Wednesday, December 29, 2010

Winter prep time

Unbalanced Reaction posted an interesting thought about when to refocus on work over the holidays -- on the 28th of December. I left a quick, off-the-cuff, answer that I work on Wednesday (which happened to be the 22nd and 29th this year), but that was before I settled down to work today and started noticing the time stamps on various files that needed attention.

You guessed it: Lots of them were dated around the 28th, regardless of the day of the week over the past five years or so! Fascinating.

There were quite a few others dated around the 20th (circa the Winter Solstice as I mentioned in my comment). Those were usually the syllabus, which I got done quite a bit earlier this semester that in the past. In fact, I noticed that I was about a week early on quite a few things this year, things that would normally get done after Christmas around the 28th.

I never noticed that work pattern until you brought it up, UR. Great insight. We both work in that natural gap between Christmas and New Years (what feels like a Wednesday), although I clearly also like to work about 5 days before Christmas.

So, no, Unbalanced Reaction, I'm not generally the model of efficiency, but this year has been an unusually good one for keeping focus on priorities by doing them as soon as I could rather than when they absolutely had to get done. I don't know if I was more stunned that I thought of working on the massive revision of my teaching plan while proctoring, or that I actually did it.

Well, I worked on work things, at least. Christmas lights didn't get put up nearly as soon as they should have.

And blogging. You should see the queue of semi-written articles. I think they will all get finished up tomorrow, but at least this one is going out only a day late.


Read Entire Article......

Wednesday, December 15, 2010

Productivity

My blog friend Dean Dad has a couple of bees under his bonnet that rival those of much older, semi-senile faculty. One of them is his obsession with the Credit Hour, even when (at other times) he worries about articulation and the transfer of course "credit" from one institution to another. He reminds me of one college that got rid of grades to foster creative risk taking, only to discover that no one wanted to hire their students because it took too much effort to evaluate the individual portfolios. That college no longer exists.

His latest version of the argument, in a blog last month, is to blame it for a lack of productivity in academia:

Third, we've defined what we do in a way that defeats productivity improvements. We measure learning in units of time. Until we stop doing that, no amount of efficiency-tinkering will make enough of a difference. A three-credit class required forty-five hours of seat time thirty years ago; it still does.

Actually, it doesn't. We only require 42 or 43 hours of seat time (plus final exams) at our college. They have to learn the same physics content in 28 weeks of class time that they once had 30 weeks to learn. You see, we measure learning in units of chapters in physics books, and engineering schools expect the same prerequisite knowledge they always did. AFAICT, everyone deals with this by cutting back somewhat on topics that students never learned anyway, but that does not help the overall learning cycle for the core material in the course.

Side comment: Somehow the same has not happened in math. Calculus takes longer now than it did when I first encountered teaching it, although that change might be making up for the lost weeks I mentioned above.

Besides, this is all a red herring. Productivity is not what one student learns, it is the cost of producing that learning. Productivity is about the difference between one professor running a tutorial for a single student and having an 80 student lecture/discussion class. And there, I know my productivity has increased significantly in just the past decade because my annual enrollment has grown by more than 50%. That means a lot more money is paying for my time, which is all that matters for the college's bottom line.

Side comment: An assistant chairman in my distant past also argued that failing students was a way to increase productivity in the department. It enabled them to squeeze twice as much money for the same amount of learning. This doesn't always work, of course, because repeats can displace other students who might be more likely to learn the material and pass the class.

And my productivity has increased because, in the days of smaller enrollment, I was also teaching labs. That is three hours of my time that only generates one credits worth of income from a small number of students rather than three credits worth for twice as many students -- a factor of SIX in income for my time! Using an adjunct instead of a tenured professor has lowered our cost there while freeing me to generate more income for the college.

One complaint I don't understand is
Fourth, unlike almost every other sector except health care, we have to invest in technology even when it doesn’t improve our own productivity.

No, you don't. That is a cop out. Managers like yourself did not have to replace blackboards in every classroom with SmartBoards and projectors without doing any study to see if they improved learning. (By the way, that is not a one-time capital expense. Projector bulbs are expensive and projectors wear out. There is also more security required because of a significant theft problem.) Indeed, they didn't even do a study to see if increased use of Powerpoint might reduce learning!

