| Students of introductory level business statistical methods courses and inexperienced users of statistical techniques often find it difficult to recall the basic assumptions and recognize the appropriate statistical models or techniques to employ for a given type of data analysis. This frequently leads to the improper use of the available statistical methods.;The purpose of this research was to examine a prototype knowledge-based system, titled STATHELP, used as a decision aid in the selection of the appropriate statistical techniques within a limited domain. A sample of 156 undergraduate students were randomly assigned to one of four treatment groups. Groups 1 and 2 consisted of students who had completed the second semester statistical methods course during the previous school year. Groups 3 and 4 were composed of students currently enrolled in this course.;Group 1 used only a textbook as a performance aid when determining their answers to a common examination composed of typical statistics problems, while Group 2 used only STATHELP. In contrast, Group 3 was allowed only a self-designed fact-sheet for the common examination and Group 4 was allowed to use only STATHELP as a performance aid.;The dependent variable examined in this study was examination scores. Analysis of variance was employed to test for significant differences between the average examination scores of the treatment groups. In addition, grade point averages, American College Test composite and mathematics scores, computer knowledge levels, attitude toward computers, and attitude toward statistics were introduced as blocking factors.;The results of this study indicated no statistically significant differences between the mean scores of the three blocking factor levels for each variable and no significant interaction effects between the blocking factors and treatment groups. There was no significant difference in the mean scores of the groups using the textbook and fact-sheet aid, nor between the means of the two groups using STATHELP. The results did reveal, however, a significantly higher average examination score for the two groups using STATHELP when compared to the two groups using the fact-sheet and the textbook performance aids. |
