Meta-regression analysis as the socio-economics of economic research

Meta-regression analysis (MRA) provides an empirical framework through which to integrate disparate economic research results, filter out likely publication bias, and explain their wide variation using socio-economic and econometric explanatory variables (Stanley and Jarrell, 1989, Stanley, 2001, Doucouliagos, 2005, Stanley, 2005a). In dozens of applications, MRA has found excess variation among reported research findings, some of which is explained by socio-economic variables (e.g., researcher?s gender) and most of which contains publication bias (Card and Krueger, 1995, Stanley, 1998, Stanley and Jarrell, 1998, Ashenfelter et al.,1999, G?rg and Strobl, 2001, Stanley, 2001, Doucouliagos and Laroche, 2003, Abreu, de Groot and Florax, 2005, Doucouliagos, 2005, Rose and Stanley, 2005, Stanley, 2005a). Publication bias is itself a socio-economic phenomenon. When researchers? compensation is based on their publication records, all available research degrees of freedom will be used to increase its probability. MRA can empirically model and test socio-economic theories about economic research. The socio-economics of the academy can explain why excess variation (beyond the classical, random sampling errors that conventional standard errors measure) will likely dominate many areas of empirical economic research, and MRA can explain how. Here, we make two strong claims: socio-economic MRAs, broadly conceived, explain much of the excess variation routinely found in empirical economic research; whereas, any other type of literature review (or summary) is biased.