By Susan Lloyd, Matthew Beckman
Information
In many introductory statistics textbooks, hypothesis testing is presented as a rote multi-step procedure. This training has led many research communities to believe that only statistically significant results are meaningful. Although there are some traditionalists who are resistant to change, many feel that eliminating the notion of statistical significance would produce more ethical research. Due to the desire to shift away from the dichotomous formal decision framework, interval estimation is of increasing importance in the research community. This study serves as a prototype for developing, validating, and analyzing an instrument that includes interactions among learning outcomes. We take an innovative approach to assessment development by employing a fractional factorial design to highlight the interaction effects among key learning objectives related to confidence intervals. We conduct qualitative think-aloud interviews to validate the instrument and identify students’ epistemic understandings about confidence intervals. We discuss preliminary results and analyses based on data we collected from two participating large-enrollment introductory statistics classes at Penn State to measure undergraduate students’ statistical literacy surrounding confidence intervals upon the conclusion of an introductory statistics course using simulation-based inference methods. The results of this study help to inform the findings of previous research studies on statistical literacy that include confidence intervals as one of many topics being assessed and suggest curricular and pedagogical improvements related to teaching confidence intervals.