This study aimed to investigate to what extent the use of hands-on science activities influences on students? academic achievement in science. Review of literature revealed several research studies focusing upon such aim and thus, a meta-analysis of these researches was carried out to obtain an overall effect size estimate of hands-on science activities on science achievement. Of the available studies, 15 with multiple outcomes satisfied the pre-determined inclusion criteria. In addition to the estimation of overall effect size using fixed- and random-effects models, subgroup analyses were also run through a mixed-effect model to determine whether heterogeneity in effect size estimates is due to the influence of moderator variables. Results showed that the estimated effect size was statistically significant (Z = 8.57, p < 0.01). The magnitude of the overall effect size estimate indicated that the hands-on activities had a very large impact on students? science achievement (Hedge`s g = 1.55, 95% CI = [1.20-1.91]). The effect size estimates for each moderator variable was also statistically significant at alpha level of 0.01, but no difference was found within sub-categories of moderator variables.