Announcement

Collapse
No announcement yet.

Effect size dependent sample

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Effect size dependent sample

    Hi, I am trying to calculate an effect size for a dependent sample. I have read around tonnes of research and there does not appear to be a standardised approach, in fact a lot of people are using the mean difference/SD of the difference, which is certainly incorrect. Could anyone please provide me with an equation and evidence of this to compare the effect size between a dependent sample please? Any help would be greatly appreciated! Thank you in advance.

  • #2
    Re: Effect size dependent sample

    Hi Lucy,

    One thing to consider before you measure effect size is:
    1. Same sample size and similar SD -> Cohen's d
    2. Same sample size and different SD -> Glass \Delta
    3. Different sample size -> Hedge's g (not your case, here, since you have repeated measures).

    When calculating effect size, what you're really doing is just comparing "how much" two sample means are different from each other. Like all statistical tests, it is a measure of variance. Some just use Cohen's D for independent samples applying it to dependent samples. Even though the test was not originally designed for comparing dependent samples, it probably still works better than when Cohen's D is used with different sample sizes and largely different SD. With that in mind, personally, I prefer to use Pearson's r in order to avoid those issues. To obtain this r, you will need to code your variables with 0 and 1, performing what is known as a point-biserial correlation. This will give you a measure of effect size. Lastly, the sign of Pearson's r, in this case, is only dependent on the way you code your variables, and can be disregarded for the purposes of effect size estimation.

    Hope this helps!
    GSH

    Here's some literature for reference:

    Field, A. P. (2005). Is the meta-analysis of correlation coefficients accurate when population correlations vary? Psychological Methods, 10(4), 444-467. doi:10.1037/1082-989X.10.4.444

    Field, A. P. (2001). Meta-analysis of correlation coefficients: A monte carlo comparison of fixed- and random-effects methods. Psychological Methods, 6(2), 161-180. doi:10.1037//1082-989X.6.2.161

    Rosenthal, R., & DiMatteo, M. R. (2001). Meta-analysis: Recent developments in quantitative methods for literature reviews.
    Annual Review of Psychology, 52(1), 59-82. doi:10.1146/annurev.psych.52.1.59

    Comment


    • #3
      Re: Effect size dependent sample

      This is excellent thank you so much for your input.

      Comment

      Working...
      X