A meta-analysis was conducted on the observational modelling literature (n = 55 studies) to quantify overall between-participant treatment effects (deltau Bi) gained for child and adult populations. The overall observational modelling treatment estimate obtained from movement dynamics (i.e. form) data was larger for adults (deltau Bi = 0.80) than children (deltau Bi = 0.24); the modelling advantage over a practice-only control condition was significant for adults only. In contrast, the overall observational modelling treatment estimate obtained for movement outcome data (i.e. task goal) was negligible (deltau Bi = -0.02) for adults, whereas it was more substantial (deltau Bi = 0.48) and significant for children. These findings are consistent with the suggestion that, as a result of observation, children show a greater tendency to achieve movement outcome goals (e.g. task accuracy) than adults, who tend to focus more on matching existing intrinsic dynamics with the observed task dynamics. We suggest that the developmental effects influencing observational modelling result from differences in individuals' intrinsic dynamics, and hence the opportunity to replicate the observed motor skill. The current meta-analysis describes reasons for equivocality in the extant observational modelling literature and elucidates a research agenda for pedagogists aiming to better understand the benefits of modelling techniques.