Posts tagged ‘educational psychology’
Lecia Barker had a terrific paper in SIGCSE 2015 that I just recently had the chance to dig into. (See paper in ACM DL here.) Here’s the abstract:
Despite widespread development, research, and dissemination of teaching and curricular practices that improve student retention and learning, faculty often do not adopt them. This paper describes the first findings of a two-part study to improve understanding of adoption of teaching practices and curriculum by computer science faculty. The paper closes with recommendations for designers and developers of teaching innovations hoping to increase their chance of adoption.
I’ve published in this area before. Davide Fossati and I wrote a paper about the practices of CS teachers (based on interviews with about a dozen CS university teachers): how they made change, what convinced them to change, and how they decided if the change worked. (See blog post about this here.) The general theme was that these decisions rarely had an empirical basis.
Lecia and her co-authors went far beyond our study. She interviewed and observed 66 CS faculty from 36 institutions, explicitly chosen to represent a diverse set of schools. The result is the best picture I’ve yet seen of how CS faculty make decisions.
Lecia found more evidence of teachers using empirical evidence than we did, which was great to see. But whether students “liked” it or not was still the most critical variable:
On the other hand, if students don’t “like it,” faculty are unlikely to continue using a new practice. At a public research university, a professor said, “You can do something that you think, ‘Wow! If the learning experience was way better this term, the experiment really worked.’ And then you read your teaching reviews, and it’s like the students are pissed off because you did not do what they expected.”
Lecia discovered a reason not to adopt that I’d not heard before. She found that CS teachers filter out innovations that didn’t come from a context like their own. Those of us at research universities are filtered out by some teachers at teaching-oriented institutions:
Faculty trust colleagues who have similar teaching and research contexts, share attitudes toward students and teaching, or teach similar subjects. In describing what conference speakers he finds credible at SIGCSE, a professor at a private liberal arts university acknowledged, “I do have the anti- ‘Research One’ bias. Like if the speaker is somebody who teaches at <prestigious public research university>, the mental clout that I give them as a teacher—unless they’re a lecturer—I drop them a notch. When someone stands up to speak and they’re from a really successful teaching college <names several> or universities that have a real reputation of being great undergraduate teaching institutions, I give them a lot of merit.”
The part that I found most depressing (even if not surprising) is that research evidence did not matter at all in adopting new ways to teach:
Despite being researchers themselves, the CS faculty we spoke to for the most part did not believe that results from educational studies were credible reasons to try out teaching practices.
Lecia’s study is well done, and the paper is fascinating, but the overall picture is rather dismal. She points out many other issues that I’m not going into here, like the trade-off between cost and benefit of adopting a new practice, and about the need for specialized equipment in classrooms for some new practices. Overall, she finds that it’s really hard to get higher education CS faculty to adopt better practices. We reported on that in “Georgia Computes!” (see post here) but it’s even more disappointing when you see it in a large, broad study like this.
The New York Times weighs in on the argument about active learning versus passive lecture. The article linked below supports the proposition that college lectures unfairly advantage those students who are already privileged. (See the post about Miranda Parker’s work for a definition of what is meant by privilege.)
The argument that we should promote active learning over passive lecture has been a regular theme for me for a few weeks now:
- I argued in Blog@CACM that hiring ads and RPT requirements should be changed explicitly to say that teaching statements that emphasize active learning would be more heavily weighted (see post here).
- The pushback against this idea was much greater than I anticipated. I asked on Facebook if we could do this at Georgia Tech. The Dean of the College of Engineering was supportive. Other colleagues were strongly against it. I wrote a blog post about that pushback here.
- I wrote a Blog@CACM post over the summer about the top ten myths of computing education, which was the top-visited page at CACM during the month of July (see post here). I wrote that post in response to a long email thread on a College of Computing faculty mailing list, where I experienced that authority was able to sway CS faculty more than research results (blog post about that story here).
The NYTimes piece pushes on the point that this is not just an argument about quality of education. The argument is about what is ethical and just. If we value broadening participation in computing, we should use active learning methods and avoid lecture. If we lecture, we bias the class in favor of those who have already had significant advantages.
