To succeed in any new territory, first learn the language. In that sense, Matt Masten’s Mod•U videos on causal inference could be considered the Rosetta Stone series of econometrics.
“Every field has its own jargon and methods,” said Masten, an assistant professor in the economics department who researches econometrics. “The first goal of our videos is to help bridge that gap between what’s being done in economics and what students outside that field know.”
The Social Science Research Institute created Mod•U as a short, simple, straightforward way for students taking research courses outside of their major to learn the concepts and terminology they need to know to make the most of a class tangential to their field. Unlike traditional MOOCs (Massive Open Online Courses) such as Coursera and Lynda that try to replicate a traditional lecture online, Mod•U videos aim to explain concepts in bite-size pieces.
“You could watch one on the bus or while walking between classes,” Masten said.
The videos, at modu.ssri.duke.edu, are grouped by concept category—“Noncompliance,” “Basics of Regression Discontinuity,” “Your Guide to Instrumental Variables” and more—and each topic has a half dozen to two dozen short videos, most presented lecture style, as if the viewer has a private tutor explaining the concept one-on-one.
Rather than take the place of a textbook, the videos are meant to supplement what’s taught in class—a refresher for upper-level students who have not thought about the concepts since their early days at Duke, or for students who need to get up to speed right away while taking an upper-level elective and don’t have time to take an entire prerequisite class.
The modules also enable a professor to bring forth new concepts that haven’t yet appeared in classic textbooks. “Part of the idea of the videos was to push some of these ideas that aren’t taught at the undergraduate level and make them accessible to undergrads,” Masten said.
Masten put together his videos to support Bass Connections classes, in which undergraduates, graduate students, and faculty come together to conduct multidisciplinary research. Mod•U enables professors to customize learning and squeeze out extra time in a lecture. The videos fill in gaps in a student’s knowledge that the professor won’t cover in class, leaving more time to synthesize and analyze ideas.
Will Goldsmith, a doctoral student in history, made some of the videos over the course of a year. The modules are not so much a substitution for a classroom lecture as they are an amplification, he said. Each of the topics he chose could have been a 50-minute lecture by itself, but “we were trying to create a condensed understanding,” he said.
Goldsmith, working with Jim Speckart, program manager for instructional design at SSRI, recorded lessons pertinent to the history of education and education policy, such as major public policy challenges that school reformers tried to address in creating a public school system. Each video runs from 5 to 15 minutes long. He sees them used in a homework setting, as a way to frame assigned readings so as to be prepared with what questions they want to take up in class.
“Videos stick with you more than reading a monograph at 2 in the morning,” he said.
Making the videos was far more time-consuming than preparing a PowerPoint presentation for a lecture. Goldsmith and Speckart looked for creative ways to take the videos beyond a traditional talking head. They interspersed images and pull-quotes, then went back through the recordings to sharpen them with editing.
“It was a lot of work,” Goldsmith admitted, but the videos can be reused for years because the concepts don’t change much over time.
Masten, because he had so many topics to cover, opted for a less polished production. Speckart recorded him sitting at a desk, explaining a concept. Writing a script would be too time consuming, Masten said, so his only aid was a computer screen off-camera with slides of his prompts as he would have for any lecture he delivered. But he sometimes did 15 or more takes to make sure he was clear and succinct in the posted video.
The resulting modules consisted of a network of ideas that encouraged non-linear learning.
“You get the concepts you need right now the fastest, without looking at all the side concepts that you would get in a usual class,” he said. “Once you have the concept, you can go to
the examples you care about and skip the ones you don’t.”
The grab-and-go format appealed to economics professor Chris Timmins, who wanted to bring some of the undergraduates in his environmental economics class up to speed in basic econometric methods. His class is interdisciplinary, and he doesn’t want to turn students away for lack of an understanding of econometrics. He asked Masten and Speckart to bundle some videos that would give students a quick sense of what’s happening when they begin implementing data in their work groups.
The video content ranges from the most basic concepts through quite advanced material. Students can watch any combination of the videos as they see fit. Timmins doesn’t require students to watch any of the videos, but he lets them know that they will be able to participate better in their group work if they have the tools explicated in the modules.
“Students can watch them at their leisure and get themselves up to speed,” he said.
The egalitarian nature of the videos appeals to Masten. They are available online without charge—neither Duke nor SSRI makes any profit from them.
Anyone anywhere in the world can watch them for free and get a feel for what these modern methods are,” Masten said.
“Hopefully, we’re making it as easy as possible to access and consume this material, so people can learn these concepts that are fundamental to making good decisions and good policy.”