Chi-squared test of independence between two categorical variables (Version 1.0)
By Matthew Aiello-Lammens
This resource is an activity designed to be completed in Swirl, an interactive platform for learning and teaching R in the RStudio console. Instructors can learn more about Swirl and how to implement a Swirl lesson here.
In this lesson, students will have the opportunity to work through a chi-squared test of independence between two categorical variables. By the end of this lesson, the student should be able to 1) construct a contingency table using R, 2) use the `chisq.test` function to perform a chi-squared test of independence, and 3) interpret the results of this test. The example in this lesson uses data from Roberts, J. 1993. Regeneration and growth of coolibah, Eucalyptus coolibah subsp. arida, a riparian tree, in the Cooper Creek region of South Australia. Australian Journal of Ecology 18, 345–350. A more detailed analysis for this case study can also be found in Logan, M. 2010. Biostatistical Design and Analysis Using R. Wiley-Blackwell. PP 478-480.
This resource was designed for a non-calculus based Introduction to Statistics for the Life Sciences course, which was comprised of primarily second and third year Biology, Biopsychology, Health Science, and Environmental Science students. The students worked through this Swirl lesson after receiving a lecture on Chi-squared tests (approximately 1 hour). The lecture material is provided in this resource and was primarily based on those available from the OpenIntro Statistics (https://www.openintro.org/stat/textbook.php?stat_book=os) resources (https://docs.google.com/presentation/d/1BQ7kVN8IEWTCXENiFLdGXjsfbZijxCIgqbMpxLkujAc/edit#slide=id.g18ef082036_0_0). Students were encouraged to work in pairs as they proceeded through the lesson.
Aiello-Lammens, M. (2019). Chi-squared test of independence between two categorical variables. Reducing Barriers to Teaching with R in Undergraduate Biology, QUBES Educational Resources. doi:10.25334/Q4244V