University Students' Ability to Apply Statistical Procedures

Abstract

Statistics educators have previously noted that university students experience some difficulty in knowing when to use statistical concepts that they have encountered in their courses. In the present study, statistics educators were asked to rate the importance of various descriptive and inferential statistical procedures for inclusion in introductory statistics courses for the behavioural sciences. Items describing research situations and presenting sample data were written, each item representing a different statistical procedure. A sample of 23 undergraduate and postgraduate students enrolled in various departments in two universities in Melbourne, Australia was presented with these items. Students attempted as many of these items as they could within a 45-minute period. They were asked to identify which procedure(s) they considered appropriate for answering the research question, to justify their choice, to state other choices they had considered and rejected, to express their level of confidence in their choice, and to rate their familiarity with the various procedures. Quantitative and qualitative data are reported, and provide detailed confirmation that the skill of identifying appropriate statistical procedures in new situations is indeed difficult. Error patterns have been identified that provide a basis for some alternative approaches to teaching this skill.

Keywords:

• Misconceptions
• Problem-solving
• Research design skills

Acknowledgments

This research was supported by a grant from the Research Committee, Faculty of Education, Monash University. The willing cooperation of staff and students in the two universities that participated in the study is gratefully acknowledged. The project (No. 97/363) was approved by the university’s Standing Committee on Ethics in Research on Humans on 16 September 1997. The authors also acknowledge the helpful comments of the three anonymous reviewers on an earlier version of this article.