The cognitive organization of statistical knowledge: An information processing study

The theoretical basis for this study was based on recent information-processing theories of human cognition based on the work of Newell and Simon, and later by John Anderson. This construct has been used in previous work in the study of expertise in fields such as algebra, physics, and computer programming. The purpose of this study was to apply this construct to investigate similarities and differences in expert and novice performance with statistics. A summary of the need for this kind of research was given, as well as a summary of previous research which has applied the information-processing theory of cognition to other technical fields.; Four cognitive probes which had been used in previous studies were administered to a group of undergraduate business students who were enrolled in the first semester of business statistics and to a group of professors who taught statistics, all volunteers from a major state university. The four cognitive probes used were a problem-sorting task, the memorizing and recall of problems, having subjects state aloud their basic approach in solving particular problems and the reasons why they chose their approach, and the actual solving of statistics problems. The results obtained from the four probes were generally in agreement with previous results from other studies. However, some of the results seemed to indicate that the performance of experts and novices was not as stereotypical as depicted in other studies. The performances of novices and experts can be similar at times and very different at other times. The similarities and differences with previous work were noted and some explanation given. Ideas suggested by the data for future research were also noted.