1

I'm evaluating different experimental studies that all included an experimental group, a treatment group and a two measurements (pre-test, post-test) and usually report the mean, standard deviation and sample size for all measurements. They don't report pre-test post-test correlations though.

To calculate the effect size I'm using the procedure advised by Morris (2008) using the mean pre-post change in the treatment group minus the mean pre-post change in the control group, divided by the pooled pretest standard deviation.

https://doi.org/10.1177/1094428106291059

See the formula here (p.369):

Effect Size Estimate Using Pooled Pretest SD

I'm wondering about how to calculate a confidence interval around this effect size with the given information (means, SDs, sample size per group). Also, I don't see the full name of this effect size (except for Δ delta). However I suppose that would be confusing due to other effect sizes with a similar name such as Cliffs' Δ or Glass'Δ. So should it just be called Δ or is it known under a different name?

Rasul89
  • 153
  • See https://stats.stackexchange.com/questions/3466/best-practice-when-analysing-pre-post-treatment-control-designs – kjetil b halvorsen Feb 07 '20 at 15:27
  • 1
    Thank you. I read through the discussion and would probably have chosen for repeated-measures ANOVA myself as method for the analysis when setting up a pre-test post-test control design. However, regarding my question, I'm looking at a table with no more than the different means scores, sd, and sample sizes and therefore decided to follow Morris' approach (2008). I can compute an effect size like that but would like to compute the confidence interval around it as well. – Rasul89 Feb 07 '20 at 19:36

0 Answers0