He also blames tenure, although he is actually blaming a seniority-based pay system rather than tenure. You can have tenure without automatic pay increases and you might need step pay increases without tenure, to keep your best people. Besides, as I alluded to above, one way every college has increased productivity is the use of contingent faculty, particularly at universities where the benefits are the greatest. I say this because the fraction of classes taught by adjuncts at my college has been stable for a long time at about 50%. (I am counting classes rather than people for a good reason: we have a significant number of adjuncts who only want to teach one or two classes.) I think this is possible because the salary disparity is not as great as at universities.

And productivity gets harder to define when you shift from a Community College environment (where he and I work on the teaching side) to a Research University environment (where I used to work on the research side). Is a professor's time better spent in the classroom generating credit hours or in the lab generating grants with overhead and jobs for students that help support enrollment? I think we all know that the answer to the last question is "yes" at an R1 institution, where it even includes the creation of non-teaching faculty positions that exist solely to bring in additional contract dollars.

And that last detail is why I think, in another article from last month, Dean Dad completely misses the point made by Historiann. Historiann is at a Wannabe Major University, just the sort of place where managers do profit (pay increases and job jumps up the ladder) by shifting resources to areas where they are more likely to get more research grants that generate more "overhead" (indirect cost recovery) and more administrative positions. There isn't much (make that ANY) value to the university if Historiann publishes another book. There is a lot of value in the 40% that gets siphoned off of a grant, and even more if the professor's salary and benefits and all other expenses (office, light, heat, staff support) can get charged to the grant while an adjunct with no benefits and few of those expenses teaches hir class that semester.

I have little doubt that what I just wrote is "far removed from any reality I [Dean Dad] can recognize", but it is a reality I am very familiar with.

But I would not put all of the blame on the managers who made it happen. Many faculty are complicit in the expansion of the university research enterprise because their lives are devoted to research and graduate and post-doctoral education. It is an unfortunate reality that history cannot compete with biochemistry at this game, and tight budgets will push money to where it creates the most return for the people managing the university.


Read Entire Article......

Saturday, December 11, 2010

Students helping students

Dot Physics offers an excellent suggestion for communicating good study habits to new students: have this year's students tell next year's students what they need to do. Better yet, he posted what his student's wrote.

I know it is a good idea because I have been doing something similar for several years. It started with an end-of-semester question about prerequisites. (If you follow my blog, you know that I have identified the failure of students to comprehend the meaning of "prerequisite" as a long-standing problem for advanced classes like calculus and physics.) That info got shared with the math faculty who taught those classes, and I think I have seen an impact on what students learned and brought with them to physics (and calculus).

Later, based on discussions with a colleague at another school, I tried something similar to what Rhett Allain is trying: collect advice from current students (mine is entirely anonymous) and distribute it via Blackboard to the incoming class. (That mechanism is still a weakness because our students don't get access to their Blackboard shell until the first day of classes. They really need this kind of info before then.) They seem to appreciate it, but I'm less certain how much it helps.

Measuring changes in the success rate of any class is tricky. There are lots of variables. (To name just one that could be measured, I started getting a significant number of kids with AP calculus experience after the depression of 2008 hit.) The biggest is that they are busy (work or play or both) and sometimes just lazy and unprofessional, still looking for the easiest way to pass. Thus, even though my students, like Rhett's, offer the excellent advice to read the textbook before class and start their homework early, they just won't do it no matter who tells them about it. But they will form study groups, and that sort of collaborative learning has grown significantly in the last few years.

Has anyone else tried this? Any suggestions on how to get them to read?


Read Entire Article......

Tuesday, December 7, 2010

Dot Physics gets Power wrong

Rhett Allain has a great blog at Wired.com, but I will not create yet another account just to comment there so my comments will be here instead.

Usually he is on the money, but in this case his bad experiences with ESPN Sport "Science" gets in the way of his analysis of a video about the power of NASCAR cars. For convenience or future reference, I'll embed the video here



and then get to the analysis.

Rhett first objects to the statement that the weightlifter being shown "exerts about 1 hp per rep". Yes, they meant "during", but what is wrong with that? The numbers are right if you take the 275 pound lift as being 2 feet (61 cm) rather than 50 cm (about 20 inches) in 1 second. The weightlifter is producing pulsed power during the lift, which is about half of that 1 second rep, but hardly resting during the other half.

