The main objective of this study is to investigate whether the free fruit policy resulted in overallbetter academic performance on the 5th and 8th grade national tests in Math, Reading and English. In addition, we will investigate whether the impact of the intervention could be moderated by sex,sosioeconomic status, country of origin (western vs. non-western) and centrality of the school. The methods and analyses are outlined below.
METHOD
Participants and data
The data will be based on all tests in 5th and 8th grade registered by the Directorate of Education inthe years 2007 to 2018 (possibly 2019 if data is available), within cohorts of children born in 1994-2008. The pupils were 9 to 10 years old at the time of the 5th grade tests, and 12 to 13 years old at the time of the 8th grade test. For the cohorts born 1997 to 2005 results for both the 5th and 8th grade tests are available.
Measures and variables
Outcome: The outcome measure is the scores on the National tests in Math, Reading and English. We assume that the data includes a continuous score for all tests in all years. If scores are standardized in the data of some of the years, we will apply the same level of standardization for all years.
School type: Correspond to treatment (combined first and secondary elementary school) vs. control (pure first elementary school). For analyses of the 5th grade test we will assume that to be the school of participation from grade 1 to grade 5. For analyses of the 8th grade test, we will use the grade 7 school, and assume that to be the school of participation at grade 1 to 7. In cases where the school changes status from combined to separate or vice versa during the period, the data will be omitted from analyses.
