Inside the sausage factory

I've been disecting a couple of papers that I'm heavily citing for a working paper I'm trying to finish. Thanks to google and other academic literature databases, it is now easy to trace the linkages and cross citations of papers. One would discover that an earlier paper of an author has a life cycle. Once a method of analysis is learned and the techniques developed and results get published in a major journal, the same technique and approach is then applied to a somewhat similar problem and in a different setting, these results are then published in second tier journal and so on. No surprise the ad hoc procedures used in the early papers are perpetuated as if an ad hoc procedure that was left to slide in an earlier article was a license to do it again. Some of the best social science papers I think are those whose procedures and results could be replicated by anyone in the discipline with ease. Unlike in hard science were replication of results is encouraged as a check of the validity of the approach, in the social sciences this is not a tradition.