A meta-analysis is reported of 88 studies, examining 100 study populations, in which the handedness of 284665 individuals has been assessed. The overall incidence of left-handedness was 7.78%. The incidence of left-handedness was not related to the method of measurement, or the length or number of response items included in inventories. Study populations with lower response rates and smaller study populations showed some evidence of higher incidences of left-handedness, presumably due to response biasses. There was no evidence that the incidence of left-handedness was related to the year of publication of studies; however the incidence of left-handedness was lower in older subjects and in those from earlier birth cohorts, the two effects not being statistically distinguishable.Information was available from 64 study populations concerning the incidence of left-handedness in males and females; overall 8.52% of males were left-handed compared with 6.69% of females, the male incidence being 27.4% higher than that in females. Although there was some suggestion that the sex difference was greater in larger studies, and in studies whose main purpose was not the study of handedness, these differences were not significant. It is concluded that the size of the sex difference is unrelated to any of the moderator variables we have studied.It was not possible to carry out a meta-analysis of degree of handedness due to wide-spread differences in the method of reporting of degree of handedness.We recommend that future studies of handedness should, as a minimum, use one of three standard methods of assessment, so that comparison of studies is facilitated. Note: This manuscript was originally prepared in 1993 but due to problems at a major journal, described briefly, was never eventually published. It has however been cited on a number of occasions, and has been available at https://www.ucl.ac.uk/medical-education/publications/unpublished-manuscripts/meta-analysis-of-handedness . A major meta-analysis of handedness in 2019 by another author has now been submitted which cites this manuscript, and therefore it needs to be available in a more archivable format.