README
======

Dataset title:
Data for: Performance Incentives in Education: The Role of Goal Mismatch, https://doi.org/10.5878/yveq-q046

Recommended Citation:  
Campos-Mercade, P., Thiemann, P., & Wengström, E. (2026). Data for: Performance Incentives in Education: The Role of Goal Mismatch (Version 1) [Data set]. Lund University. https://doi.org/10.5878/yveq-q046

See the Researchdata.se catalogue entry for this dataset - accessible via the DOI link above - for complete metadata, including the dataset description, contact information and information about how to obtain the dataset. 

Responsible organization:
Lund University

Description: This data and the accompanying replication files are meant to reproduce all main results in the paper "Performance Incentives in Education: The Role of Goal Mismatch" (accepted for publication in the Journal of Human Resources). The original data combines survey data and administrative student records.

The input dataset, goals_JHR.dta, is a de-identified and reduced version of the original data, designed to ensure replicability while protecting anonymity. It includes a subset of the variables used in the original paper: raw outcomes, treatment variables, and outcomes residualized (demeaned) with respect to the control variables used in the paper. It also includes the variables used for the heterogeneity analyses.

Note: The replication dataset does not exactly reproduce the results reported in Tables 3-6 of the paper. This is because the replication package provides residualized outcomes, whereas the paper uses the underlying control variables directly. As a result, small deviations in the results occur, for instance, when collinear controls were dropped from the original regressions or when heterogeneity analyses are performed.
	
Access restrictions: This dataset contains pseudonymized data and is subject to legal protections. This dataset can be requested from the Researchdata.se research data catalog using the link above.

Essential information for dataset reuse.
----------------------------------------

This dataset is meant for replication purposes only. Details about data collection are provided in the original paper.

To run the code and obtain replication results, complete the following steps:
1) At the top of each dofile, set the directory path to the location where goals_JHR.dta is stored. Figures and tables will be stored in the same folder.
2) Run each dofile individually in Stata.
3) The dofiles produce tables as .tex files. To view the formatted output, compile the .tex files in a LaTeX editor, or alternatively, view each file in a text editor.

Dataset organization
--------------------
README.txt
manuscript.pdf
goals_JHR.dta
replication_script.do
replication_results.pdf

Datafile descriptions
----------------------
README.txt - readme file
manuscript.pdf - original paper
replication_script.do - dofile that reproduces Figure 3 and Tables 3-6 in the original paper
replication_results.pdf - results of the replication exercise for Tables 3-6 using goals_JHR.dta and the dofiles above

goals_JHR.dta - replication data set
Columns
	macropoints [0-100]. Macropoints, raw scale.
	sd_macropoints [standardized, mean 0, SD 1]. Macropoints.
	std_uppgift [standardized, mean 0, SD 1]. Homework points.
	std_examgrade [standardized, mean 0, SD 1]. Exam points.
	avg_target [standardized, mean 0, SD 1]. Target grade.
	sd_macropoints_resid [residual]. Residual from OLS regression of sd_macropoints on all control variables (see manuscript.pdf).
	std_uppgift_resid [residual]. Residual from OLS regression of std_uppgift on all control variables (see manuscript.pdf).
	std_examgrade_resid [residual]. Residual from OLS regression of std_examgrade on all control variables (see manuscript.pdf).
	avg_target_resid [residual]. Residual from OLS regression of avg_target on all control variables (see manuscript.pdf).
	incentives [0/1]. Student was assigned to any of the three treatment groups.
	treat_exo_high [0/1]. Student was assigned to the "exogenous-high" treatment group.
	treat_exo_low [0/1]. Student was assigned to the "exogenous-low" treatment group.
	treat_endo [0/1]. Student was assigned to the "endogenous" treatment group.
	control [0/1]. Student was assigned to the control group.
	treat_match [0/1]. Student's assigned goal matches the chosen goal.
	treat_nomatch [0/1]. Student's assigned goal does not match the chosen goal.
	treat_match_ability [0/1]. Student's assigned goal matches their ability.
	treat_nomatch_ability [0/1]. Student's assigned goal does not match their ability.
	female [0/1]. Female.
	base_high [0/1]. High baseline grade.
	choose_a [0/1]. Student chose goal A in the survey.

Reference information
---------------------
Source code information:
All analyses were conducted in Stata (version 19.5).

Data Quality:
The replication dataset does not exactly reproduce the results reported in the paper. It does not contain all the variables used in the original paper and is not meant to be used for research purposes, only for replication.