Check out this video of a power lifter doing 26 reps on the NFL 225 pound lift, or this one where the guy does 72 reps at 225 pounds. The first one takes around one second per repetition, locking the arms out each time. The second one is a much shorter, but faster, lift that might make for an interesting video analysis to see what his power output is.

Estimating the average power is trickier, because you really can't use the work done ON the weights as your metric. If you did, the average power would be zero because there is negative work done on the weights as you lower them! !! However, if you shift your focus to the work done by (within) the muscle, it might be more than 1 hp for the entire time the weight is moving. Controlling a weight as it comes down is not quite as hard as lifting it, but it isn't being done for free!

What I like best about this example is that "power lifting" is one of the few cases where a physics term is used correctly in sports. Power lifting, where the emphasis is on multiple reps, is entirely about the rate of doing work in a way that reflects what is done in competitive athletics rather than just lifting the most weight. That is why the NFL tests on the number of reps of 225 pounds. (The NFL record is supposedly 43.) Yes, that is power. And I think it is more obviously power than the similar output required to climb a mountain on a bicycle even though that is probably the most extreme case of continuous power output by humans.

Rhett next complains about a statement that he actually misinterprets. The statement in the video (around 0:50) is that horsepower of a car engine is "calculated by measuring torque". This is 100% correct. Rhett says "First, horsepower is not measured by calculating torque (at least not in physics)." Right but not relevant, because they don't calculate torque. They measure torque and rpm and calculate power by multiplying the two together. Rhett says "I guess the only problem here is using “fast” to describe the relationship between torque and power." except that is not what they are doing. They are using fast to describe the angular velocity, just as you might use "fast" to describe the linear velocity if you said that power was about how fast you can apply a force (Power = force * velocity). This is 100% good physics. Rhett, you messed up this time.

For the record, in physics and engineering and the real world of dynamometers, you determine horsepower by measuring a torque curve (torque in foot-pounds as a function of angular velocity in rpm) with a load cell (which measures force) on the end of a lever that is connected to the load on the engine. Modern ones do the multiplication and plot both power and torque versus rpm, but the actual measurement is torque (or, if you like nits, force that gets autoscaled into torque on the graphical output).

I'll go along with the final nitpick about lifting the space shuttle. Yes, they should have included "in one second" at the end of that last sentence. 850 hp is, indeed, like bench pressing the space shuttle in one second. Time is important. But no one would confuse using a jack (in his video example) with "benching". Everyone knows that you bench press a weight in less than a second unless you are totally whipped, so the same would apply to benching the space shuttle.

My negative nit pick: The video correctly describes the historical origin of horsepower as a marketing term, but the draft horses shown in the video (e.g. at about 0:30) produce more than 1 hp. James Watt used the small horses used in mines as his reference point for selling his steam engines.

My other negative nit pick is that engine size is not nearly as important as the rate of fuel consumption. After all, a top fuel dragster only needs about 550 cu.in. (compared to 358 cu.in in NASCAR) to make over 8000 hp (rather than 850 hp). It is all about the fuel and the rate you can burn it -- and how long the engine lasts! You have to put power in to get power out. Like the co-host commented during his 259 mph test drive of the Bugatti Veyron Super Sport on "Top Gear", before the pro took it to 267, you can actually see the gas gauge moving when you are burning 1.7 gallons per minute pushing out about 1200 hp. Wide Open Throttle is like that. (I'll have to save for another day the effort to figure out the Reynolds number comparison between air and treacle they used. I like it, but I'm not buying it.)

But I also have a positive nit pick. I loved their description of the added power from opening up the exhaust, although they oversimplified it a lot. Part of it is to "tune" the exhaust so it resonates at a frequency that matches the rate at which you want to pull exhaust out of the cylinder. Back pressure from the exhaust makes the engine less efficient. Getting a rarefaction as the exhaust valve opens is ideal.

However, the distinctive engine sound they played comes more from the Doppler effect than the resonating pipes. You need to stand next to one to appreciate that.

PS - The best thing about the Top Gear Bugatti video is you can actually see the exponential approach to terminal velocity as the spinning of the digital speedo slows down.

PPS - This was started ages ago, but only finished up and posted at the end of December. I'll try to monitor comments to be sure they don't sit too long in the moderation queue.


Read Entire Article......