Gas station without pumps

2013 September 20

Some MOOC retention rates amazingly low

Filed under: Uncategorized — gasstationwithoutpumps @ 23:42
Tags: , ,

Mark Guzdial, in his post Lessons Learned From First Year College MOOCs at Georgia Tech (and SJSU), points out the astonishingly low completion rates for some MOOCs (about 1% of registered students, 2–6% of those who did the first assignment):

Karen Head has finished her series on how well the freshman-composition course fared, published in The Chronicle. The stats were disappointing—only about 238 of the approximately 15K students who did the first homework finished the course. That’s even less than the ~10% we saw completing other MOOCs.

Georgia Tech also received funding from the Gates Foundation to trial a MOOC approach to a first year of college physics course. I met with Mike Schatz last Friday to talk about his course. The results were pretty similar: 20K students signed up, 3K students completed the first assignment, and only 170 finished.

Note that the composition course had only 3 major assignments, one of which was to produce a video, so it probably did not do a lot for the writing skills of even the students who completed it, compared to a normal freshman composition class with feedback from experienced instructors.

I can’t help thinking that for the cost of creating the MOOCs, a lot more students could have completed conventionally taught courses. After all, 238 students is only 10 sections of normal freshman comp, at about $6k per section or $60k, and 170 physics students is one lecturer and 6 discussion sections, or about $50k. (And a real physics class would have had an associated lab.)

I’m sure that the MOOCs cost a lot more than $60k—or they wouldn’t have needed Gates Foundation grants to run them.  If the courses had really managed to get a significant number of students to finish, they might have been worth the huge price tag (despite the arguably lower quality of the education).  But having fewer completions than higher quality, cheaper traditional education does not make for a convincing argument in favor of MOOCs.

2013 September 17

Broken soldering iron

Filed under: Circuits course — gasstationwithoutpumps @ 11:51

I’ve previously recommended that students get a cheap soldering station like the one I have, and even recommended that the School of Engineering buy a dozen or so for use in the applied circuits lab.

My son recently found out why they are cheap: the ferrule that holds the tip in is not firmly mounted—it just has a friction fit, and after a while it comes loose and the tip falls out:

The soldering iron after the tip has come out.

The soldering iron after the tip has come out.


A closeup of the ferrule and handle. Pushing it back in and recrimping the tube to hold it tighter seems to have no effect.  One of the reviewers on Amazon recommended supergluing the ferrule in (that’s what they did when theirs failed).

It looks like I’ll be buying a new soldering iron soon. I’m undecided between getting a hot-air rework station with a soldering iron, or separate tools for the hot-air rework and for regular soldering. A combined tool is cheaper and takes less bench space, but I don’t often need the hot air, so a smaller soldering iron would be more convenient most of the time. Also, the cheap hot-air rework stations that include a vacuum pickup tool don’t usually have a soldering iron as well, though for $160 I can get an Aoyue 968A+ that does.

If I get a new soldering iron, do I get another cheap one ($25) and regard it as disposable, or do I get a high-quality digital Weller iron  for $145, or the intermediate Weller analog unit for $90?

2013 September 12

Olin College of Engineering tour

Filed under: home school — gasstationwithoutpumps @ 18:56
Tags: , ,

Today (Thursday 12 Sept 2013) we toured the Franklin W. Olin College of Engineering.

Olin is a very new school, with the first class of freshmen entering in Fall 2002. It is still a very small school with only 350 students—smaller than most high schools.  The college made a big splash in the engineering world when it started, because of very good marketing and because of the hands-on approach they are taking to teaching engineering.

The hands-on, project-based approach was very appealing to my son, who has learned most of his science and engineering (particularly CS) that way.  He’s done some group projects, but mostly he’s been working on his own or been the lead engineer on the projects he’s been part of, so he liked the idea of being part of a team with other good engineering students.

We started out the day walking around campus—literally.  The campus is small enough that you can circumnavigate all the buildings in about 10 minutes.  (There is a large wooded area that is somewhat swampy that we did not visit.)  The Olin campus is small, but the student population is so small that the campus felt deserted—even more so than the Caltech campus did.

After checking in at the admissions office, we went to sit in on the one computer class he could attend today: ENGR 3410, Computer Architecture (known to the students as CompArch).  The “3” digit at the beginning indicates that this is considered a junior-level course, though the equivalent at other engineering schools would be a sophomore-level course.  The difference in levels is partly from the required design courses taken by all freshmen and sophomores, which delay more field-specific courses by about a year.  The class we saw was well taught—the teacher has 2 sections of the course with about 20 in each, which allowed a fair amount of student activity in the class (mainly board work in pairs or larger groups interspersed with short lectures to prepare students for the next round of board work).  The material was a bit dull (number systems), but appropriate for early in the semester of a computer architecture class. Forty students taking CompArch means that about half the Olin students end up taking this course.

