More research on large active learning spaces: the TEAL project

Recently, I posted a 1000-word literature review article on a foundational work in active learning spaces: the original SCALE-UP report by Bob Beichner and others. Let's follow up with another article, also considered foundational, that actually appeared before Beichner's article but is an outgrowth of the work of the SCALE-UP team:

Dori, Y. J., & Belcher, J. (2005). How Does Technology-Enabled Active Learning Affect Undergraduate Students’ Understanding of Electromagnetism Concepts? Journal of the Learning Sciences, 14(2), 243–279. http://doi.org/10.1207/s15327809jls1402_3

This paper describes and reports on the implementation of a complete redesign of university physics courses. The project --- Technology Enabled Active Learning, or TEAL --- was carried out in the physics department at MIT during the early 00's.

What questions does this paper try to answer?

The overarching question this paper addresses is the same as SCALE-UP: How can we redesign classroom experiences to focus on active learning more, and to do so more effectively? TEAL uses a multi-pronged approach that involves fundamental changes in pedagogy, technology and space to provide a radically different experience from the usual large-lecture setting of university physics classes.

To approach this question, the authors tap into several veins of theoretical background. The pedagogical approach of TEAL is rooted in social constructivism, the idea that learning is not static and cannot be transfered from one person to another, but rather knowledge is constructed by each individual through interactions with other learners. Linking social constructivism to the content of the focus course --- an undergraduate course on electromagnetism --- the researchers focused the course not on lecture but on activities that involve active learning and which promote conceptual understanding.

The authors note that up to that point, physics education research had focused on mechanics rather than electromagnetism --- problematic, since EM contains concepts that are harder for students to grasp or experiment with. Therefore a central part of the TEAL experience is the use of digital technologies, especially those that can simulate the ideas under study and provide visual imagery. A centerpiece of the technology used in TEAL is the "micro-based lab", what we today would call an interactive computer simulation that allows students to perform "desktop experiments" and visualize the results of experiments in real time.

TEAL uses classroom space in a similar way as SCALE-UP. Rather than having fixed seating in a lecture hall, TEAL classes are conducted in a large open space, using round tables seating 9 students each, segmented into three groups of three with one laptop provided per team-of-three. The TEAL class in the study had 13 such tables in the room, so 117 students were in a single section of the course.

The room also features plentiful whiteboard space, lots of projection for the laptops, and a student-centered design where the instructor --- although she has a station in the middle of the room --- is not the center of attention.

The TEAL project was initiated in 2000 and reached full implementation in 2005. The study here is a report from Fall 2001, Spring 2002, and Fall 2003. The authors state three research goals:

  1. To characterize student interactions while studying in small groups during class (the social domain)
  2. To assess changes in students' conceptual knowledge about electromagnetism during a class heavily focused on active collaborative work (the cognitive domain)
  3. To analyze students' attitudes toward the TEAL environment and study their preferences regarding the teaching methods (the affective domain)

What were the methods?

This is a quasi-experimental study, which means there were control and experimental groups involved but students were not randomly assigned to those groups. The same course was targeted, a freshman-level EM class at MIT. The Fall 2001 sections (N = 176) and Fall 2003 sections (N = 514) were the "experimental" groups, set up and taught using the TEAL framework. The Spring 2002 section (N = 121) was the control, taught using traditional lecture in a traditional classroom. All groups studied the same topics in the same sequence and were given the same sets of evaluations:

  • Pre- and post-tests using 20 multiple-choice conceptual questions developed by the researchers, with different forms used to avoid the effect of prior exposure in the pretest
  • In the post-test, one open ended conceptual question asking students to summarize the main ideas of electromagnetism according to their order of importance without using any formulas or equations

Using the conceptual tests, students were sorted into ability levels (low, intermediate, high) for further analysis. Students were also given open-ended surveys and participated in focus group discussions to gather qualitative data. Finally, observations were conducted during class time to gather field data on student interactions.

What was discovered?

In the cognitive domain, the study found:

  • The failure rates were less than 5% in the experimental groups, versus 13% in the control group.
  • The normalized gains on the conceptual tests were statistically significantly higher for the experimental groups as a whole than for the control group, with the largest gains coming from students whose conceptual test scores classified them as being of "low" ability. For high ability students (e.g. students who started EM in the Fall as freshmen because of AP scores), there were no significant differences between the gains in scores of the experiment and control groups. "Intermediate" level students in the TEAL classes also had significantly higher conceptual gains but not as much as the "low" students.
  • There was a small but statistically significant correlation between the pretest mean scores and the course grades, and a large and significant correlation between the posttest means and course grades. That indicates that we can take the conceptual tests as a reliable proxy for overall understanding of the subject -- including the parts of the course that focused on computation and not concepts.

In the social domain, the paper goes into detail to report on one of the observations (more detail than a 1000-word report can hold). It paints a picture of a highly interactive environment in which students are acquiring and modifying their conceptual understanding of electromagnetism at a rapid pace and converging on correct scientific understanding, each student on his own terms and it's clear that the TEAL environment facilitated this.

In the affective domain, students rated the problem solving experience in TEAL as being the most beneficial element for their learning. 70% of the Fall 2001 TEAL class said they would recommend TEAL to others; only 53% of the Fall 2003 class said the same, though.

What does this mean for teachers and other ordinary people?

The TEAL study indicates that the pedagogical approach taken --- lots of focused active learning centered around well-designed tech-driven activities, with a view toward socially constructed conceptual understanding --- can have a powerful, positive effect on students, especially the weakest ones. And as with SCALE-UP, it can be done even in large sections of courses where many instructors just assume that lecturing is the only way to go.

It's also worth noting that even though the focus in TEAL is on conceptual understanding, which presumably involves spending less time covering content and working out computations, students in the TEAL sections improved in all phases of the course including basic content knowledge and computational fluency. This is similar to results from research on peer instruction. Focusing on conceptual understanding has a Pareto effect in which investing time on a small, critical subset of material in the course causes disproportionate gains across the board. If you're a teacher and you're concerned that taking an approach like TEAL will mean you can't cover all the material or that students' basic skills will suffer: Don't be.

What are the strengths and weaknesses?

As in many other learning-space oriented studies of this time period, the extent to which the results of the study can be decoupled from the learning space itself is unclear. I said above that you don't need a full-on SCALE-UP or TEAL classroom to reap some of the benefits of the approach, but do you need the space to get the full effect? This study can't say. We would need a secondary study where one class is done TEAL-style in a TEAL classroom and another is done TEAL-style in a regular classroom, and see what happens. Studies like this have since been done, and I may report back on one of those later.

It's reasonable also to have some concerns about the external validity of the results, coming from the fact this was done at an elite STEM-focused institution with a unique academic culture. The "low" ability students at MIT are likely not analogous to the "low" ability students at, say, Grand Valley State University or a community college, for example. This is no disrespect to MIT, just a concern that could be addressed by replicating this study at multiple, varied sites.