12: Meta-Analysis

meta-analysis

set of statistical methods for quantitatively aggregating the results of several primary studies to arrive at an overall summary statement or conclusion about the relationship between specified variables

• secondary research design → a primary study is a data point

  • quantitative literature review procedure

    • empirically summarize relationships between variables

    • test for/examine effects of moderators

  • tend to receive more citations than primary studies

advantages of meta-analysis

• able to summarize large volumes of literature

  • more standardizes and relatively more objective than qualitative literature review

  • uses effect size as a common metric across studies

• can resolve conflicts between two or more bodies of literature by comparing effect sizes across them

• ability to investigate relationships not investigates in original primary studies

• focuses on magnitude of effects (instead of statistical significance)

• sample-weighting of effect sizes controls for sampling error

  • NOT all studies contribute equally

meta-analysis issues

• garbage in-garbage out

• apples and oranges comparison

• number of primary studies available

• selection of primary studies for inclusion

• file drawer problem

• judgement calls

hi

hi!