Examining the effect of sample size in repeated measures analysis of variance: Malpractice insurance example

The aim of this study was to examine the effect of different sample sizes on the results of repeated measures ANOVA (rANOVA). A questionnaire including questions related to sociodemographic characteristics, 33 knowledge questions regarding malpractice insurance, and attitude questions was used to collect data. Three repeated measurements were conducted on four different sample sizes (n=20,40,60,80). The pre-test was administered without any training on malpractice insurance. Post-test was administered after training. Retention-test was conducted three weeks after post-test. Sphericity assumption was tested using Mauchly's test of sphericity. Due to violation, Greenhouse-Geisser correction was conducted. One-way rANOVA was used to examine the relationship between students' pre-test, post-test, and retention-test results. Bonferroni method was conducted for post-hoc analysis. A target value of 0.90 was aimed in post-hoc power analysis. rANOVA revealed significant differences regarding questionnaire scores and post-hoc comparisons found significant differences between the pre-test, post-test, and retention-test scores in all sample size groups (p<0.001). The targeted power was achieved in all sample size groups. Consistently high effect sizes and partial eta squared values observed across all groups are considered to contribute to maintaining high statistical power across different sample sizes. In all sample size groups, it was determined that with the training given, the knowledge levels of students increased in the post-test, and that the acquired knowledge was largely retained in the retention-test. Students stated that they had knowledge gaps (97.5%) and expressed a desire to receive training (91.25%) and wished malpractice insurance to be included in the mandatory curriculum (93.75%). They suggested increasing number of medical law courses in the curriculum (90%). It is planned to recommend the inclusion of malpractice insurance as a separate topic in the medical education curriculum and to increase the number of medical law topics in curriculum as a part of efforts to improve the NCEP.