Abstract
The interrater reliability of three different interaction scales—the Maternal Behavior Rating Scale, the Rating Scales of Mother-Child Interaction, and the Teaching Skills Inventory—were estimated with four different statistical techniques: Pearson correlations, percentages of agreement, comparisons of raters' means, and generalizability theory techniques. The estimates varied widely, both within and across the three interaction scales. The meanings of each statistical approach are discussed, and some suggestions are given to help a researcher decide which approach to use in particular circumstances.
Get full access to this article
View all access options for this article.
