ABSTRACT
The essence of the Columbian exposition of research data seems to consist of intuitive interpretations of heavily partialed tables. While almost anyone is in favor of the partials, the intuitive part raises some problems. Without significance tests, how is one going to tell when a finding is worth interpretation? While some of the argument is trenchant and some of it plain foolish, it is the author opinion that it is not the absence of significance tests but the absence of any formal criterion for arriving at a conclusion which typifies this approach. Furthermore, without some communicable decision process, it is even less possible for different investigators to arrive at the same conclusions from data. In a very literal sense, the new survey methodology comes close to an art form in which a verbal argument is presented, accompanied by complex and fascinating counterpoint on the IBM machine.
