Welcome new readers!

The "New to the blog? Start here" page will give you an overview of the blog and point you to some posts you might be interested in. You can also subscribe to receive future posts via RSS, Facebook or Twitter using the links on the right-hand side of the page, or via email by entering your address in the box. Thanks for reading!

Monday, November 22, 2010

Social norms

Speaking of having students set some criteria, I've also been thinking about having students set some of the class rules (that I would typically set down for them). I was originally thinking about this in the context of my writing class, having students decide things like what the penalty would be for late assignments (and what constitutes 'late'), etc. But then I heard about the Cornell instructor who got upset when a student yawned super-loud in one of his classes. While I can see why people think he over-reacted, I am completely sympathetic - whenever this has happened to me, I have been sorely tempted to stop class and make a sarcastic comment to the yawner. I mean, it's just so freakin' rude! [Note: I'm not talking about just simple yawning here (though personally, I was taught that if you're going to yawn while someone is talking, you at least cover your mouth and try not to be obvious about it). Visually obvious I can live with; it's the ones who actually make a loud "aaaaahhhhhhmmmm" that make me livid (thankfully, this has been incredibly rare in my classes). You may not be able to control whether you yawn or not but you sure as heck can control how much noise you make while doing it.]

But the main thing that keeps me from calling out these idiots (aside from my impeccable sense of professionalism, of course :-)), is that I wonder how the other students will react. I have to assume that at least some of them are as appalled as I am, but I suspect that many will react as people have been reacting to the Cornell guy and think I'm way over-sensitive. But what if I asked students at the beginning of the semester to come up with a specific code of conduct? This would include having them determine what is and is not acceptable behavior (reading papers in class? cell phones going off? coming in late/leaving early?) and what my/the class reaction should be when such behavior does happen (ignore it? kick people out? dunce cap in the corner?). On the one hand, I can imagine most students already know what is, and is not, appropriate behavior, so I don't expect them to come up with a list that is all that surprising; but I also imagine that most students think that when these behaviors do happen in class, there is some good reason for them (at least, I have to believe they think that, or else these behaviors would never happen, right?). So I have no idea what they think should be done about these behaviors. For example, most students know they should turn their cell phones off; they (and I) also know that sometimes they forget. So my guess is that they will say it isn't appropriate behavior but I shouldn't necessarily do anything about it when it happens. But I wonder if simply having the conversation, having them tell me what is and is not appropriate behavior (and hearing what their peers consider appropriate or inappropriate), will make them work a little harder to act appropriately? I can't really imagine it would lead to more inappropriate behavior... If anyone has had experience with this sort of things (i.e., having students themselves establish the code of conduct for a class), please let me know how it went!

Wednesday, November 17, 2010

Letting students set the team criteria

For some reason, I woke up this morning thinking about next semester (yeah, I've got issues...). Since this is my first semester teaching the data class, or using TBL, of course I have a list of stuff that I want to change next time around. What I was thinking about this morning was how to create the groups. This fall, I created the teams by asking students for some basic information and then I just tried to make sure each team had a good mix of students (i.e., mix of good and not-so-good grades, gender, econ versus business interests, etc.). For the most part, it's worked out well but I'm wondering if there's a better way.

In my Econ for Teachers class, they have group projects (but not semester-long teams) and I did an exercise where I asked students to brainstorm what characteristics they would want in their group members. Their responses boiled down to four main factors: subject knowledge (this is a course of mostly social science majors with a handful of econ majors and the social science majors wanted at least one econ major in each of the groups), reliability (attendance, doing the work you say you're going to do, etc.), "niceness" (wasn't sure what else to call this but students mentioned things like being supportive and helpful), and flexibility (students seemed worried about trying to find time for groups to meet outside of class to get work done). Once we had these four traits, I asked students to rate themselves in each category, simply on a three-point scale (high, average, low). Then I made groups by distributing the "high" and "low" responses as equally as I could across the groups. One thing I thought was pretty funny is that while many students rated themselves "low" on knowledge or flexibility, not a single student rated themselves "low" for reliability or niceness (of course, there is some sample selection since some of the least reliable students were not even in class that day).

