1. Changes over time in union relative wage effects in Great Britain and the United States

As all economics data files have weaknesses – measurement error, unmeasured variables, sample survey quirks – and all model specifications are questionable, contaminated by data mining, any 'finding' ought to be replicated on several data sets and under 'plausible' model specifications before one accepts it as valid. Replication with additional data and specifications contrasts sharply with the practice of econometricians who postulate a 'true' model, use maximum likelihood search procedures to extract its parameters from data, and stop, as if technical prowess rather than robustness of results was the key to credibility. In fact, as all practitioners know, any single piece of complex econometric analysis rarely convinces anyone, for the more sophisticated the econometrics, the greater the danger the results derive from the model than from the world. In economics it is the accumulation of disparate lines of evidence, not the elegance of the statistical technology for a single estimator, that is compelling. (Freeman 1989, p. xi)