Data drawn from 30 journal articles and ERIC documents reporting on gender differences in natural science achievement were re-examined. Three meta-analysis methods were used: (1) vote counts and vote-counting estimation procedures; (2) tests of combined significance; and (3) analyses of effect sizes. The three methods produced seemingly contradictory conclusions which were explained in terms of differences in the hypotheses tested by the methods, as well as the statistical properties of the methods. Effect-size analyses were found to be more informative for the study of single outcome constructs, whereas combined significance tests with or without effect-size analyses did not provide more information. Sex differences were not evident in many subject areas. However, males had higher achievement than females in physics by one-half of a standard deviation and in biology by about one-sixth of a standard deviation. (MGD)