During the COVID-19 pandemic, governments changed rules and procedures related to Medicaid enrollment. These changes…
Methods and examples to improve biostatistics education, including for students with limited mathematical backgrounds
Statistical analysis has become a crucial part in public health research and practice. Because of this, biostatistics education is a significant part of the curriculum for master of public health (MPH) students. However, methods for teaching statistical methods vary from field to field and little research has investigated best practices for teaching biostatistics to MPH students.
A new study conducted by Qi Zheng, PhD, professor in the Department of Epidemiology and Biostatistics at the Texas A&M School of Public Health, delves into statistics teaching practices to further define how best to teach biostatistics to MPH students. The study, published in the Athens Journal of Health & Medical Sciences, used examples from online courses in introductory categorical data analysis to explore a novel hands-on approach that focused on conceptual understanding as well as practical knowledge.
Frequently, MPH students lack the calculus heavy background needed to grasp the complex mathematics underlying many statistical methods. This has led to recipe-style teaching of statistics relying on statistical software packages. Although this approach enables students to conduct analyses, it does little to develop understanding or interest in the concepts involved, and the nature of online education appears to reinforce this old methodology.
In his paper, Zheng outlined methods used in an online categorical data analysis course over a five-year period to teach common biostatistics methods such as the likelihood ratio test. Interestingly, his first example was about Fisher’s exact test, which he often taught in MPH students’ first biostatistics courses. The inner workings of Fisher’s exact test appears to be mathematically complex, which leads many educators to skip in-depth discussion of the technique. However, as Zheng’s paper clearly illustrates, only a basic understanding of high school algebra concepts such as the binomial coefficient is needed to allow MPH students to grasp the test’s underlying principles.
The approach detailed in Zheng’s paper is based on the observation that abstract concepts are typically best explained by relying on basic principles and concrete examples. The important concept of the likelihood function can be made concrete using a small date set and a few lines of computer code. Zheng devised a large set of computational exercises that enabled MPH students to grasp concepts based on the likelihood function despite having limited knowledge of calculus.
Compared to a traditional teaching approach, students were more enthusiastic about the concepts taught and were more likely to view the information as valuable. This is important as MPH students have rarely expressed interest in learning abstract concepts. Traditional education methods that focus on teaching procedures as if they were a recipe can prevent students from developing an interest in learning more. Additionally, opening conceptual knowledge in such a manner helps boost student confidence, which may guide how they use biostatistics further in their studies, research and practice after graduation.
This approach also benefits instructors who might see the limited mathematical backgrounds of MPH students as too great an obstacle to overcome. The importance of integrating conceptual knowledge and statistical reasoning is well known, but previous efforts to do so have been limited by a lack of case studies. This paper fills that need by providing examples from an actual online classroom environment.
Although this study focused only on introductory categorical data analysis, it is possible that the novel approach could be used in other specialized biostatistics areas such as analyzing longitudinal data.
Relying largely on rudimentary logical reasoning and high school algebra, Zheng’s approach to teaching biostatistics appears to overcome the obstacles presented by MPH students’ limited mathematical background. The methods and examples outlined in this study highlight ways to improve biostatistics education in public health. With better approaches students will be able to learn more effectively and with more enthusiasm, and instructors will be better able to teach students in a way that overcomes their mathematical limits and promotes a desire to learn more and understand more deeply.
Media contact: Dee Dee Grays, firstname.lastname@example.org, 979.436.0611