Correlation, necessity, and sufficiency: Common errors in the scientific reasoning of undergraduate students for interpreting experiments.

Gaining an understanding of how science works is central to an undergraduate education in biology and biochemistry. The reasoning required to design or interpret experiments that ask specific questions does not come naturally, and is an essential part of the science process skills that must be learned for an understanding of how scientists conduct research. Gaps in these reasoning skills make it difficult for students to become proficient in reading primary scientific literature. In this study, we assessed the ability of students in an upper-division biochemistry laboratory class to use the concepts of correlation, necessity, and sufficiency in interpreting experiments presented in a format and context that is similar to what they would encounter when reading a journal article. The students were assessed before and after completion of a laboratory module where necessary vs. sufficient reasoning was used to design and interpret experiments. The assessment identified two types of errors that were commonly committed by students when interpreting experimental data. When presented with an experiment that only establishes a correlation between a potential intermediate and a known effect, students frequently interpreted the intermediate as being sufficient (causative) for the effect. Also, when presented with an experiment that tests only necessity for an intermediate, they frequently made unsupported conclusions about sufficiency, and vice versa. Completion of the laboratory module and instruction in necessary vs. sufficient reasoning showed some promise for addressing these common errors.