There were two other computer courses today (ENGR 3520,Foundations of Computer Science and ENGR 3599 Special Topics in Computing: Computer Networks), but they conflicted with the info session and tour that he had signed up for.  There are only 4 computer courses in total this semester—not a lot of variety to choose from.

We had lunch at the Olin dining hall, which was mediocre dining-hall fare by Sodexo.  (UCSC used to contract with Sodexo, but the quality was so low that students got the university to terminate the contract—the dining hall food quality rose substantially after UCSC started running the dining halls themselves.)  Because Olin is about 1.5 miles from any restaurants, the staff and faculty each lunch in the dining hall also. This situation is different from Harvey Mudd College, where the faculty and staff eat in the dining hall despite nearby restaurants—at Olin there is little alternative.  I’ve been told that some students go to Babson next door for a change (they can use the Babson dining halls on the same meal plan), but supposedly the dining halls at Babson have even worse food.  Almost everyone at the Olin dining hall was eating with others—like at Harvey Mudd, they seem to have done a good job of getting students to eat together.

After lunch we had an information session and tour, each lasting an hour.  We were the only family in either one—the smallest tour we’ve had yet.  The tour guide gave us a lot of information, but he did seem to repeat himself a fair amount.  It must be difficult giving a one-hour tour of such a tiny campus.  We did not see a dorm room, but we did see a dorm kitchen, which is exactly the same size as one of the dorm rooms.  As might be expected from such a recently built campus, the dorm rooms are fairly large and all the facilities in excellent condition.  Olin makes a big deal out of student involvement in improving the campus, so a lot of the things we were shown were prefaced by remarks about how students had decided they wanted x, and got the funding from the college to do it themselves.  That’s a nice approach—particularly on a brand-new campus, but I don’t know how long the college will have the space and the funding to support student initiatives. I don’t think that they’ll run short in the next four years though.

Olin seems to be preparing engineers primarily for industry—only about 35% go on to grad school.  One of the bigger recruiters on campus is Microsoft, and students are provided with a Windows laptop when they first register on campus, loaded with lots of software useful to mechanical engineers (but not much for CS students).  Based on what we saw of the design labs and project labs, Olin looks like a great place to do undergrad studies in mechanical engineering and materials science, but not so great for computer science and computer engineering.  They had just enough computer engineering to be able to do robotics, but not much depth past that.

We had dinner in the dining hall with a couple of Olin students, including one from FWOP (Franklin W. Olin Players), the main theater club on campus.  They do a couple of plays a year, plus a few shorter one-act productions.  Students can also do theater at Babson and Wellesley colleges—male actors are particularly in demand for Wellesley productions, because Wellesley College is still an all-female college.  Overall, it sounded like there was plenty of opportunity for acting at Olin—not as many options as at Brown, but fewer students competing for the parts available.

The big plus for Olin, project-based curriculum, was offset by the small, isolated campus and lack of depth in computer science. Overall, my son rated Olin fairly low on his list, along with Caltech.  Both would be great schools for some students, and he could do reasonably well at either one, but they did not seem like as good a fit as schools higher on his list.

Update 2013 Sept 14: I forgot to mention of Olin’s big pluses: the final admissions decisions are made by faculty, not non-academic staff.  This means that applications to Olin can include technical material that most admissions officers would just be mystified by.

This is probably the last college tour we’ll do this fall. He may apply to some colleges that we haven’t toured yet, but visit only if he is accepted at them.

2013 September 11

Not just a job ticket

Filed under: Uncategorized — gasstationwithoutpumps @ 08:31
Tags: , ,

On my MIT tour post, Nita commented

As an engineer myself, I find it rather interesting that people automatically think that going to MIT will secure a certain type of position or opportunity. So not true for the field. There is such a need for developers that many of them don’t even have 4 year degrees to be viable in this field. Just getting a Certification in Oracle, Cisco and more will get you in the door to a ‘Help Desk’ or ‘Jr. Programming’ job without a degree and if you are good, your salary and the salary of your peers will be the same in about 5 to 10 years … all without a degree. Now if you want to become a Software Manager or Engineering Architect, the highest degree most have is a Masters—but it’s not necessary, what they want you to have is another ‘Certification’ a PMP which takes about a week to get. Really spend time talking to people in your child’s chosen career makes a big difference.

I started to respond to the comment, but as my response got longer and longer, I decided to turn it into a separate post.

