Identifying Your Effect Size

Effect sizes are summary statistics, like correlations (r) or effect size differences (d), used to summarise results to answer questions such as “What is the correlation between attitudes and intentions?” or “Does forming implementation intentions change behaviour?” Meta-analysis synthesizes effect sizes extracted from a set of studies to produce an overall, pooled, effect size to represent the literature. This chapter will cover several issues about effect sizes: what they are (correlations and effect size differences); how to describe their direction (positive, negative, or null); how to classify their magnitude (null, small, medium, large). Examples from the literature and practice data will be used to achieve this aim, providing you with the knowledge you need to make sense of any meta-analysis.