Resumo

EDWARDS JR, Brent. The Inability of Big Data to Escape the Limitations of Impact Evaluations.Traduzido porDulce Lomelí. RMIE [online]. 2019, vol.24, n.82, pp.871-878.  Epub 20-Fev-2020. ISSN 1405-6666.

This article analyzes the methods used to carry out “impact evaluations”, a type of research that supposedly can determine the effect of a policy or program on educational results. Thanks to the increased compilation of data on students and schools in the field of education, these methods have become ever more popular, in spite of their serious limitations. The study addresses the deficiencies of the two most common focuses of impact evaluations: regression analysis and random experiments (Randomized Controlled Trials, or RCT). The observation is that although large amounts of data and the methods of impact evaluation can be useful inputs for policy discussions, it is important to understand their limits.

Palavras-chave : educational research; quantitative method; data analysis; decision making; public policies.