Sacha Kapoor, Matthijs Oosterveen, and Dinand Webbink, "The Price of 
Forced Attendance", Journal of Applied Econometrics, Vol. 36, No. 2, 
2021, pp. 209-227.

Our paper uses administrative data from a public Dutch university. A
description of the dataset and its variables can be found in the Data
section of the paper. According to the General Data Protection
Regulation (GDPR), the data contains highly sensitive information on
students. 

Our goal is to make our findings replicable while respecting these
regulations. In consultation with the privacy officer of Erasmus
University Rotterdam, we have archived a stripped-down version of the
dataset on the online open access repository figshare. Access to this
striped-down version can be granted by submitting a research request
to the corresponding author: kapoor@ese.eur.nl.

The archived dataset is stripped from all control variables. Personal
identifiers are anonymized. The stripped-down dataset contains all
second-year student-course observations in our baseline sample and
includes the following essential variables:

* Student id
* Course id
* First-year GPA
* Second-year attendance
* Second-year grades

This dataset, together with the do file in this folder, allows for a
replication of our results. Not all of our results can be (exactly)
replicated, as, for example, the models cannot include demographic
control variables.

The file forced.do is an ASCII file in DOS format. It is zipped in the
file kow-files.zip. Unix/Linux users should use "unzip -a".