While it is certainly possible to get a job in IT without a degree, I think my son would not be happy in a “Help Desk” or “Jr. Programming” job for very long (by which I mean more than a few weeks). He is patient with people who don’t know what he knows, and (I’m told) he’s been a good teaching assistant for the home-school Python class, but I think he’d get frustrated dealing with the rude behavior by stupid people that most help desk personnel have to put up with. A junior programming job would not be very rewarding either, as people in such positions are not trusted to make any decisions about the code, but just to implement what they are told or slap patches on badly written code that is already so patched by incompetents that it is nearly impossible to maintain. The most tedious and unrewarding parts of programming are given to junior programmers, with almost no opportunity given to show what one is really capable of.

His goal in going to college is not “to get a good-paying job”, but to learn cool stuff and to do cool stuff (which for him, at the moment, mainly means writing programs, though he’s gotten a bit interested in designing digital hardware as well).  It is unfortunate that computer science has recently been oversold as a job ticket, because it means that in most colleges he’ll be surrounded by people who are just there for the money and have little love for the learning. Computer programming suffers from even wilder boom-and-bust swings than other engineering fields, and I suspect that a lot of the current pushing of CS as a ticket to a good job is deliberate hype to keep labor costs low.

Part of what we’re looking for on our college tours is a place where many of the students are there for the learning, not the job ticket. Of course, since most parents and students have been taught to think of college as a job ticket, the information sessions and campus tour guides often spend a fair amount of time talking about how good the job prospects are for their graduates.  Both Brown and Stanford made a big deal out of Google’s active recruitment of their students, though Stanford included Google only as one of the many examples of employers who pay big bucks to recruit Stanford students.

We’ve been trying to read between the lines to find the places where he’d get the learning he wants and where he’d be surrounded by other students with similar motivations.  A high rate of students going on to grad school in CS and doing well is a good sign, for example, because it means that the students loved learning enough to forego high-paying jobs in order to learn more.  For students primarily interested in getting one of those high-paying jobs, the number of people getting job offers immediately after graduation and the size of those job offers may be more important than the number going on to grad school. Google recruiting is a good sign for both groups of students—Google pays well, but they are also interested in finding people who can do new things, not people who’ve just gone through cram-and-forget training to get a piece of paper with BS on it. Having a large number of successful startups formed by students or recent alumni is also a good sign for both groups, at least if the startups are doing cool new things, and not just random tweaks to whatever the current fad is.

Getting paid well to do cool stuff would be nice, but my son is not primarily driven by money.  His motivations might be different if we were rich and spent all our time comparing ourselves to still richer people, or if we were poor and spent all our time trying to make ends meet and pay off crushing debts.  But our family is solidly middle-class, with a paid-up mortgage and just enough income to indulge our cheap tastes and save for retirement and college. Our biggest expenses are food and education.

My son’s future is undetermined: It’s possible that he’ll get rich from creating a startup that hits the big time.  It’s possible that he’ll become a hot-shot programmer or engineer at a large company. It’s possible that he’ll end up as a professor or as a senior researcher in a national lab.  It’s also possible that he’ll end up as an actor, doing various contract programming and web-design jobs as day jobs to pay the rent between acting gigs.  He’s preparing himself for any of those possibilities (and probably others).

Right now, being a student is most attractive to him, and he is trying to optimize that experience.  If learning continues to be his main passion, he’ll probably stay in school through a PhD.  If something else beckons along the way, he might stop (or perhaps pause) with a BS and pursue the other opportunity.

2013 September 10

MIT tour

Filed under: home school — gasstationwithoutpumps @ 22:53
Tags: , , ,

Tuesday, we toured MIT, a school I’ve always been impressed by, both for the reputed intensity of its students and the research that comes out of MIT.

Before the tour we attended an information session in room 35-225.  At MIT, everything is given a number, rather than a name (majors, courses, and buildings)—it is part of how MIT creates a culture of insiders and outsiders.  So 35-255 means room 255 in building 35 (we started in the Admissions office, better known as 10-100).  Room 35-255 turned out to be a very run-down lecture hall with paint peeling off the ceiling, and an impressive tangle of power cords and data cables hanging off the back of the podium (including parts that were clearly intended to be fastened to a solid surface, not hanging by a wire).  The lecture hall has a very steep rake, so that everyone can see over the person in front of them, but even from the second row, the speaker looked like he was a long way away, at the bottom of a deep well.  There was a class scheduled in the room right after our information session, so I know that this wasn’t a mothballed room pulled into service for information sessions, but an active classroom.

One thing that MIT did that no other college has done so far was to provide a customized welcome letter for my son, providing information about the three subjects that he had expressed the most interest in when he signed up for the info session.  It was a simple template-based letter, just putting in three paragraphs from the form he filled out, but no other campus has shown that level of care in the recruiting.  This gesture almost made up for the very rundown teaching facilities—but the admissions process doesn’t last long, and the teaching facilities have to be lived with for 4 years.

