Otras formas de entender la d de Cohen

Abstract

Cohen’s d (d) is quite a used measure of the size of the effect and its report is compulsory necessary in sta-tistical analyzes. Nevertheless, researchers report that the difference between two distributions is small (d > .20). However, the interpretation of this coefficient is not clear in psychology studies. In this sense, it is necessary to con-vert the d into a probability measure to facilitate the inter-pretation of the distributions that are object of comparison. Among the most frequent measures are: Cohen’s U3, the superposition coefficient (OVL), the probability of superi-ority (PS) and the number needed to treat (NNT), which can be considered as alternative measures of the magnitude of a difference. For such purposes, R codes that can be easi-ly used by the researchers are provided, as well as a table showing the modifications of the alternative measures be-fore the increase in the size of the effect.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 26 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.