Now that I know the students better, I believe that they were not being falsely modest about being nice but some do have some issues recognizing their own level of reliability. So I would need to think about that. But I wonder how much simply having the conversation about these characteristics may have affected how students see themselves as members of their team. Do students who rated themselves as 'high' on niceness or reliability feel any obligation to be nicer or more reliable? I'll have to figure out a way to ask them in the end-of-semester evaluation... But regardless, I think this could be a good approach for my TBL courses next semester. At the very least, it would give students some idea of concrete things to think about when evaluating their teammates later on.

Tuesday, November 16, 2010

Budget simulations

A bunch of econ blogs have been talking about an interactive graphic puzzle on the New York Times website where you can 'fix the budget' by choosing which programs to cut and which taxes to raise. The options include some of the most recent policy proposals (and some that are totally outside the realm of political possibility). One thing that is cool about the list, particularly for those less familiar with these policies, is that you can easily see the relative contribution of each policy to the deficit.

I just wanted to make sure econ teachers are aware of two other budget simulations, both of which are updated regularly so are likely more useful for teachers than the NYT site (though at the moment, neither is quite as timely as the NYT policy options). My favorite is Budget Hero from American Public Media, partly because the graphics are fun :-) but also because it starts out by asking you to specifically decide on general policy priorities before you get into spending and tax decisions. You select three 'badges' (e.g., safety net, national security, environment, etc.) and then as you make specific spending and tax decisions, the simulation keeps track of how well you are meeting those priorities. When I have my students do this, we discuss how their specific budget decisions line up (or don't) with their stated priorities; students often decide that they are willing to make trade-offs they would not have otherwise thought to make, once they find out the specific dollar amounts attached to their priorities.

Budget Explorer starts out by having you estimate how much of the budget is spent on various categories, a good exercise for students who likely have no clue how much of the budget is devoted to, say, the military versus interest on the debt. Then you can change the budget by increasing/decreasing spending for specific departments and agencies (for example, Social Security Administration), and increasing/decreasing revenues from different types of taxes (i.e., income, social insurance, excise, etc.). Similar to the NYT graphic, you can see at a glance the relative contribution of various factors to the budget; however, while it is interesting to see the complete list of governmental agencies, there is no detail about them (clicking on the agency name takes you to the agency website), and you don't have to make choices about HOW to reduce spending or increase taxes.

Friday, November 5, 2010

Texting in responses to open-ended questions

I've been using clickers for several semesters now and I can't imagine teaching without them. But one drawback has always been that I can only ask multiple-choice questions. When I teach 500 students, I don't see any other option (at least for things where students' answers will count toward their grade in some way). But this semester, I have had a few application exercises in the data class where I wanted groups to come up with short responses to open-ended questions. I have thirteen teams in one section and ten in the other so grading their responses is not a big deal but I had to figure out how to collect them. One of the tenets of TBL applications is that teams should report simultaneously - easy enough with responses to multiple-choice questions (either with clickers or cards) but more difficult with longer responses.

I ended up using Poll Everywhere, a very cool site that allows anyone to create a multiple-choice or open-ended poll and people can respond via text, Twitter or website. I had asked students at the beginning of the semester whether they had laptops and/or smartphones that they were willing and able to bring to class so I know that each team has at least one person capable of submitting a response, at least via the website; it turns out every team texted in their responses. I asked for one-sentence responses and then showed those on the computer screen as they came in. It's not quite simultaneous (some teams were a bit slow in getting their responses in) but close. The free version of Poll Eveywhere allows 30 responses per poll so that works for teams, though I couldn't use it for individual responses (but premium versions allow for unlimited responses, as well as registering students so you know who responded with what).

In both sections, I did a trial-run before using it for anything that 'counted'; that is, my first poll question was "What is your team number?", just to make sure everyone understood how it worked, and there were no problems. I was using the free version so I couldn't identify who submitted which answer but I just told the teams to put their team number at the end of their responses. Since I was showing the responses on the projector, I did have one incident where someone tried to be funny (someone sent in a response that said, "Hey girl, this is cool"). I laughed and then gave my "I expect you to act like adults but if you aren't able to do that, I can certainly treat you like children" speech.

Personally, I don't have issues with students using cell phones in class. I give them the same "I will treat you like adults until you prove me wrong" speech at the beginning of the semester and they know that if I hear a ringing phone, I will stop everything until it is answered, and I almost never have phones go off. I'm not sure if I would use something like Poll Everywhere in a summative assessment situation like a test (that might be too much temptation for students to resist cheating) but given the way I typically use clickers, it could work for almost everything else.

