Targeting the inverted classroom approach

Image via Wikipedia

A while back I wondered out loud whether it was possible to implement the inverted or “flipped” classroom in a targeted way. Can you invert the classroom for some portions of a course and keep it “normal” for others? Or does inverting the classroom have to be all-or-nothing if it is to work at all? After reading the comments on that piece, I began to think that the targeted approach could work if you handled it right. So I gave it a shot in my linear algebra class (that is coming to a close this week).

The grades in the class come primarily from in-class assessments and take-home assessments. The former are like regular tests and the latter are more like take-home tests with limited collaboration. We had online homework through WeBWorK but otherwise I assigned practice exercises from the book but didn’t take them up. The mix of timed and untimed assessments worked well enough, but the lack of collected homework was not giving us good results. I think the students tended to see the take-home assessments as being the homework, and the WeBWorK and practice problems were just something to look at.

What seemed true to me was that, in order for a targeted inverted classroom approach to work, it has to be packaged differently and carry the weight of significant credit or points in the class. I’ve tried this approach before in other classes but just giving students reading or videos to watch and telling them we’d be doing activities in class rather than a lecture — even assigning  minor credit value to the in-class activity — and you can guess what happened: nobody watched the videos or read the material. The inverted approach didn’t seem different enough to the students to warrant any change in their behaviors toward the class.

So in the linear algebra class, I looked ahead at the course schedule and saw there were at least three points in the class where we were dealing with material that seemed very well-suited to an inverted approach: determinants, eigenvalues and eigenvectors, and inner products. These work well because they start very algorithmically but lead to fairly deep conceptual ideas once the algorithms are over. The out-of-class portions of the inverted approach, where the ball is in the students’ court, can focus on getting the algorithm figured out and getting a taste of the bigger ideas; then the in-class portion can focus on the big ideas. This seems to put the different pieces of the material in the right context — algorithmic stuff in the hands of students, where it plays to their strengths (doing calculations) and conceptual stuff neither in a lecture nor in isolated homework experiences but rather in collaborative work guided by the professor.

To solve the problem of making this approach seem different enough to students, I just stole a page from the sciences and called them “workshops“. In preparation for these three workshops, students needed to watch some videos or read portions of their textbooks and then work through several guided practice exercises to help them meet some baseline competencies they will need before the class meeting. Then, in the class meeting, there would be a five-point quiz taken using clickers over the basic competencies, followed by a set of in-class problems that were done in pairs. A rough draft of work on each of the in-class problems was required at the end of the class meeting, and students were given a couple of days to finish off the final drafts outside of class. The whole package — guided practice, quiz, rough draft, and final draft — counted as a fairly large in-class assessment.

Of course this is precisely what I did every week in the MATLAB course. The only difference is that this is the only way we did things in the MATLAB course. In linear algebra this accounted for three days of class total.

Here are the materials for the workshops we did. The “overview” for each contains a synopsis of the workshop, a list of videos and reading to be done before class, and the guided practice exercises.

• Determinants: Here’s the overview and here are the in-class problems.
• Eigenvalues and eigenvectors: Overview, in-class problems.
• Inner products: Overview, in-class problems.

The results were really positive. Students really enjoyed doing things this way — it’s way more engaging than a lecture and there is a lot more support than just turning the students out of class to do homework on their own. As you can see, many of the guided practice exercises were just exercises from the textbook — the things I had assigned before but not taken up, only to have them not done at all. Performance on the in-class and take-home assessments went up significantly after introducing workshops.

Additionally, we have three mastery exams that students have to pass with 100% during the course — one on row-reduction, another on matrix operations, and another on determinants. Although determinants form the newest and in some ways the most complex material of these exams, right now that exam has the highest passing rate of the three, and I credit a lot of that to the workshop experience.

So I think the answer to the question “Can the inverted classroom be done in a targeted way?” is YES, provided that:

• The inverted approach is used in distinct graded assignments that are made to look and feel very distinct from other elements of the course.
• Teachers make the expectations for out-of-class student work clear by giving an unambiguous list of competencies prior to the out-of-class work.
• Quality video or reading material is found and used, and not too much of it is assigned. Here, the importance of choosing a textbook — if you must do so — is very important. You have to be able to trust that students can read their books for comprehension on their own outside of class. If not, don’t get the book. I used David Lay’s excellent textbook, plus a mix of Khan Academy videos and my own screencasts.
• Guided practice exercises are selected so that students experience early success when grappling with the material out of class. Again, textbook selection should be made along those lines.
• In-class problems are interesting, tied directly to the competency lists and the guided practice, and are doable within a reasonable time frame.

