Field experiments on employment discrimination test for differential treatment in real-life social settings. In these experiments, job applications of fictitious candidates are sent to real vacancies. The applications vary on treatment conditions only (e.g. ethnicity, phenotype, religious affiliation, gender, etc). Differences in call-back rates across identical applicants varying in the treatment condition provide clear and direct evidence of discrimination.  Implementing field-experiments is a complex and costly endevour.

 

Working in close collaboration with the GEMM project, the D-Lab has developed a unique software, we call  the goat. The goat is and advanced scraping and implementation engine that allows us to identify suitable online vacancies, randomize the treatments, and automatize key steps of the application process, thus greatly reducing implementation costs. Our software builds on a randomization code originally developed by our colleagues at WZB and later updated by the ICT programmers at University of Utrecht. To this code, we added new scraping and implementation tools adapted for online field-experimentation. The goat provides the software infrastructure of the D-Lab for field-experimental research in the Spanish labour market. To protect our software from the risk of cyberattacks, we apply the strictest security protocols.