We performed a meta-analysis to examine associations of occupational exposure to extremely-low frequency magnetic fields (ELF-MF) with amyotrophic lateral sclerosis (ALS). Epidemiologic studies were identified in EMBASE and MEDLINE, in reference lists and a specialist database. We included studies that reported risk estimates of ALS in association with occupational ELF-MF exposure. Summary relative risks (RR) or odds ratios (OR) were obtained with random effect meta-analysis, and analyses were stratified by type of exposure assessment. This was done to evaluate whether observed heterogeneity between studies could be explained with differences in the way the exposure had been determined. We included 20 studies in our meta-analysis. Overall, studies reported a slightly increased risk of ALS in those exposed to higher levels of ELF-MF compared to lower levels with a summary RR (sRR) of 1.14 (95% Confidence Interval [CI] 1.00-1.30) and for workers in electrical occupations (sRR 1.41, CI 1.05-1.92), but with large heterogeneity between studies (I  > 70%). Self-reported exposure or occupations determined from death certificates did not show increased risks. Highest-longest types of exposure translated into increased risks of ALS if the studies had evaluated the whole occupational history, in contrast to evaluating only few points in time (e.g., from census records); sRR were 1.89 (CI 1.31-2.73, I 0%) and 1.06 (CI 0.75-1.57, I 76%), respectively. In this meta-analysis, we observed an increased risk of ALS in workers occupationally exposed to ELF-MF. Results of studies depended on the quality of the exposure assessment. Bioelectromagnetics. 39:156-163, 2018. © 2018 Wiley Periodicals, Inc.