Measurement Issues: Large effect sizes do not mean most people get better – clinical significance and the importance of individual results
Journal of Child Psychology and Psychiatry
Published online on December 01, 2016
Abstract
---
- |2+
Background
This paper seeks to compare group statistical analysis with effect size, group measures of clinical significance (Reliable Change Index and normative comparison), and individual analysis of clinical significance.
Method
Measures of variables important to parenting and child behavior improvement (Parenting Scale, Eyberg Child Behavior Inventory, and Parenting Stress Index [PSI]) were administered pre and post for a 9‐ to 10‐week group Behavioral Parent Training Intervention. Analysis compares traditional group statistical significance testing with group measure of clinical significance and individual analysis of clinical significance.
Results
All three measures demonstrated statistically significant differences from pre to post, with large effect sizes. Group measures of clinical significance, however, demonstrated meaningful change only on the PSI, while individual analysis showed improvements of 54% of participants at best and 0% at worst.
Conclusions
Individual analysis of clinical significance provides valuable information in treatment outcomes and should be included as a standard practice in outcomes research.
- Journal of Child Psychology and Psychiatry, Page 163-166, October 2018.