The delivery method of a math course may affect the math scores of nursing students, which relates to rates of medication errors that could be fatal. The purpose of this study was to discover the relative effectiveness of a delivery method of a math course. Benner's novice-to-expert theory guided the study. A sequential explanatory, mixed-methods, nonexperimental pre-/posttest alternative treatment design was used. Phase 1 answered which delivery methods-online self-directed, face-to-face, or a mix of online self-directed with instructor lead-were associated with the best Medication Administration Competency exam results. Phase 2 included students' assessment of each learning method. The sample size was 148 students who were admitted to 1 nursing school between 2011 and 2013. The data were collected from 4 sources: (a) archival standardized entrance exam math scores, (b) archival standardized exit exam math scores, (c) a qualitative survey regarding student perceptions of the delivery method, and (d) a qualitative section of the same survey with math questions. The ANCOVA analysis showed no statistically significant difference in the delivery method used. Students with lower pretest exam scores took the posttest exam more times and also had lower posttest grades. The content analysis showed that students from all 3 groups did not see an advantage in the delivery method, but in certain teaching strategies that support learning. Therefore, the nursing school should continue to allow students to select their preferred delivery method, or offer fewer methods as they were equivalent. Positive change could come from using teaching strategies that students valued, improving their ability to provide correct dosages and increasing patient safety in the healthcare environment.