The High Tech Small Study Group Saga

When I went to work for the Instructional Technology Program at UC Berkeley in 1987, I wanted to see if technology could offer cost effective alternatives to large lecture courses. The idea of broadcasting lectures to students working at home never appealed to me. That approach just made the big lecture bigger and more impersonal. One-on-one or small class instruction would be better for students, but not cheap. Try going to any research university and see if they would be willing to turn a five hundred student lecture into fifty seminars each with one professor and ten students.

However, one idea for breaking up big lectures into small student led study groups caught my eye early on. It was an experiment conducted by J.F. Gibbons in the early seventies. [1] At that time, Gibbons was dean of Stanford’s School of Engineering, and the problem he faced was one of extending Stanford’s distance education program to Hewlett Packard students in Santa Rosa, which was just outside the range of Stanford’s broadcast station.

Gibbons decided to use an alternative technology that is now obsolete, but was all-the-rage at the time: videotape. Since the students would not be able to phone in questions in real time, he thought it would be good for them to view the recorded lectures in small study groups.

Even in the early seventies, encouraging students to form study groups was hardly a new or controversial idea. But, interestingly, the simple idea of replacing live lectures with lecture videos and a small study group environment is a very controversial idea, even to this day, if you try to do this on campus.

In any event, Gibbons called this approach to distance education “Tutored Videotaped Instruction,” which became the title of a paper he published in the journal Science in 1977.

The basic idea was quite simple: Gibbons recorded lectures of courses that were being taught both on campus and at the remote location in Santa Rosa. Gibbons didn’t want the HP students to simply watch the lectures by themselves, so he instructed the course coordinators at HP to organize their students into small study groups of three to seven students and appoint one student to serve as the group’s facilitator.

Stanford provided these groups with lecture video tapes, study questions, and problem sets for the students to work through. The tutors would make sure the groups stopped the videos to address the problems and study questions. They were also responsible for relaying questions back to the faculty member teaching the course at Stanford.

Gibbons compared the grades attained by the HP students with those of Stanford graduate students taking the same course that semester. Both groups had the same lectures, the same homework, and the same tests. But, obviously, it was not a well “controlled” experiment in the conventional sense because students were not randomly assigned to the two groups.

Before he started, Gibbons knew his on-campus graduate students were head-and-shoulders above the HP engineers in terms of GRE and SAT scores. So, he thought the HP students would not preform as well as the Stanford students. To his surprise, however, the HP student’s grades tied those of the Stanford students in the course – there was no statistically significant difference in the grades even though one would expect the Stanford students to easily outperform the HP engineers.

Over the past four decades, Stanford has been repeating this experiment with different courses and different student groups, and essentially they get the same results time and time again. They have even done a few controlled experiments with randomly assigned student groups, and they found that the students who benefited most from the TVI approach were those most challenged by the material in the course.

It’s easy to see why distance education programs might be interested in using the TVI approach. The interesting question is why campus based institutions have not shown more interest in this approach, especially as an alternative to large lecture classes.

In 1988, I did try to use TVI on the Berkeley campus. That year the campus had received a big NSF grant to explore uses of technology in engineering education. Although I wasn’t a faculty member, I put in a proposal and to my surprise a mechanical engineering professor and I received a grant to explore the possibility of using the TVI approach at Berkeley.

We used the funding to create the High Tech Small Study Group project, and our target course was one the instructor was teaching on descriptive geometry (DG). As I recall, there were around thirty to forty students enrolled in the course, and it met three times a week at around eight in the morning. Right off the bat, he liked the idea of video tapping all the lectures because his current approach to teaching the course was very labor intensive.

The course involved drawing some very complex diagrams on the chalk board, so his normal procedure was to come in an hour early to lightly draw the diagrams on the chalk board so only he could see them. Then, when class started, he could retrace the drawings with heavy chalk lines while explaining the process of solving DG problems as he went along. His technique was effective, but it nearly doubled the time he would otherwise spend in the classroom. Also, although the course material didn’t change much from semester to semester, he still had to repeat the process of setting up the chalk boards each time he taught the course. So, he was interested in finding an easier solution.