Another handout they gave us was a “Facts for Freshmen” data sheet, and this one undid the good impression of the customized welcome letter. They used a number of tricks to disguise and distort data:

  • putting 4,384 undergraduate students in a huge font and 6,510 graduate students in a tiny font to make it look like the undergraduates dominate the graduates.
  • having the “area” mistake, where triangles were used to show percentages of students from each geographic area, but the heights of the triangles were proportional to the percentage, rather than the areas being proportional to percentage.  I guess their graphic designer never read any of Tufte’s books.  Oh, wait—Tufte is from Yale, and maybe people from MIT don’t read stuff from Yale.
  • total financial aid is reported as $31,232 per year, but they don’t say on this sheet that loans are included in that figure.  If you take the total scholarships ($8.9M) and divide by the number of students (4,384), you get an average scholarship of $20k—a lot less than the $31k they misleadingly report.

The MIT information session was reasonably informative, though the speaker (an admissions officer) spent far too much time talking about athletics at MIT.  One positive point was the amount of time he talked about the Undergraduate Research Opportunity Program (UROP), which students can participate in from the beginning of their MIT careers.  About 85% of students participate and most are paid for their research work (it wasn’t clear whether students coming up with their own projects got paid, or only those working on funded research for faculty).  A lot of really cool research is done at MIT, and being part of one of the better projects may be a wonderful experience.  There is certainly a strong expectation that students will do research and hands-on projects, and not just book learning, matching what is claimed in the MIT motto “Mens et Manus” (Mind and Hand).

At the end of the session I asked one rather pointed, even rude, question: “I’ve heard a lot about how stressful MIT can be—how does the suicide rate compare with other selective colleges?”  I had a personal reason for concern about this—my best friend from high school committed suicide at MIT in the spring of his sophomore year.  I also have seen some figures suggesting that MIT had a much higher suicide rate than other schools, though those are old figures for 1990–2001, and may no longer be representative.  The answer I got was not very reassuring—they have a lot of support mechanisms in place (though the only ones mentioned were studying together and support services that could be called) and their rate is not really higher than comparable institutions (a claim I’ve not been able to justify or refute online, though what little I’ve been able to find does suggest that the rate has dropped to a more normal 7/100,000/year).  I would have been more reassured if they had said “we had a problem in 2000 and 2001, and took a lot of steps to address it—since then our rate has been x, comparable to a rate of y at other selective schools”.  They are still saying  to the press that male engineering students are more likely to commit suicide than other students (ref), which is not reassuring to the parent of a male student likely to be studying engineering.

Our MIT tour guide was reasonably well-informed and audible, giving us a few factoids we did not previously know (how much the copper roof of Kresge Auditorium weighs and how the car-on-the-roof hack was managed), but we spent far too much time in the Zesiger Sports and Fitness Center talking about athletics.  He had some interesting information about the buildings, including the incompetent design of the Green Building by I.M. Pei.  He talked about the good acoustics of the Chapel, but not the abysmal acoustics of the Kresge Auditorium.  He did not, however, talk much about academics at MIT, other than a couple of pro forma speeches about the gen-ed requirements, and a fairly good description of UROP research opportunities.

I looked at the CS requirements for the 6-3 program and combined gen-ed/CS requirements, and it looks like a fairly rigid, formulaic program, with very few electives until senior year (choice of 3 core courses—out of the 3 offered, choice of 3 header courses—out of the 3 offered, and so forth).  The MIT program seems to be guaranteeing quality by making sure that everyone has taken the same courses and not missed anything, with just a little customization in the senior year—rather the opposite of Brown University’s philosophy of every student creating their own unique path.

The population density of MIT seemed reasonable when the sun was shining, but when it started sprinkling, everyone headed inside and the corridors became packed with students and faculty trying to get from one building to another without going outside. It was like the mad scramble to get from one class to another in a big mid-western high school, but with clear rules (explained to us clueless tourists many times) about staying to the right in all corridors. My son has not experienced crowded corridors much before (most California classroom doors open directly to the outside), so the little sprinkle of rain we got today was good for giving him a more representative view of MIT.

At lunch in the CS building (32G), I noticed that a lot of people were eating in groups, but a lot of people were sitting alone, and that no one approached them, even when there were no empty tables left.  It might be hard for a shy person to make friends at MIT, given the general unwillingness to disturb anyone who seemed to be by themselves.

Overall, my son had a reasonably favorable impression of MIT, putting it in an equivalence class with CMU and UCB.  All three are research powerhouses that are somewhat dominated by the grad programs.  Personally, I thought that of the three, MIT did the most to include undergrads in research and UCB did the least, but that the UCB campus was the nicest environment and MIT the least pleasant.  He put all three lower on his list than Harvey Mudd, Stanford, and Brown, though he has not yet articulated exactly why he ranks the schools the way he does.

« Previous PageNext Page »

%d bloggers like this: