Evaluation of the Danish School Lunch Program

How to cite

Andersen, Simon Calmar; Nielsen, Helena Skyt. 2026. "Evaluation of the Danish School Lunch Program." AEA RCT Registry. AEARCTR-0018291. https://doi.org/10.1257/rct.18291-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Simon Calmar Andersen* , Helena Skyt Nielsen

* Primary investigator

Registry ID
AEARCTR-0018291
DOI
https://doi.org/10.1257/rct.18291-1.0
First registered
Last updated
Status
In development
Trial dates
2026-01-01 – 2029-12-31
Countries
Private
JEL codes
I18; I24; 38
Registry record
http://www.socialscienceregistry.org/trials/18291

Abstract

This pre-analysis plan concerns the evaluation of the Danish School Lunch Program (DSLP). The program is rolled out as a Randomized Controlled Trial (RCT) in the period 2025-2028 in around 130 public and around 50 private independent schools in Denmark with the purpose of improving health, well-being, and academic performance.

Intervention

Schools offer either free lunch (around 70% of schools), lunch offered at a price of 15 DKK per day (around 15%) and lunch offered at a price of 25 DKK per day (around 15%).

Experimental design

We treat each of the large schools that signed up two or three grade level tiers as a stratum. The grade level tiers are grades 0-3, grades 4-6, and grades 7-9. We randomly draw one of the tiers that the school had signed up as a treatment tier, and the other one or two tiers as control tier(s). In our main analyses, we disregard small schools and large schools signed up with only one tier.

Design details

Randomization method
Randomization done by computer (Stata with set seed).
Randomization unit
The clusters are grade level tiers. Strata are schools.
Clusters
130 class level tiers in treatment group, about 200 in control group, i.e., about 330 in total.
Observations
About 26,000 treated students and about 39,000 students in control group in the first year.
Arms
130 class level tiers are treated, about 200 are in control group.
Primary outcomes
Number of days absent
Secondary outcomes
See pre-analysis plan (PAP).

Cited by

No citation data available for this registration yet.

Keywords