ASSESSING STATISTICAL REASONING IN DESCRIPTIVE STATISTICS: A QUALITATIVE META-ANALYSIS

To date, there are abundant studies on statistical reasoning in descriptive statistics and inferential statistics. Nevertheless, the types of statistical reasoning assessments used in those studies are different from each other. Hence, this qualitative meta-analysis is aimed to explore the methods utilized in assessing statistical reasoning among students from all levels in descriptive statistics. A total of 36 studies on reasoning about measures of central tendency, variability and distribution were found and reviewed in this paper. It was noticed that six major types of methods were employed to assess students’ statistical reasoning in descriptive statistics, namely interview, survey or questionnaire, tasks, tests, minute paper, and teaching. This study contributes considerably to the statistical reasoning area as it provides new information on statistical reasoning in descriptive statistics. For future studies, some recommendations are proposed to improve statistical reasoning assessments.