However, one of my big concerns with using something like Poll Everywhere in place of clickers entirely is that it does require students to have either a cell phone with a texting plan, or a smartphone. What I've been seeing this semester is that almost every student has a phone with texting; it's certainly not an issue when I'm asking for one response per team of five or six students. But since I'm considering whether to use Poll Everywhere more extensively in the future, perhaps for individual-response questions, I specifically asked my students about their texting capabilities, using a clicker question. In one section, 29 out of 38 (76%) said they have a cell phone with unlimited texting; that was 25 out of 30 (83%) in the other section. 8 of 38 (21%) and 5 of 30 (17%) said they have limited texting (free up to X texts per month); only 1 student, in either section, said they have a phone with a fee per text; and NO students said they do not have a phone with texting capabilities.The question was right at the beginning of class on a day with unusually poor attendance but my guess is that the sample is still pretty representative, at least of my typical econ students.

So it seems like having students use a service where they text in their answers would be costless, on the margin, for the vast majority of students (for the students who have limited texting, if the texts they send for my class crowd out texts they send to their friends, well, I'm not going to worry about those costs). For the students with a fee per text, let's assume that's $0.10 per text and they have to send maybe 75 texts over the semester (roughly 25 class meetings where I use clickers, average of three questions per meeting). So that's only $7.50 (plus tax?), which is way cheaper than buying a clicker (for those who don't already have one). Even if were double that, it's way cheaper than most of the 'lab fees' students pay for many other classes. I should also add that student reaction to Poll Everywhere seemed pretty positive. A number of students asked me about the site itself ('Can just anyone use this?' 'Are you paying for that?'), and they liked that they could give open answers instead of all multiple-choice.

Monday, November 1, 2010

Trade-offs stink

I have 135 students across two sections of my Data Analysis class, and 40 students in my other class, Economics for Teachers. This weekend, I graded a set of short (2-4 page) papers from the Econ for Teachers class and it took me about five hours, total, spread over a couple days - I have a basic rubric and I do not make very extensive comments because I feel like I simply don't have time, plus the papers were fairly straightforward and the writing was generally fine. Tomorrow, I will get a bunch of 3-4 page papers from my data students and I expect those to be much more difficult to grade, both because the content is more complex and the students generally don't know how to write like economists. So I'm sitting here doing the math and anticipating it will take me somewhere on the order of twenty hours (that's hopefully). And because of the other stuff I need to get done (mostly class prep), and the fact that I just can't grade for more than a few hours at a time, realistically, I know I should tell the students not to expect the papers to be graded for at least two weeks.

But the part of me that always wants to be Super-Prof is appalled at that. Aren't we supposed to give students useful and timely feedback? Two weeks doesn't seem very 'timely' to me. So I think, "Well, I could just hunker down over the weekend and try to knock out most of the grading so I can at least get them done within one week". But then the part of me that is desperate for better life balance responds, "Yeah, but what about L's birthday? And J is traveling for work all week so what about spending time with him when he gets back? And what about simply not wanting to kill yourself?"

And then there's the fact that I'm also trying to work out the Final Project for this class, which will be turned in at the end of the semester. Given that the class is supposed to be all about critical thinking and not just plug-and-chug statistics, I decided not to give a final exam but to have students do a data project instead; however, I haven't actually written up the assignment yet. So I'm trying to do that now and realizing that if I have the students do what I had originally wanted them to do, the result will have to be papers of at least 8-10 pages and I will be grading until New Year's. So I find myself wondering: what is the minimum I can have them do without feeling like I'm compromising their education too much? Or to put it the other way around, what is the most I can have them do without killing myself?

This isn't necessarily meant as an indictment of my teaching load - I realize that there are instructors out there who teach more sections and have more students, and are able to accomplish more with their students (right? You do exist, don't you?). I just know that given what I am expected to do for the rest of my job (i.e., research and service), and what I demand for my own happiness (i.e., at least a few hours a week when I can be NOT working), there are limits to what I can do as a teacher when I have this many students in upper-division courses. The economist in me is realistic enough to understand those trade-offs but the Super-Prof wannabe in me feels incredibly guilty. I know there's no such thing as a free lunch but why does it feel like either I or my students have to starve?