This is very impressive when compared to effects observed in psychology (figure, full; if it’s helpful, this effect size is similar to the gain in height American girls experience between 14 and 18 years of age).
Would you be able to share where you got this analogy from? I often read research that deals with effect sizes like standardised mean difference, hedges G, Cohen’s d and so on, but I don’t have a very good intuition for what those effect sizes mean. I’d love to see more examples like this that helps to build my intuition.
If you are trying to train yourself to have an intuition on the sizes, you can use visualization tools, and there are two main ones at present:
this visualization tool for group differences. The tool is based on Cohen’s d, and since Hedges’ g is just Cohen’s d with a sample size correction, the tool will help you grasp to what extent two groups overlap/differ.
this visualization tool for association between two variables. This tool is based on Pearson’s r correlations, but you can conceptually use it to develop intuitions for other tools measuring association (like Spearman rho, R2 = R-squared, or eta-squared—let me know if you’d like me to elaborate on this!)
If you want someone else to have an intuitive grasp of the size of an effect you’re discussing, you can refer to lists of benchmarks. My height example comes from Cohen (1988′s Statistical power analysis for the behavioral sciences), but more modern examples can be found here and here.
Would you be able to share where you got this analogy from? I often read research that deals with effect sizes like standardised mean difference, hedges G, Cohen’s d and so on, but I don’t have a very good intuition for what those effect sizes mean. I’d love to see more examples like this that helps to build my intuition.
If you are trying to train yourself to have an intuition on the sizes, you can use visualization tools, and there are two main ones at present:
this visualization tool for group differences. The tool is based on Cohen’s d, and since Hedges’ g is just Cohen’s d with a sample size correction, the tool will help you grasp to what extent two groups overlap/differ.
this visualization tool for association between two variables. This tool is based on Pearson’s r correlations, but you can conceptually use it to develop intuitions for other tools measuring association (like Spearman rho, R2 = R-squared, or eta-squared—let me know if you’d like me to elaborate on this!)
If you want someone else to have an intuitive grasp of the size of an effect you’re discussing, you can refer to lists of benchmarks. My height example comes from Cohen (1988′s Statistical power analysis for the behavioral sciences), but more modern examples can be found here and here.