Faith Hathenbruck and Faculty Mentor: Dr. Jamie Jensen, Department of Biology
Introduction: The Lawson’s Classroom Test of Scientific Reasoning1(LCTSR) is a content-independent measure of scientific reasoning abilities including conservation, proportional reasoning, identifying and controlling variables, probabilistic reasoning, correlational reasoning, and hypothetico-deductive reasoning. A relationship has been seen between scores on this test and a student’s decision to major in a STEM degree2 as well as with their performance in science, technology, engineering and mathematics, also known as STEM, courses3. Thus, the LCTSR can be a clever tool in predicting a student’s performance and persistence in STEM degrees. This test, however, has more potential than just predicting STEM success; we hypothesize that it may serve as a predictor for all academic success. In preliminary research, we have seen evidence that at least some of these reasoning skills do indeed transfer between disciplines as disparate as social studies, art, and music (data unpublished). If reasoning skills correlate to success in STEM fields, then is it possible that reasoning skills could predict performance in other academic areas? This is our research question.
In order to address this question, we administered the Lawson’s Classroom Test of Scientific Reasoning (LCTSR) to over 1300 students, Predominantly from non-majors Bio 100 courses. We then collected their science and non-science GPA’s and ran a Pearson Correlation to determine the correlation between LCTSR and non-science GPA as well as LCTSR and science GPA. As an added element we also recorded ACT scores and measured the correlation between LCTSR and ACT scores as the ACT is meant to be a predicter of academic success. The LCTSR was administered for class credit in introductory level biology courses and an alternative assignment was given for those students who did not want to participate in the research.
Once we ran the correlation we had to eliminate some of the data due to clerical error and student error. For example, we received some data that showed a zero GPA. Since this is likely due to an error with the data or possibly a student who withdrew from a STEM class we eliminated those data points. With the LCTSR we eliminated any student who scored below a nine, indicating that they are a pre-operational thinker (typical of children under 8 years of age). In this case, we made the assumption it was more likely that the student did not perform to the best of their ability (as the test was not administered for a grade), and this is why their LCTSR score was so low. Students who get into a highly-selected university, such as BYU, would likely have higher reasoning ability. Therefore, we excluded those data points. Once, the data was cleaned we ran the correlation again and recorded those values.
Results: The science and non-science GPA’s had a correlation to the LCTSR of .199 and .070 respectively. However, since the sample size was so large the correlation between science GPA and LCTSR was still significant at the .01 level. Likewise, the correlation between the LCTSR and non-science GPA was extremely minimal but significant at the .05 level, so it was significant that the correlation between non-science GPA and LCTSR was so small. Interestingly, the most highly correlated elements were ACT to LCTSR score, with a correlation coefficient of .362, which again, due to sample size, is significant at the .01 level. This all points to the conclusions that the LCTSR is only a weak predictor of science GPA, a moderate predictor of ACT score, but cannot predict non-science GPA.
Discussion: The LCTSR has been a means of testing the ability of students to reason though scientifically-oriented problems. This test has proven to correlate to their performance in STEM classes, and through our own experimentation, we have been able to confirm this correlation. However, our hypothesis that the LCTSR would not only correlate to academic success in STEM majors but also to overall academic success was rejected. This means that while the LCTSR is a decent source for predicting performance in STEM courses, it fails to apply to any other kind of discipline. When we tried to correlate this test to non-science GPA we found that there was virtually no correlation. There are many factors that could contribute to this lack of correlation. We had no way of controlling for age, major, year in school, and many other factors that could influence GPA.
When we ran the regression on the LCTSR and ACT it turned out the LCTSR was a better indicator for how a student would perform on the ACT than predicting their overall GPA and therefore their performance in college. This is an interesting observation because it means that increasing scientific reasoning ability (those skills measured by the LCTSR) has the potential to increase performance on the ACT. This could lead to more students getting into college, increasing the amount of college educated professionals in the work force. So, while the LCTSR may not necessarily predict the degree of success a student will have in college, it could serve as a valuable tool to help students be better equipped at getting accepted into college.
Conclusion: Through this experiment we have been able to conclude that the LCTSR is a valuable measure of academic success on some fronts, although it does not correlate to overall GPA. Thus, our hypothesis that a student’s score on the LCTSR would correlate to their non-science GPA was rejected. However, we were also able to conclude that the LCTSR does correlate to science GPA and ACT score. This could lead to a new set of experimentation to determine if increasing student reasoning ability will also lead to an increase in ACT score, as our correlation would suggest.
1 Lawson, A. E. (1978). The Development and Validation of a Classroom Test of Formal Reasoning. Journal of Research in Science Teaching, 15(1), 11-24.
2 Jensen, J. L., Neeley, S., Hatch, J. B., & Piorczynski, T. (2015, Online). Learning scientific reasoning skills may be the key to retention in science, technology, engineering, and mathematics. Journal of College Student Retention. doi: 10.1177/1521025115611616
3 Lawson, A. E., Banks, D. L., & Logvin, M. (2007). Self-efficacy, reasoning ability, and achievement in college biology. Journal of Research in Science Teaching, 44(5), 706-724