Thanks to both Jeff Gray and Briana Morrison who brought this article to my attention.
Yet a growing body of evidence suggests that the lecture is not generic or neutral, but a specific cultural form that favors some people while discriminating against others, including women, minorities and low-income and first-generation college students. This is not a matter of instructor bias; it is the lecture format itself — when used on its own without other instructional supports — that offers unfair advantages to an already privileged population.
The partiality of the lecture format has been made visible by studies that compare it with a different style of instruction, called active learning. This approach provides increased structure, feedback and interaction, prompting students to become participants in constructing their own knowledge rather than passive recipients.
Research comparing the two methods has consistently found that students over all perform better in active-learning courses than in traditional lecture courses. However, women, minorities, and low-income and first-generation students benefit more, on average, than white males from more affluent, educated families.
My Blog@CACM post this month makes a concrete proposal (quoted and linked below). We (all academic computing programs) should incentivize faculty to use active learning methods by evaluating teaching statements for hiring, tenure, and promotion more highly that reference active learning and avoid lecture.
On my Facebook page, I linked to the article and tagged our Dean of Engineering, the Vice-Provost for Undergraduate Education, and the RPT Chair for our College, and asked, “Can we do this at Georgia Tech?” The pushback on my Facebook page was the longest thread I’ve ever been part of on Facebook.
The issues raised were interesting and worth discussing:
- Would implementing this put at a disadvantage new PhD’s who have no teaching experience and don’t learn about active teaching? Yes, but that incentivizes those PhD programs to change.
- My blog post title is “Be It Resolved: Teaching Statements must embrace Active Learning and eschew Lecture.” I chose the word “eschew” deliberately. It doesn’t mean “ban.” It means “deliberately avoid using” which is what I meant. Lecture has its place — I wrote a blog post defending lecture which still gets viewed pretty regularly. The empirical evidence suggests that we should use active learning more than lecture for undergraduate STEM education.
- Should such a requirement for teaching statements emerge from faculty talking about it, or should it be done by administrative fiat? I lean toward the latter. As I’ve pointed out, CS faculty tend to respond to authority more than evidence. The administration should do the right thing, and deal with educating teachers (e.g., what are active learning methods first? how do we use them? even in large classes?) later. Faculty will learn the active learning methods in order to create those teaching statements. The incentive comes first.
- Lots of respondents thought I was saying that we should require all teaching to be active learning. I wasn’t, and I don’t know how to enforce that anyway. By evaluating teaching statements more heavily that emphasize active learning, we create an incentive, not a requirement.
- Some faculty pushed back, “How about students that like lecture? Tough luck for them?” Since we know that active learning is better, even for students who like lecture — yes.
- Several respondents suggested that active learning is just too hard, that faculty are over-stressed as it is. Faculty are over-stressed, but active learning isn’t that hard. In fact, it’s hard for faculty because they have to be quiet and listen in class more. It is hard to make change, but that’s the point of incentives. We start somewhere.
- The biggest theme in the thread is that we should first aim to get faculty to care about teaching and to take active steps to improve their teaching. I don’t think that’s enough. Libertarian paternalism (see Wikipedia page) suggests that we set the incentive at the minimal acceptable level (use of active learning) then encourage choice above that (choosing among the wide variety of active learning methods). We don’t want people to choose options that won’t be in the best interests of the largest number of people.
The discussion went on for four days (and hasn’t quite petered out yet). I do wonder if active learning methods will be forced upon faculty if we don’t willingly pick them up. The research evidence is overwhelming, with articles in Nature and hundreds of studies reviewed in the Proceedings of the National Academy of Sciences. How long before we get sued for teaching but not using the best teaching methods? One of the quotes in the blog post says, “At this point it is unethical to teach any other way.” We should take concrete steps towards doing the right thing, because it’s the right thing to do.
Here is something concrete that we in academia can do. We can change the way we select teachers for computer science and how we reward faculty.
All teaching statements for faculty hiring, promotion, and tenure should include a description of how the candidate uses active learning methods and explicitly reduces lecture.
We create the incentive to teach better. We might simply add a phrase to our job ads and promotion and tenure policies like, “Teaching statements will be more valued that describe how the candidate uses active learning methods and seeks to reduce lecture.”
