April 14, 2016
Last updated
April 14, 2016 3:20 PM EDT
Primary Investigator
Yale University
Other Primary Investigator(s)
PI Affiliation
Inter-American Development Bank
PI Affiliation
Yale University
Additional Trial Information
Start date
End date
Secondary IDs
We contribute to the policy debate about the effectiveness of different teaching methods. We use a randomized control trial to study the impact of a large scale educational program in Chile that provided technical and pedagogical support to teachers in order to help them improve the ways in which they teach the academic curricula and contents.
External Link(s)
Registration Citation
Bassi, Marina, Konstantinos Meghir and Ana Reynoso. 2016. "Teaching practices and students' learning: an experiment in Chile." AEA RCT Registry. April 14. https://doi.org/10.1257/rct.1177-2.0.
Former Citation
Bassi, Marina et al. 2016. "Teaching practices and students' learning: an experiment in Chile." AEA RCT Registry. April 14. http://www.socialscienceregistry.org/trials/1177/history/7676.
Experimental Details
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
- Test scores of students in standardized SIMCE tests (Math, Language, and Science) by gender and family income.
- Classroom Assessment Scoring System (CLASS) scores of teachers (principal component).
Primary Outcomes (explanation)
To produce the CLASS measures, trained coders watch and analyze videotaped classes and assign a score for teacher-students interactions in 11 dimensions. The outcome of interest is the first principal component of the 11 measures of student-teacher interactions.
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
Plan Apoyo Compatido experimental design.

In 2010, the Ministry of Education of Chile awarded the PAC program (Plan Apoyo Compartido) randomly among eligible schools throughout the country. Eligibility was defined in terms of baseline SIMCE test scores (eligible schools should have an average in the 2005-2009 SIMCE below the national average) and number of students (at least 20 students). Schools that met these criteria were ranked by their 2005-2009 average SIMCE scores in Language and Math and the bottom 1,000 schools were automatically considered eligible. Since participation in the program was voluntary, refusal to participate was expected, so in order to reach a target of around 1,000 eligible schools in the fi rst year of the program, the Ministry increased the sample within each local Departmetn of Education (DEPROVs) by 50%, going up in the the SIMCE ranking. The final sample of eligible schools (1480 schools) were offered participation in the program according to two methodologies. First, a non
experimental method was applied to 632 schools located in "small" DEPROVs (DEPROVs with 40 schools or
less). These schools were automatically invited to participate in the program and we exclude them from our
analysis. Second, an experimental method was applied to the other 848 schools located in "large" DEPROVs.
These schools were randomly assigned to being o ffered participation or not. All in all, the experimental sample
that we consider in this study includes 651 schools that were invited to participate (treatment schools) and 197
schools that were not (control schools).

The PAC was implemented in 2011, 2012, and 2013. This study uses administrative data from 2011 and 2012. This data set contains
student level information on treatment status, test scores, and baseline demographic characteristics.

CLASS experiment.

In 2012, among the entire PAC experimental sample, a subsample of 210 schools (105 from the PAC treamtent
group and 105 from the PAC control group) was randomly selected to also participate in the CLASS experiment.
137 invited schools agreed to participate in the filming sessions.
Experimental Design Details
Randomization Method
Randomization done in office by a computer
Randomization Unit
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
PAC: 848 schools

CLASS experiment: 210 schools
Sample size: planned number of observations
PAC: 31,384 students in 2011, 35,835 students in 2012.
Sample size (or number of clusters) by treatment arms
PAC: 651 schools in the treatment group and 197 schools in the control group.

CLASS: 105 from PAC treatment group and 105 from PAC control group.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB Name
IRB Approval Date
IRB Approval Number
Post Trial Information
Study Withdrawal
Is the intervention completed?
Is data collection complete?
Data Publication
Data Publication
Is public data available?
Program Files
Program Files
Reports and Papers
Preliminary Reports
Relevant Papers