In research samples, how much of a change in scale scores should be considered "real" change?

Answer:  In research designed to compare the effects of different interventions with each other or with no interventions, changes can be tested as follows: Scores for groups of children receiving the different conditions should be statistically analyzed to see whether there are (a) significant changes from pre-intervention to post-intervention and (b) whether there are significant group differences in changes.  Statistical analyses of group data thus provide a basis for determining whether changes are "real" according to statistical criteria for chance vs. nonchance changes. 

Comments are closed