Integrating research evidence: A commentary on meta-analysis

Discusses (1) the need to specify the null hypothesis and the danger of posing overly broad questions; (2) the adequacy of the empirical data base and its poor correspondence to clinical realities; (3) the frequent confounding of variables, particularly when data are aggregated across studies; (4) the limitations of multivariate statistical methods; (5) the variable quality of research, with particular stress on the proper treatment of poor-quality research; and (6) the apparent objectivity of meta-analytic methods. (14 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved).