Abstract:

We report findings from a meta-analysis on all published and unpublished studies from our labs (total N = 9,656) examining the priming effect of the American flag on political attitudes. Our analyses suggest that, consistent with the studies we originally published in 2011 (T. J. Carter et al., 2011b), American flag primes did create politically conservative shifts in attitudes and beliefs during the initial time period when data were collected (even excluding the published studies), but this effect has since declined over time to be roughly zero, though we believe that other interpretations, including false positives, are plausible. We discuss possible interpretations of this decline effect and the importance of considering the historical context inrelation to the priming effects of symbols whose meaning is not static over time. We also highlight the value of publicly posting data, emptying file drawers, and conducting direct as well as conceptual replications.