What happens to the student? The neglected variable in educational outcome research

Disputes about the superiority of teaching methods often remain unresolved. The essential question we continuously want to answer is: Which teaching methods yield the best knowledge and skills in students? Abundant literature, in medical education and in education in general, on research with educational methods as independent variables and measures of outcome (e.g., test scores) as the dependent variable often point at "no significant difference" or only small differences between methods. Many factors do influence the educational outcome in students and large statistical power (such as meta analysis) should be helpful to eliminate many sources of error. However, one source we cannot tackle this way. That is, students will usually adapt quantity and quality of studying to meet testing requirements. In doing so, they may compensate for teaching quality. Some teaching may generate more effort in students than other teaching. Since test scores reflect primarily student activities, it is their efforts that may bring differences in teaching methods close to equality in test scores. Therefore, knowledge and skills should not be considered the primary outcome of teaching but the outcome of learning activities. If we want to discriminate between teaching methods, we must at least consider what happens to students.