Time period:

Target group:

Number of participants:

Around 50.000 children for each grade level and test.     

Intervention:

In Denmark, students in public schools are mandated national testing in reading in the 2^nd grade and in math in the 3rd grade. A key feature of the initiative is that schools report the results of each test to parents together with a description of the test content. The letter reports the child’s test performance as one of the following: Considerably below average, Below average, Average, Above average, or Considerably above average. We evaluate the impact of this information provided to parents.     

Research:

We use a regression discontinuity design to estimate the impact of the content of the performance feedback.

We have access to detailed student-level national test data from the period 2010-2014. The data includes the performance labels and the underlying finer-grained and numeric test score distribution. As parents do not observe their children’s underlying numeric test score—only the performance label—the local variation in test results gives rise to a sharp contrast in performance feedback to parents. This allows us to use a regression-discontinuity design to study the effect of how these labels affects later student well-being and performance.

Specifically, we compare children whose numeric test result is just below or just above the threshold value that separates the performance labels considerably below average and below average. We do the same around the other thresholds allowing us to study whether the impact differs for academically disadvantaged and advantageous students.

We use the national wellbeing-survey and schools’ absence reports to evaluate the impact on students’ well-being and behavior.    

Results:

Expected Spring 2017.