These would serve as guidelines for any inverted classroom approach, but they are especially important for making sure that student learning is as great or greater than the traditional approach — and again, the idea of distinctness seems to be the key for doing this in a targeted way.

What are your suggestions or experiences about using the inverted or “flipped” classroom in a targeted way like this?

Advertisement

Finding passion

I’m finally through one of the busiest three months I think I’ve ever spent in this business, so hopefully I can get around to more regular posting here. The last big thing that I did as part of this busy stretch also happened to be one of the coolest things I’ve done in a while: I got to do a clicker workshop for some of the senior staff of the Johnson County Humane Society.

It turns out that someone had donated a set of 50 TurningPoint RF cards and a receiver to the Humane Society for use in educational programming — but nobody at the Humane Society knew how to use them or had any idea what they could do with them. One of the leaders in the Humane Society saw an email announcing a workshop I was doing on campus and contacted me about training. We had a great workshop last Friday and came up with some very cool ideas for using clickers in the elementary schools to teach kids about proper care of animals, in training new volunteers at the animal shelter in identifying animal breeds and diseases, even in board meetings.

The thing that stuck with me the most, though, about the folks from the Humane Society was their authentic passion for what they do. They really care about their work with the Humane Society and want to think of new and creative ways to express and share it with others.

This got me thinking: How can you tell what a person or small group of people are passionate about? It seems to me that there’s a two-step process:

1. Give those people a break and let them do whatever they want. Remove all the programming you have planned for them, just for a little bit. And then:
2. See what it is they talk about when there is no structure.

Whatever gets talked about, is what those people are passionate about — at least at the time. If they don’t talk about anything, they aren’t passionate about anything.

For teachers: What does this observation, assuming it’s not totally off-base, say about how we conduct our teaching? It seems to me that we fill the spaces that our students have with all kinds of programming — more topics, more homework, more of everything — until there is no space left to fill, and then when there is time to discuss anything students want, they’d rather stay silent. The passion has been beaten out of them. Might students benefit from a little more space, a little more time to play, and a lot less time trying to get to the next topic or the next example or prepare for the next test?

Speaking of the inverted classroom

On Wednesday, I gave a talk at Indiana University – Purdue Universty – Indianapolis (IUPUI, for short) to the teaching seminar for math graduate students on the inverted classroom. It was sort of a generalization of the talk I gave on the inverted linear algebra classroom back at the Joint Mathematics Meetings in January. Carl Cowen was in attendance at that talk and invited me to make the 20-minute drive from my house to IUPUI to do something like it, and I was happy to oblige.

Since putting the talk up on Slideshare yesterday morning, it’s gotten over 200 views, 2 favorites, a handful of retweets/Facebook likes, and is currently being highlighted on Slideshare’s Education page. So I thought I would share it here as well. Enjoy and ask questions!

Technology making a distinction but not a difference?

This article is the second one that I’ve done for Education Debate at Online Schools. It first appeared there on Tuesday this week, and now that it’s fermented a little I’m crossposting it here.

The University of South Florida‘s mathematics department has begun a pilot project to redesign its lower-level mathematics courses, like College Algebra, around a large-scale infusion of technology. This “new way of teaching college math” (to use the article’s language) involves clickers, lecture capture, software-based practice tools, and online homework systems. It’s an ambitious attempt to “teach [students] how to teach themselves”, in the words of professor and project participant Fran Hopf.

It’s a pilot project, so it remains to be seen if this approach makes a difference in improving the pass rates for students in lower-level math courses like College Algebra, which have been at around 60 percent. It’s a good idea. But there’s something unsettling about the description of the algebra class from the article:

Hopf stands in front of an auditorium full of students. Several straggle in 10 to 15 minutes late.

She asks a question involving an equation with x’s, h’s and k’s.

Silence. A few murmurs. After a while, a small voice answers from the back.

“What was that?” Hopf asks. “I think I heard the answer.”

Every now and then, Hopf asks the students to answer with their “clickers,” devices they can use to log responses to multiple-choice questions. A bar graph projected onto a screen at the front of the room shows most students are keeping up, though not all.

[…]

As Hopf walks up and down the aisles, she jots equations on a hand-held digital pad that projects whatever she writes on the screen. It allows her to keep an eye on students and talk to them face-to-face throughout the lesson.

Students start drifting out of the 75-minute class about 15 minutes before it ends. But afterward, Hopf is exuberant that a few students were bold enough to raise their hands and call out answers.

To be fair: This is a very tough audience, and the profs involved have their work cut out for them. The USF faculty are trying with the best of intentions to teach students something that almost assuredly none of them really want to learn, and this is exceedingly hard and often unrewarding work. I used to teach remedial algebra (well short of “college algebra”) at a two-year institution, and I know what this is like. I also know that the technology being employed here can, if used properly, make a real difference.

But if there’s one main criticism to make here, it’s that underneath the technology, what I’m seeing — at least in the snapshot in the article — is a class that is really not that different than that of ten or twenty years ago. Sure, there’s technology present, but all it seems to be doing is supporting the kinds of pedagogy that were already being employed before the technology, and yielded 60% pass rates. The professor is using handheld sketching devices — to write on the board, in a 250-student, 75-minute long lecture. The professor is using clickers to get student responses — but also still casting questions out to the crowd and receiving the de rigeur painful silence following the questions, and the clickers are not being used in support of learner-centered pedagogies like peer instruction. The students have the lectures on video — but they also still have to attend the lectures, and class time is still significantly instructor-centered. (Although apparently there’s no penalty for arriving 15 minutes late and leaving 15 minutes early. That behavior in particular should tell USF something about what really needs to change here.)

What USF seems not to have fully apprehended is that something about their remedial math system is fundamentally broken, and technology is neither the culprit nor the panacea. Moving from an instructor-centered model of learning without technology to an instructor-centered model of learning with technology is not going to solve this problem. USF should instead be using this technology to create disruptive change in how it delivers these courses by refocusing to a student-centered model of learning. There are baby steps here — the inclusion of self-paced lab activities is promising — but having 75-minute lectures (on college algebra, no less) with 225 students signals a reluctance to change that USF’s students cannot afford to keep.

Coming up in January

Fall Semester 2010 is in the books, and I’m heading into an extended holiday break with the family. Rather than not blog at all for the next couple of weeks, I’ll be posting (possibly auto-posting) some short items that take a look back at the semester just ended — it was a very eventful one from a teaching standpoint — and a look ahead and what’s coming up in 2011.

I’ll start with the look head to January 2011. We have a January term at my school, and thanks to my membership on the Promotion and Tenure Committee — which does all its review work during January — I’ve been exempt from teaching during Winter Term since 2006 when I was elected to the committee. This year I am on a subcommittee with only three files to review, so I have a relatively luxurious amount of time before Spring semester gets cranked up in February. A time, that is, which is immediately gobbled up by the following:

• I’ll be at the Joint Mathematics Meetings in New Orleans from January 6–9. This will be my first trip to the Joint Meetings since 2002, and I’m pretty excited about it. I will be giving two talks, one in the MAA Session on Undergraduate Cryptology (PDF) about my five-day micro-unit on cryptology for freshmen and the other in the MAA Session on Innovative and Effective Ways to Teach Linear Algebra (PDF) on experimenting with the inverted classroom model in linear algebra. Both of those sessions are loaded with interesting-sounding talks, so I hope to attend the entire session. I also hope to catch up with friends I haven’t seen since, well, 2002 — and maybe connect with some new ones. If you’re attending, let me know!
• The second iteration of the MATLAB course is coming up in the spring as well, and I will be doing some significant redesign work on it based on experiences and data from the first iteration. I’m constantly humbled and gratified by the interest and positive responses that the course has generated in the MATLAB community and elsewhere — and by how much interest and attention the course has received. I’ve had a chance to observe and talk to the alumni from the first run of the course during their Calculus III course that used MATLAB significantly, and their usage habits and feedback have given me some ideas for what should be positive changes in the course. I’ll elaborate on that later.
• I am teaching Linear Algebra again in the spring, as I have done for the last 4-5 years, and this year I am targeting that course for a more robust implementation of inverted classroom techniques. A lot of the students in that course will be MATLAB course alumni, so they will be used to all that inversion. But I’ve had enough experience with peer instruction and classroom response system (“clicker”) use on the one hand from this past semester (which I never blogged about, and I’ll try to remedy that) and inverted classroom approaches in MATLAB on the other that Linear Algebra seems well-positioned to benefit from a combination of these approaches. I’ll be sketching out and planning the course in January.
• Like I said, I used a lot of peer instruction and clickers in calculus this semester with great success (I think; at least the students say so). I’m teaching two more sections of calculus in the spring and will be refining my teaching using these tools. But calculus in the spring has a different flavor than calculus in the fall, so we will see how it goes.
• What I’m reading this January: Teaching with Classroom Response Systems by Derek Bruff; Learning to Solve Problems by David Jonassen; The Craft of Research by Booth, Colomb, and Williams; and catching up on a mountain of articles that accumulated during the semester.
• I’m also reading Geometry and Symmetry by Kinsey, Moore, and Prassidis leading up to an MAA review of the book. The “Prassidis” in the author list is Stratos Prassidis, who was my Ph.D. dissertation advisor.

Throw a couple of consulting projects on top of all that, and you’ve got yourself a busy January!

The case of the curious boxplots

I just graded my second hour-long assessment for the Calculus class (yes, I do teach other courses besides MATLAB). I break these assessments up into three sections: Concept Knowledge, where students have to reason from verbal, graphical, or numerical information (24/100 points); Computations, where students do basic context-free symbol-crunching (26/100 points); and Problem Solving, consisting of problems that combine conceptual knowledge and computation (50/100 points). Here’s the Assessment itself. (There was a problem with the very last item — the function doesn’t have an inflection point — but we fixed it and students got extra time because of it.)

Unfortunately the students as a whole did quite poorly. The class average was around a 51%. As has been my practice this semester, I turn to data analysis whenever things go really badly to try and find out what might have happened. I made boxplots for each of the three sections and for the overall scores. The red bars inside the boxplots are the averages for each.

I think there’s some very interesting information in here.

The first thing I noticed was how similar the Calculations and Problem Solving distributions were. Typically students will do significantly better on Calculations than anything else, and the Problem Solving and Concept Knowledge distributions will mirror each other. But this time Calculations and Problem Solving appear to be the same.

But then you ask: Where’s the median in boxplots for these two distributions? The median shows up nicely in the first and fourth plot, but doesn’t appear in the middle two. Well, it turns out that for Calculations, the median and the 75th percentile are equal; while for Problem Solving, the median and the 25th percentile are equal! The middle half of each distribution is between 40 and 65% on each section, but the Calculation middle half is totally top-heavy while the Problem Solving middle half is totally bottom-heavy. Shocking — I guess.

So, clearly conceptual knowledge in general — the ability to reason and draw conclusions from non-computational methods — is a huge concern. That over 75% of the class is scoring less than 60% on a fairly routine conceptual problem is unacceptable. Issues with conceptual knowledge carry over to problem solving. Notice that the average on Conceptual Knowledge is roughly equal to the median on Problem Solving. And problem solving is the main purpose of having students take the course in the first place.

Computation was not as much of an issue for these students because they get tons of repetition with it (although it looks like they could use some more) via WeBWorK problems, which are overwhelmingly oriented towards context-free algebraic calculations. But what kind of repetition and supervised practice do they get with conceptual problems? We do a lot of group work, but it’s not graded. There is still a considerable amount of lecturing going on during the class period as well, and there is not an expectation that when I throw out a conceptual question to the class that it is supposed to be answered by everybody. Students do not spend nearly as much time working on conceptual problems and longer-form contextual problems as they do basic, context-free computation problems.

This has got to change in the class, both for right now — so I don’t end up failing 2/3 of my class — and for the future, so the next class will be better equipped to do calculus taught at a college level. I’m talking with the students tomorrow about the short term. As for the long term, two things come to mind that can help.

• Clickers. Derek Bruff mentioned this in a Twitter conversation, and I think he’s right — clickers can elicit serious work on conceptual questions and alert me to how students are doing with these kinds of questions before the assessment hits and it’s too late to do anything proactive about it. I’ve been meaning to take the plunge and start using clickers and this might be the right, um, stimulus for it.
• Inverted classroom. I’m so enthusiastic about how well the inverted classroom model has worked in the MATLAB course that I find myself projecting that model onto everything. But I do think that this model would provide students with the repetition and accountability they need on conceptual work, as well as give me the information about how they’re doing that I need. Set up some podcasts for course lectures for students to listen/watch outside of class; assign WeBWorK to assess the routine computational problems (which would be no change from what we’re doing now); and spend every class doing a graded in-class activity on a conceptual or problem-solving activity. That would take some work and a considerable amount of sales pitching to get students to buy into it, but I think I like what it might become.