Adapting education style to improve relevance and practical skills

Part of my educational duties this past semester was as a teaching assistant for an undergraduate introductory biostatistics course. We went over the usual topics—calculating probabilities from tables, test statistics, hypothesis testing, linear and logistic regression—and I felt that the curriculum made a great effort to contextualize the material by organizing the content into goal-oriented modules. For example, linear regression was introduced as a tool for the specific goal of explaining college students’ GPA based on alcohol consumption-related characteristics. Whether or not the dataset that we gave to the students was the best source of information for investigating this relationship, I felt that there were two pedagogical ideas behind this module that were well implemented. First, the relationship being explored (I would guess) is one that piques the interest of a large percentage of the students. A lot of college students drink and almost everyone knows someone who drinks. It was great that the theme for this module was self-motivating. Second, the structure of the main assignment for the module forced students to write in a substantive way. The students had to come up with their own linear regression models to explore the relationship between GPA and drinking-related characteristics, and they were asked to write a report of their process, findings, and model interpretations—all things that are essential to understand when reading, writing, and discussing research findings.

I want to focus on these two teaching ideas for a bit and give my perspective on what we might need to do to adapt education for the future. Current events have precipitated a lot of intense and particularly emotion-backed discussion about racial injustice and inequality in general. From discussions with people much more knowledgeable and well-spoken than I am, I would have to say that perhaps the most essential asset that a lot of people don’t get from their education is an ability to think critically and argue rationally. These are hard things to do, and I think that school is the place to get people to practice.

As I’ve explained in previous posts, case-based learning is a great way to motivate the concepts being taught in class so that students can see their practical uses and therefore remember the ideas for the long-term. In creating examples and giving context to classroom content, we need to think hard about what those examples will be. Perhaps the best way to really engage students and get them to care about what they’re learning is to use the most current issues possible. For example, I think that someone who decided to create a case-based statistics course for the current generation might have a lot of success using both news and research articles about social inequality. I think the key is to relate what is being presented in the news to what research has actually been performed and get students to closely examine the relationship between these sources of information.

Rational argumentation is another skill that is lacking in our generation. Very often we tend to talk to and become friends with like-minded people, and we are not as easily forced to question our views and see fallacies in the bases for our opinions. Forcing conversation in an educational setting is a great way to break that comfort zone because there is such a diversity of opinion. I love that the statistics course that I taught for emphasized writing because it forced students to at least put some words behind what they were learning. Just having those words is several steps above rote calculation and memorization in terms of really embedding meaning and understanding. But the way to truly make the most of those words is to construct them carefully to tell a story, to make a point. Words as a list of facts do little to reinforce understanding. To this end, I think that educators should elevate the role of writing to the level of speech and debate. I see this having great potential in mathematics and statistics classrooms. Throughout a statistics course we teach students about tools and the assumptions behind them that are used to draw conclusions from data. To truly assess their understanding of statistical concepts and to put these concepts in a meaningful context, perhaps one of the key assignments or activities of the course would be to read scientific papers and have a debate. In this way, we can prepare students for the types of discussions that they’ll have throughout their life by helping them recognize reasoning flaws in the arguments of others as well as their own.


2 thoughts on “Adapting education style to improve relevance and practical skills

  1. I see you are an idealist, which is great. Allow me for a moment to be a grinch on this Christmas eve day.

    While I think it would be fantastic if every graduate course offered a completely engaging learning experience, it simply isn’t practical in a world where our most precious resource (time) is finite. If an instructor wanted to use an example from a current news story every time they taught a course, it would require them to find this new and relevant dataset, analyze it, identify the key findings that are relevant, and put them in a format to effectively communicate the information to students. Each of these steps can take LOTS of time. Professors are busy. It simply isn’t practical to expect them to put in the amount of time and effort that it would require to do this well in an introductory course, where the focus is on teaching basic techniques, and not on creating engaging discussion.

    Setting aside the burden this places on the instructor though, I just don’t think this is the most effective way to teach. For large introductory courses (such as the one you referred to), the examples have been carefully selected and refined in order to communicate specific concepts. If the instructor chose a different example each year, it likely would diminish its impact at teaching these concepts. Maybe an example is chosen to illustrate effect modification, for example. It would be a shame to go through an entire analysis and then realize that there is no effect modification in this example. Maybe this can teach the point that you don’t always observe effect modification, but then the student doesn’t get to see certain results, plots, etc. that illustrate effect modification and show how it impacts the data. The result is the student might not have as clear an understanding of the concept.

    Moreover, by using the same lecture material over and over, that material can be refined and improved based on the instructor’s experience of seeing what works and what doesn’t. I know I am always better at teaching or presenting something the second time around than the first. It is a much more efficient and effective use of one’s times to revise and perfect previously-prepared material then spending much more time doing a superficial analysis that might not even convey a point very well.

    While I agree that an important part of statistical analysis is communication, I don’t think an introductory stat class is the best place for it. There simply isn’t enough time to have these kinds of spirited debates that you are referring to. If you are interested in gaining skills for forming rational arguments about specific topics, there are a whole host of topic-focused courses at JHSPH that do exactly this.

    Perhaps there could be more advanced courses in applied statistical analysis, statistical consulting, etc. that could take some of these approaches, such as analyzing current datasets and more discussion, and these classes do exist. I just don’t think an introductory stat course is the right place for them.

    • Jon, I agree with you about efficiency being a major factor in determining how we design courses. And I concede being a bit whimsical about the feasibility of doing yearly curriculum updates just to keep material current. My intent in writing about this was in part to be a reminder to myself about potential teaching goals but also to hear other people’s thoughts. If given the chance to design and teach a small course on introductory statistics, I would want to experiment with the ideas that I presented in this post so that I could get a better idea of the types of activities, lessons, assignments, etc. that work well. And I do not find helpful discussion such as yours to be grinchy at all. Happy Christmas 🙂

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.