To help us run the study groups, we partnered with two study group programs – The Minority Engineering Program and The Professional Development Program – who had both been in operation at Berkeley for many years. They agreed to work with the descriptive geometry course, but they were also interested in trying out the TVI idea in the large physics, chemistry, and math courses they normally worked with. But, there was one catch, it didn’t make sense to force students to attend the live lectures and then to view them again in study groups. It simply made more sense to take the students out of the live lectures, and just let them watch recorded lectures in conjunction with their normal study group activities.

So, the plan was to use the videotapes created in the first semester of the project as part of a TVI course in the subsequent semester. We planned to compare grades, just as Gibbons had done, and, if we found similar results, we planned to try other experiments, such as using randomly assigned groups, and to expand the project to include other courses.

Unfortunately, when other faculty found out about our plans, we had to call the project off. We were called into the office of a senior faculty member in the Mechanical Engineering department who essentially read us the riot act. He didn’t say we needed to call off the experiment because the students in the TVI course would be short changed in some way. His concern was that the experiment might be a success and, therefore, might constitute a threat to the faculty. He thought we were running the risk of innovating faculty out of their jobs; he didn’t see it as an opportunity to redeploy faculty resources to better uses, such as teaching more small seminars. So, that was the end of the High Tech Small Study Group project.

Almost twenty years later, the year I retired in fact, I had another opportunity to bring up the TVI idea. The year was 2007, and the unit I was working with at the time – Educational Technology Services – had recently introduced a service that videotaped lectures and put them on the web so students who missed a lecture – due to illness, for example – could still view it online.

This was a popular program with many big lecture courses, but a funny thing started happening when students learned they could view lectures at a time of their choosing – they stopped coming to class. Faculty found that their lectures – some with hundreds of students still enrolled – were all but empty. They also were finding that unsupervised, any-time, any-place viewing of lecture videos doesn’t work out for some students. Those with poor study habits, for example, try to watch all the lectures right before an exam with predictably bad results.

A lively debate ensued on teach net, an email list used by Berkeley faculty to discuss classroom issues. Eventually, a live forum was organized to discuss the issue. Several faculty who had recently won teaching awards were invited to present. One economics professor who regularly taught one of the big introductory econ classes focused her presentation on why she thought students should attend live lectures. Her main thesis was that live lectures – even lectures with hundreds of students like her’s – were more interactive than recorded lectures, and, therefore, better for students.

Since this was my last year at Berkeley, I decided to stick my neck out. During the Q&A session that followed her presentation, I gave a brief description of TVI and my experience with the High Tech Small Study Group project. Then I ask her if she would be interested in re-starting the HTSSG project using her class as a way to test her thesis.

At that point, the instructor who had just made interactivity the main thesis of her presentation did something that really surprised me – she switched to an entirely new thesis. She gave no direct response to my question and instead offered this: Live lectures were more inspiring than recordings and, therefore, students would be more motivated to learn in the future.

I then asked if she could think of a way to measure student inspiration and motivation as part of an experiment. She didn’t think we could, at least not in the short run, since the benefits might not show up until years after the course.

I was tempted to ask more follow up questions: Did she think we needed to give motivational lectures three times a week, fifteen weeks a semester? Might we get the same effect if we gave a few motivational lectures each semester instead of fifteen? Could faculty do something else with the time they saved that could be both motivational and educational? For example, could faculty make a few motivational lectures and spend the rest of their time visiting the small study groups?

But, I could see that she was uncomfortable with the questions I’d asked so far, as were others in the room. So, I decided to wait and see if anyone else would follow up with similar questions. To my dismay, there was no further discussion of alternatives to large lectures. Instead, the conversation quickly turned to techniques faculty could use to get students to come to lecture. I remember on professor asking if giving pop quizzes might work.

Up until that point, when it briefly looked like we might be able to restart the High Tech Small Study Group project, I was reconsidering my decision to retire that year. But, when I heard her defense of large lectures, and when no one else elected to join the discussion, I decided my initial decision to retire early was probably for the best. So, now at least, I have time to post my memories of Berkeley to this blog.

[1] Gibbons, James (1977) Tutored Videotape Instruction.

5 thoughts on “The High Tech Small Study Group Saga

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 )

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

%d bloggers like this: