April 1983

Psychotherapy and Neurotics-Reply

Author Affiliations

School of Psychiatry University of New South Wales Sydney, Australia

Arch Gen Psychiatry. 1983;40(4):471. doi:10.1001/archpsyc.1983.01790040125026

In Reply.—  Our reanalysis of the data from Smith et al was necessarily confined to their data base, and we were not free to add or subtract articles. Further, because we wanted our results to be directly comparable with those of the original authors, we confined ourselves to the methods of analysis they used. In meta-analyses of other sets of data, we carried out the procedures recommended by Higgitt and Fonagy to confirm that the findings remain robust (1) when the study and not the effect size is the unit of analysis,1 (2) when the number of effect sizes derived from a study is limited by the number of subjects in the study,1,2 and (3) when effect sizes are weighted to control for varying numbers of subjects in the treatment groups.With respect to our article in the Archives, we have now made the two procedural corrections requested by Higgitt

First Page Preview View Large
First page PDF preview
First page PDF preview