Students are in the classroom to learn. This is not a controversial statement. What the statement lacks, however, is the conceptual definition of “learning”. It is my belief that my classroom’s purpose is to disseminate tools students may use to process the world around them, increase understanding of it, and then devise actionable solutions to the problems that interest them. This comprises my definition of “learning.” In the field of economics, these tools may be as practical as a method for handling statistical issues in a dataset, or they may be as general as frameworks for understanding human action.
Tools have three components for usefulness. First, one must have a tool in order to use it. Second, one must understand the context for the tool – its purposes and its limitations. A Phillips-head screwdriver is useful for removing or tightening screws of a specific type, and it useless when the problem at hand is a nail. Finally, the construction and development of the tool itself speaks to process by which new tools are developed. My teaching philosophy is guided by these three components, and my teaching experience shows direct evidence of this.
Teaching the Tools
Explaining and understanding the world around us requires us to abstract to simplifications. Models and theories allow us to do this, and thus are the core tools taught in my classroom. In “Statistics for Environmental Scientists,” which I taught in Fall 2016 at Duke University, I presented statistical modeling techniques ranging from simple t-tests to ANOVA models to linear and non-linear regression as tools for explaining environmental data and phenomenon. A large part of my belief in tools-based teaching is that we learn by doing. Thus, I took a hands-on approach, providing students with actual data and actual questions, showing them how to use, for example, a t-test, and connecting the results of the empirical test back to the world as a whole. We question the nature of the population, we take a sample, we make inference on that sample, and we discuss how this relates to the population as a whole. As part of this course, students were asked to form groups and write a final paper answering a question of their choice. The results were rich and varied, with students answering questions about myriad topics, and every tool taught made an appearance.
While this statistics course centered on statistical models, tools are not limited to modeling techniques. Economic and political theories that provide frameworks for understanding human decisions (e.g. behavioral economics) or policy outcomes (e.g. punctuated equilibrium) are teachable tools in my portfolio as well. Tools, in general and in my view, include critical thinking, problem-solving, and many other more abstract concepts.
Context and Limitations
Learning about a tool requires more than executing it. It is vital to teach the context for a tool – the settings in which it is applicable, the assumptions it requires, and the limitations in its conclusions. In the context of “Statistics for Environmental Sciences,” I presented the underlying assumptions of each model, and following the “learn by doing” philosophy, wrote labs that required students to select the right tool, test its assumptions, and justify its use. For part of the final assignment, students were required to identify a statistical issue in the chosen model and independently find a solution to this issue. For example, a student who chose to analyze time series data identified an issue in the residual-versus-fitted plot of his model. Although the course did not cover serial correlation, he researched the issue and identified an appropriate method for adjusting standard errors. This continuation of my hands-on approach ensures that students are well-equipped to solve new problems in the future, which is vital to my teaching philosophy.
Construction and Development of New Tools
As students progress in their understanding and application of tools, it is important that they have a foundation in how and why existing tools have been developed. The purpose of this is to foment new tools – new frameworks and theories to explain the unexplained phenomenon, and new ways of testing these theories. As an example of one means of accomplishing this, I would assign students in an econometrics course the task of identifying and presenting the methods and history of one specific model (e.g. difference-in-differences, synthetic counterfactual method, propensity score matching), giving them an understanding of how a model works as well as why the model was developed.
My philosophy is not limiting in the definition of “learning.” Instead, it recognizes that students’ progress in life will be measured in challenges successfully met – in questions chosen and successfully answered. My hands-on, tools-based philosophy encourages the growth of researchers, problem-solvers, and critical thinkers in an efficient manner, and with measurable results.