Return to search

Empirical Benchmarks for Interpreting Effect Sizes in Child Counseling Research

The goal of this study was to establish empirical benchmarks for Cohen's d in child counseling research. After initial review of over 1,200 child intervention research studies published from 1990 to 2016, 41 randomized clinical trials were identified in which intervention and control groups were compared with children 3-12 years old (N = 3,586). Upon identification or calculation of a Cohen's d for each study, I calculated a weighted mean d by multiplying the effect size of each study by the number of participants in that study then dividing by total number of effect sizes. The weighted mean accounted for study sample size and served as the suggested medium effect size benchmark. Results indicated effect size is impacted in large part by type of reporter, with parents apparently most sensitive to improvement and yielding higher effect sizes overall; teachers relatively less sensitive, perhaps due to difficulty observing change in a classroom setting; and children self-reporting lowest levels of improvement, perhaps reflecting a lack of sufficient measures of child development. Suggested medium benchmarks for Cohen's d in child counseling literature are .70.

for parent report, .50 for teacher report, and .36 for child self-report. Small and large benchmarks are suggested based on the use of standard deviations of the mean Cohen's d for each reporter.

Identiferoai:union.ndltd.org:unt.edu/info:ark/67531/metadc984168
Date05 1900
CreatorsWeisberger, Andrea Godwin
ContributorsRay, Dee C., Bratton, Sue, Henson, Robin K. (Robin Kyle), Stulmaker, Hayley L.
PublisherUniversity of North Texas
Source SetsUniversity of North Texas
LanguageEnglish
Detected LanguageEnglish
TypeThesis or Dissertation
Formatv, 130 pages, Text
RightsPublic, Weisberger, Andrea Godwin, Copyright, Copyright is held by the author, unless otherwise noted. All rights Reserved.

Page generated in 0.0018 seconds