Effect of a Statistical Test Selection Help System

Abstract

Many students and scholars use statistical tests in research. Oftentimes, however, there is more than one test available for use with any one study, meaning that it is necessary to select the one which is most appropriate to the needs of the research. Although they have already learned something about statistics, many students, and sometimes even scholars, find it difficult to choose the most relevant test without the advice of a statistician. The reason for this is, of course, the need to take into consideration a large number of variables—the manner in which the sample and/or scores were taken, the nature of the population, etc. All of these factors enter into determining the statistical test most appropriate to a given set of data and this makes it difficult to select the best test. The purpose of this study was to develop a statistical test selection help or support system which would contribute towards the alleviation of the above problem. Hopefully, it will be able to act as a "statistician" for those students and scholars who are still unskilled in the selection of statistical tests. Experimental use of this help system has shown that those who use it are able to select tests relevant to specific research designs but that those who do not are not able to do so.