Background Older meta-analyses of the effects of psychological treatments of social anxiety disorder have found that these treatments have moderate to large effects. However, these earlier meta-analyses also included non-randomized studies, and there are many featured studies in this area which were published after the recent meta-analysis.Method We conducted a systematic literature search and identified 29 randomized studies examining the effects of psychological treatments, with a total of 1628 subjects. The quality of studies varied. For the analyses, we used the computer program comprehensive meta-analysis (version 2.2.021; Biostat, Englewood, NJ, USA).Results The mean effect size on social anxiety measures (47 contrast groups) was 0.70, 0.80 on cognitive measures (26 contrast groups) and 0.70 both on depression (19 contrast groups) and general anxiety measures (16 contrast groups). We found some heterogeneity, so we conducted a series of subgroup analyses for different variables of the studies. Studies with waiting-list control groups had significantly larger effect sizes than studies with placebo and treatment-as-usual control groups. Studies aimed at subjects who met Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria for social anxiety disorder had smaller effect sizes than studies in which other inclusion criteria were used.Conclusions This study once more makes it clear that psychological treatments of social anxiety disorder are effective in adults, but that they may be less effective in more severe disorders and in studies in which care-as-usual and placebo control groups are used. © 2008 Cambridge University Press.