Field
Trial Status
|
Before
on_going
|
After
completed
|
Field
Last Published
|
Before
January 23, 2019 09:17 AM
|
After
April 24, 2019 04:02 AM
|
Field
Primary Outcomes (End Points)
|
Before
We measure financial literacy of students by a test that contains 13 questions. This test consists of questions that measures both financial knowledge and financial behavior. We measure the financial literacy of teachers by a separate questionnaire. We are able to measure the intensity of treatment because we observe the activity of the teachers in the online training module (for example the number of clicks on each item).
|
After
We measure financial literacy of students by a test that contains 9 questions. This test consists of questions that measures both financial knowledge and financial behavior. We measure the financial literacy of teachers by a separate questionnaire. We are able to measure the intensity of treatment because we observe the activity of the teachers in the online training module (for example the number of clicks on each item).
|
Field
Sample size (or number of clusters) by treatment arms
|
Before
Wave 1: second year of high school
Control group = 739 students, 13 schools
Treatment group 1 = 408 students, 12 schools
Treatment group 2a = 481 students, 10 schools
Wave 2: third year of high school
Control group = 111 students, 5 schools
Treatment group 1 = 91 students, 4 schools
Treatment group 2b = 119 students, 4 schools
Total sample
Control group = 850 students, 18 schools
Treatment group 1 = 499 students, 16 schools
Treatment group 2 = 600 students, 14 schools
Average number of schools per condition = 16
Average number of students per condition = 649.7
Average number of students per school = 40.6
|
After
Wave 1: second year of high school
Control group = 739 students, 13 schools
Treatment group 1 = 408 students, 12 schools
Treatment group 2a = 481 students, 10 schools
Wave 2: third year of high school
Control group = 111 students, 5 schools
Treatment group 1 = 91 students, 4 schools
Treatment group 2b = 119 students, 4 schools
Total sample
Control group = 850 students, 18 schools
Treatment group 1 = 499 students, 16 schools
Treatment group 2 = 600 students, 14 schools
Average number of schools per condition = 16
Average number of students per condition = 649.7
Average number of students per school = 40.6
Please note that as teacher characteristics were included as control variables in the final analyses, this resulted in missing values for a selection of students. Consequently, the final sample reported in the article is somewhat smaller:
Wave 1: second year of high school
Control group = 675 students, 13 schools
Treatment group 1 = 401 students, 11 schools
Treatment group 2a = 481 students, 10 schools
Wave 2: third year of high school
Control group = 111 students, 5 schools
Treatment group 1 = 91 students, 4 schools
Treatment group 2b = 86 students, 3 schools
Total sample
Control group = 786 students, 18 schools
Treatment group 1 = 492 students, 15 schools
Treatment group 2 = 567 students, 13 schools
|
Field
Power calculation: Minimum Detectable Effect Size for Main Outcomes
|
Before
We base our computation on List et al. (2011) and account for intracluster correlation in the calculation of the minimal detectable effect size. In our experimental setting, there on average 16 schools in each experimental condition. Each school on average contains 40.6 students. We computed an intracluster correlation of 0.12. In the analysis, we can control for characteristics of schools and students, which would decrease the intracluster correlation. With the conventional power of 0.8 and a significance level of 0.05, we are able to detect a treatment effect of 0.37 standard deviations or larger.
Details of calculation:
To calculate the minimum detectable effect size, we follow List et al. (2011). They show that in a clustered design, the minimum number of observations in each experimental group can be computed as follows:
n=2(t_(α/2)+t_β)²(σ/δ)²(1+(m-1)ρ)
This implies that the minimum detectable effect size is equal to:
δ=σ/√(n/(2(t_(α/2)+t_β)²(1+(m-1)ρ)))
Or the minimum detectable effect size expressed as a fraction of a standard deviation is equal to:
δ/σ=1/√(n/(2(t_(α/2)+t_β)²(1+(m-1)ρ)))
δ/σ=1/√(649.7/(2(1.96+0.84)²(1+(40.6-1)0.12)))=0.37
Reference
List, J., Sadoff, S. and Wagner, M. (2011), So you want to run an experiment, now what? Some simple rules of thumb for optimal experimental design, Experimental Economics 14, 439-457
|
After
We base our computation on List et al. (2011) and account for intracluster correlation in the calculation of the minimal detectable effect size. Please note that this computation was based on the sample sizes before the inclusion of teacher characteristics resulting in missing values. In our experimental setting, there on average 16 schools in each experimental condition. Each school on average contains 40.6 students. We computed an intracluster correlation of 0.12. In the analysis, we can control for characteristics of schools and students, which would decrease the intracluster correlation. With the conventional power of 0.8 and a significance level of 0.05, we are able to detect a treatment effect of 0.37 standard deviations or larger.
Details of calculation:
To calculate the minimum detectable effect size, we follow List et al. (2011). They show that in a clustered design, the minimum number of observations in each experimental group can be computed as follows:
n=2(t_(α/2)+t_β)²(σ/δ)²(1+(m-1)ρ)
This implies that the minimum detectable effect size is equal to:
δ=σ/√(n/(2(t_(α/2)+t_β)²(1+(m-1)ρ)))
Or the minimum detectable effect size expressed as a fraction of a standard deviation is equal to:
δ/σ=1/√(n/(2(t_(α/2)+t_β)²(1+(m-1)ρ)))
δ/σ=1/√(649.7/(2(1.96+0.84)²(1+(40.6-1)0.12)))=0.37
Reference
List, J., Sadoff, S. and Wagner, M. (2011), So you want to run an experiment, now what? Some simple rules of thumb for optimal experimental design, Experimental Economics 14, 439-457
|
Field
Pi as first author
|
Before
No
|
After
Yes
|