Context matters: Interpreting effect sizes in education meaningfully

Abstract

Assessing group differences through effect size estimations has become standard practice in educational research. However, the interpretation of these effect sizes is often discussed without considering contextual factors such as prior knowledge, motivation, and socio-economic background that influence the learning outcome. In this methods paper, we discuss the significance of considering affective, cognitive, and sociographic factors when interpreting effect sizes. Furthermore, we propose a theoretical framework for two primary types of effect size distribution based on prior knowledge. This framework may enable researchers in the learning sciences to interpret and contextualize their findings beyond testing null hypotheses and reporting abstracted effect sizes. - Contextual factors, including prior knowledge or motivation, should be taken into account when interpreting effect sizes. - In relationship to prior knowledge, effect sizes often follow a normal distribution or an extension thereof following one period of a sine-like wave.

Keywords

• Contextual Effect Size Interpretation