Cohen's small, medium, and large effect size benchmarks are arbitrary and should not be interpreted rigidly¶

Claim · All claims
Evidence · 1 study · theoretical r3 · q1
Currency · newest study 2013 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q1 i? The conventional d = 0.2, 0.5, 0.8 benchmarks should only be used for extremely novel findings; small effect sizes can have large practical consequences, and effect sizes are best interpreted against related effects in the literature. → Lakens 2013

Evidence¶

Lakens 2013¶

Lakens, D. (2013). Calculating and reporting effect sizes to facilitate cumulative science: a practical primer for t-tests and ANOVAs. Frontiers in Psychology. https://doi.org/10.3389/fpsyg.2013.00863

q1 · i? · theoretical · r3

Interpretive argument in the section on interpreting Cohen's d, citing Thompson (2007) and Cohen (1988). The article gives no empirical test of benchmark use; it states small effects can have large consequences, citing a suicide-rate reduction at d = 0.1, and notes there are no clear recommendations for interpretation (Fidler, 2002).

"However, these values are arbitrary and should not be interpreted rigidly ( Thompson, 2007 ). Small effect sizes can have large consequences, such as an intervention that leads to a reliable reduction in suicide rates with an effect size of d = 0.1."

Discussion¶

Related Claims¶

• Statistically significant differences in approaches to learning can constitute small effects of little theoretical or practical importance — related
• Writing To Learn Improves Understanding — related
• The test benefit of interleaved practice does not diminish over time and perhaps grows larger (d = 0.79 at 30 days) — related