gms | German Medical Science

Introduction: Inquiry-based learning appears to be a promising approach to support university students in the acquisition of skills related to scientific reasoning & argumentation (SRA), including the ability to evaluate scientific evidence and to draw evidence-based conclusions [1], [3]. This study examined differences in students’ SRA skills and statistical literacy, i.e. the ability to use and understand statistical numbers [2], depending on domain and study progress.

Methods: SRA skills and statistical literacy were analyzed in a quasi-experimental 2x3 study with N=212 students from three domains (social sciences, medicine, economics) and in different stages of their studies (undergraduate vs. graduate). We developed an online test inventory for the assessment of statistical literacy based on validated measurement tools. For the assessment of SRA skills, a decision-making scenario was utilized.

Results: Regarding statistical literacy, differences were observed between the domains, F(2,211)=10.86, p=.000, partial eta²=.095. Social sciences students scored lower (M_{Social Sciences}=10.96, SD=4.94) than students from medicine (M_Medicine=14.32, SD=4.63) and economics (M_Economics=13.20, SD=4.44). We found an interaction between domain and study progress, F(2,211)=4.31, p=.015, partial eta²=.04. Medicine students scored higher in the undergraduate phase (M_Medicine=15.23, SD=4.21; M_Economics=12.15, SD=4.21), economics students scored higher in the graduate phase (M_Economics=15.38, SD=4.19; M_Medicine=13.40, SD=4.90). Regarding SRA skills, students did not differ across domains, F(2,202)=.070, p=.932. We found a main effect for study progress in social sciences and economics, F(1,122)=5.82, p=.017, partial eta²=.05. In these domains, students scored lower in the graduate phase (M=10.49, SD=4.74) than in the undergraduate phase (M=12.47, SD=3.86).

Discussion: In general, students in economics and medicine demonstrated stronger skills in statistical literacy than social sciences students. The cross-sectional results suggest a decline of statistical literacy and SRA skills between undergraduate and graduate phase, indicating an urgent need to foster these skills during university studies.