This study aims to evaluate the effectiveness of concept maps on science achievement among elementary and secondary education students, including low-achieving students. A systematic search located 55 studies about concept mapping in science achievement published in peer-reviewed journals and dissertations between 1980 and 2020. We extracted 58 independent standardized mean difference effect sizes from 55 eligible studies involving 5,364 students from Grade 3 to Grade 12 who used concept maps for learning in physics/earth science, chemistry, and biology that met the specified design criteria. A random-effects model meta-analysis revealed that the mean effect size was moderate for overall science (g = 0.776). The mean effect sizes varied from moderate to large based on the subject area (g = 0.671 for biology; g = 0.590 for chemistry; g = 1.040 for physics and earth science); these differences between groups were not statistically significant (p = 0.220). Concept maps were generally associated with increased science learning across several learning and teaching conditions, and methodological features (low-achieving students, higher teaching guidance, intermediate grades, low- or middle-income countries, journal publications, and late year of publication). However, we found significant heterogeneity in most subsets. Implications for future research and practice recommendations are discussed.