NYTimes recently had a series of op-ed articles about the role of technology in our world, specifically, “Is Silicon Valley Saving the World, or Just Making Money?” The piece by Melinda Gates (quoted below) caught my attention because she’s invoking the desire to meet students’ “different learning styles” (see blog post on this theme, and why it leads to worse learning).
There’s an important issue here (beyond me critiquing Melinda Gates, who does important work that I admire). It’s not all technology. We need other disciplines as well. Educational psychologists should be informing these developers at Facebook to tell them, “Stop. That’s a bad idea.”
I was at a workshop last year at Stanford about how to grow more CS Education Research in the United States. Andrew Ng spoke to us about the research going on at Coursera. He was clearly not previously informed about the focus of the workshop. When asked, “Would you want to hire more PhD’s in CS Education?” he answered (my paraphrase), “Sure, but we just hire CS PhD’s, and they’re smart enough to pick up anything on-the-fly.” No, that’s wrong. CS is not a superset of all other disciplines. That belief is exactly the problem I see in the below quoted piece. Scholars in other areas do know things that CS PhD’s don’t, and they bring something unique to the table. Believing that it’s all technology is exactly why Silicon Valley gets accused of being more interested in money than having actual positive impact.
One of the biggest problems in American education is that teachers have to teach 30 students with different learning styles at the same time. Developers at Facebook, however, have built an online system that gives teachers the information and tools they need to design individualized lessons. The result is that teachers can spend their time doing what they’re best at: inspiring kids.
Briana Morrison is presenting the next stage of our work on subgoal labeled worked examples, with Lauren Margulieux. Their paper is “Subgoals, Context, and Worked Examples in Learning Computing Problem Solving.” As you may recall, Lauren did a terrific set of studies (presented at ICER 2012) showing how adding subgoal labels to videos of App Inventor worked examples had a huge effect on learning, retention, and transfer (see my blog post on this work here).
Briana and Lauren are now teaming up to explore new directions in educational psychology space and new directions in computing education research.
- In the educational psychology space, they’re asking, “What if you make the students generate the subgoal labels?” Past research has found that generating the subgoal labels, rather than just having them given to the students, is harder on the students but leads to more learning.
- They’re also exploring what if the example and the practice come from the same or different contexts (where the “context” here is the cover story or word problem story). For example, we might show people how to average test grades, but then ask them to average golf scores — that’s a shift in context.
- In the computing education research space, Briana created subgoal labeled examples for a C-like pseudocode.
One of the important findings is that they replicated the earlier study, but now in a text-based language rather than a blocks-based language. On average, subgoal labels on worked examples improve performance over getting the same worked examples without subgoal labels. That’s the easy message.
The rest of the results are much more puzzling. Being in the same context (e.g., seeing averaging test scores in the worked examples, then being asked to average test scores in the practice) did statiscally worse than having to shift contexts (e.g., from test scores to golf scores). Why might that be?
Generating labels did seem to help performance. The Generate group had the highest attrition. That make sense, because increased complexity and cognitive load would predict that more participants would give up. But that drop-our rate makes it hard make strong claims. Now we’re comparing everyone in the other groups to only “those who gut it out” in the Generate group. The results are more suspect.
There is more nuance and deeper explanations in Briana’s paper than I’m providing here. I find this paper exciting. We have an example here of well-established educational psychology principles not quite working as you might expect in computer science. I don’t think it puts the principles in question. It suggests to me that there may be some unique learning challenges in computer science, e.g., if the complexity of computer science is greater than in other studies, then it’s easier for us to reach cognitive overload. Briana’s line of research may help us to understand how learning computing is different from learning statistics or physics.
I was honored to serve on Michael Lee’s dissertation committee. Mike’s basic thesis is available at this link, or you can get the jumbo-expanded edition with an enormous appendix describing everything in his software plus his learning evaluation (described below) at this link. His thesis brings together several studies he’s done on Gidget, his game in which he teaches programming. I’ve written about his work before, like his terrific finding that including assessments improves engagement in his game (see blog post here) and about how Gidget offers us a new way to think about assessing learning (see blog post here).
