Angelov, N.; Johansson, P. Lessons Learned from Applying Computer Assisted Experimental Designs. Preprints2024, 2024100575. https://doi.org/10.20944/preprints202410.0575.v1
APA Style
Angelov, N., & Johansson, P. (2024). Lessons Learned from Applying Computer Assisted Experimental Designs. Preprints. https://doi.org/10.20944/preprints202410.0575.v1
Chicago/Turabian Style
Angelov, N. and Per Johansson. 2024 "Lessons Learned from Applying Computer Assisted Experimental Designs" Preprints. https://doi.org/10.20944/preprints202410.0575.v1
Abstract
This paper discusses our experience with a computer assisted design for the analysis of an intervention on a digital platform, conducted by the Swedish Tax Agency. The interest was in analyzing the possibilities of estimating the effect of a message (the intervention) on the subsequent tax compliance of taxpayers who had foreign income. The effect size was not expected to be large. However, as the cost of the intervention is very low, even a small effect could be important economically. Due to administrative reasons the number of treated was limited to 500 individuals. Because of the limited sample size and the small expected effect it was imperative to consider an efficient experimental design. The population considered for the experiment consisted of 2,697 individuals. Instead of choosing all except the treated as controls, only 500 were deliberately chosen to examine the efficiency gains of a rerandomization design with real world data. To this end, a simulation study was conducted on the same population as was randomly sampled to be part of the experiment. An important conclusion from this exercise is that regression-based post-stratification on relevant covariates is an equally efficient alternative as rerandomization in this setting. The results are of special interest when experiments are run in organizations where computer assisted designs may be costly.
Keywords
field experiment; tax compliance; rerandomization; foreign income; automatically exchanged financial information
Subject
Business, Economics and Management, Accounting and Taxation
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.