Resumen de emagnification: A tool for estimating effect-sizemagnification and performing designcalculations in epidemiological studies
David J. Miller, James T. Nguyen
Artificial effect-size magnification (ESM) may occur in underpoweredstudies, where effects are reported only because they or their associatedp-valueshave passed some threshold. Ioannidis (2008,Epidemiology19: 640–648) and Gel-man and Carlin (2014,Perspectives on Psychological Science9: 641–651) havesuggested that the plausibility of findings for a specific study can be evaluatedby computation ofESM, which requires statistical simulation. In this article, wepresent a new command calledemagnificationthat allows straightforward im-plementation of such simulations in Stata. The commands automate these sim-ulations for epidemiological studies and enable the user to assessESMroutinelyfor published studies using user-selected, study-specific inputs that are commonlyreported in published literature. The intention of the command is to allow a widercommunity to useESMs as a tool for evaluating the reliability of reported effectsizes and to put an observed statistically significant effect size into a fuller contextwith respect to potential implications for study conclusions