Michael had several fascinating results with Gidget. One of my favorites that I have not blogged on yet was that personifying the programming tool improves retention (see his ICER 2011 paper here). When Gidget sees a syntax error, she (I’m assigning gender here) doesn’t say, “Missing semicolon” or “Malformed expression.” Instead, she says “I don’t what this is, so I’ll just go on to the next step” and looks sad that she was unable to do what the programmer asked her to do. The personification of the programming tool dramatically improved the number of game levels completed. They kept going. In course terms, they were retained.
The dissertation has yet another Big Wow result. Mike developed an assessment of computing knowledge based on Allison Elliott Tew’s work on FCS1 (see here). He did a nice job validating it using Amazon’s Mechanical Turk.
He then compares three different conditions for learning differences:
- Gidget, as a game for learning.
- CodeAcademy, as a tutorial for learning.
- The Gidget game level designer. The idea was to provide a constructionist learning environment without a curriculum. Mike wanted it be like using Scratch or Alice or any other open-ended creative programming environment. What would the students learn without guidance in Gidget?
Gidget and CodeAcademy are statistically equivalent for learning, and both blow away the constructionist option. A designed curriculum beats a discovery-based learning opportunity. That’s interesting but not too surprising. Here’s the wild part: The Gidget users spend 1/2 as much time. Same learning, half as much time. I would not have predicted this, that Mike’s game is actually more efficient for learning about CS than is a tutorial. I’ve argued that learning efficiency is super important especially for high school teachers (see post here).
Mike is now an assistant professor at the New Jersey Institute of Technology (see his web page here). I wish him luck and look forward to what he does next!
I’m currently reading Nobel laureate Daniel Kahneman’s book, “Thinking Fast, Thinking Slow” (see here for the NYTimes book review). It’s certainly one of the best books I’ve ever read on behavioral economics, and maybe just the best book I’ve ever read about psychology in general.
One of the central ideas of the book is our tendency to believe “WYSIATI”—What You See Is All There Is. Kahneman’s research suggests that we have two mental systems: System 1 does immediate, intuitive responses to the world around us. System 2 does thoughtful, analytical responses. System 1 aims to generate confidence. It constructs a story about the world given what information that exists. And that confidence leads us astray. It keeps System 2 from asking, “What am I missing?” As Kahneman says in the interview linked below, “Well, the main point that I make is that confidence is a feeling, it is not a judgment.”
It’s easy to believe that University CS education in the United States is in terrific shape. Our students get jobs — multiple job offers each. Our graduates and their employers seem to be happy. What’s so wrong with what’s going on? I see computation as a literacy. I wonder, “Why is our illiteracy rate so high? Why do so few people learn about computing? Why do so many flunk out, drop out, or find it so traumatic that they never want to have anything to do with computing again? Why are the computing literate primarily white or Asian, male, and financially well-off compared to most?”
Many teachers (like the comment thread after this post) argue for the state of computing education based on what they see in their classes. We introduce tools or practices and determine whether they “work” or are “easy” based on little evidence, often just discussion with the top students (as Davide Fossati and I found). If we’re going to make computing education work for everyone, we have to ask, “What aren’t we seeing?” We’re going to feel confident about what we do see — that’s what System 1 does for us. How do we see the people who aren’t succeeding with our methods? How do we see the students who won’t even walk in the door because of how or what we teach? That’s why it’s important to use empirical evidence when making educational choices. What we see is not all there is.
But, System 1 can sometimes lead us astray when it’s unchecked by System 2. For example, you write about a concept called “WYSIATI”—What You See Is All There Is. What does that mean, and how does it relate to System 1 and System 2?
System 1 is a storyteller. It tells the best stories that it can from the information available, even when the information is sparse or unreliable. And that makes stories that are based on very different qualities of evidence equally compelling. Our measure of how “good” a story is—how confident we are in its accuracy—is not an evaluation of the reliability of the evidence and its quality, it’s a measure of the coherence of the story.
People are designed to tell the best story possible. So WYSIATI means that we use the information we have as if it is the only information. We don’t spend much time saying, “Well, there is much we don’t know.” We make do with what we do know. And that concept is very central to the functioning of our mind.