Monday, November 7, 2011

Why do college students defect from STEM

In this weekend's New York Times Education Supplement, author Christopher Drew asks "why science majors change their mind," and why "roughly 40 percent of students planning engineering and science majors end up switching to other subjects or failing to get any degree." His short answer, "it's just so darn hard," rings true. So does his observation about "the proliferation of grade inflation in the humanities and social sciences, which provides another incentive for students to leave STEM majors."

It's in his investigation of why science is "just so hard," and what to do about it, that Drew falls short. His focus is a single data point, Notre Dame student Matthew Moniz:

He had been the kind of recruit most engineering departments dream about. He had scored an 800 in math on the SAT and in the 700s in both reading and writing. He also had taken Calculus BC and five other Advanced Placement courses at a prep school in Washington, D.C., and had long planned to major in engineering.

But as Mr. Moniz sat in his mechanics class in 2009, he realized he had already had enough. “I was trying to memorize equations, and engineering’s all about the application, which they really didn’t teach too well,” he says. “It was just like, ‘Do these practice problems, then you’re on your own.’ ” And as he looked ahead at the curriculum, he did not see much relief on the horizon.

So Mr. Moniz, a 21-year-old who likes poetry and had enjoyed introductory psychology, switched to a double major in psychology and English, where the classes are “a lot more discussion based.” He will graduate in May and plans to be a clinical psychologist. Of his four freshman buddies at Notre Dame, one switched to business, another to music. One of the two who is still in engineering plans to work in finance after graduation.

Mr. Moniz’s experience illustrates how some of the best-prepared students find engineering education too narrow and lacking the passion of other fields. They also see easier ways to make money.
From this one data point, the solution emerges naturally. It is-- you guessed it--a greater emphasis on relevance, aspiration, leadership, and student-centered project-based learning. At Notre Dame, for example,
Dr. Kilpatrick [the Dean] has revamped and expanded a freshman design course that had gotten “a little bit stale.” The students now do four projects. They build Lego robots and design bridges capable of carrying heavy loads at minimal cost. They also create electronic circuit boards and dream up a project of their own.

“They learn how to work with their hands, how to program the robot and how to work with design constraints,” he says. But he also says it’s inevitable that students will be lost. Some new students do not have a good feel for how deeply technical engineering is. Other bright students may have breezed through high school without developing disciplined habits. By contrast, students in China and India focus relentlessly on math and science from an early age.
Drew also cites Worcester, which
ripped up its traditional curriculum in the 1970s to make room for extensive research, design and social-service projects by juniors and seniors, including many conducted on trips with professors overseas. In 2007, it added optional first-year projects — which a quarter of its freshmen do — focused on world problems like hunger or disease.
Some of this sounds quite reasonable. There's perhaps no subject where project-based-learning is more appropriate than engineering. But how much sense does it make to dilute the course requirements for, say, biology and chemistry with overseas field trips and social service projects when there's so much hard material to cover to prepare students to compete for research & development jobs in STEM?

Besides field trips and social service projects, and also not to be confused with research and development, there's leadership. The University of Illinois, for example:
began this fall to require freshmen engineering students to take a course on aspirations for the profession and encourages them to do a design project or take a leadership seminar.
The underlying assumption is that the problem isn't so much that STEM majors are difficult, but that students find the classes boring and irrelevant to daily life, and that they lack confidence. Says Arthur C. Heinricher, the dean of undergraduate studies at Worcester Polytechnique, in reference to its student-centered projects:
"That kind of early engagement, and letting them see they can work on something that is interesting and important, is a big deal. That hooks students.”
More generally:
...The main goals [at Worcester] are to enable students to work closely with faculty members, build confidence and promote teamwork. Studies have shown that women, in particular, want to see their schoolwork is connected to helping people, and the projects help them feel more comfortable in STEM fields, where men far outnumber women everywhere except in biology.
It doesn't seem to occur to Drew, or to any of his interviewees, to consider a more obvious reason why STEM courses are "so darn hard" for students these days, and why they lack confidence: the decreasingly poor preparation that they are receiving in high school math and science classes. There are many reasons for this, ranging from the No Child Left Behind-inspired dumbing down of the curriculum, to the decline in AP course offerings and ability-based grouping, to the ravages of Reform Math (which begin in elementary school). But among these reasons is precisely the kind of child-centered, project-based learning that Drew and his interviewees are advocating (one need look no further than Philaelphia's project-based Science and Leadership Academy, and its dismal test scores in science). While it often sounds great in theory, project-based learning is an inefficient and disorganized way of learning the core curriculum necessary for college-level science courses and STEM r&d jobs.

Furthermore, if you're ill prepared for college-level math and science classes, and therefore don't understand what's going on in class unless it's hands-on and student-centered,of course non-hands-on courses will seem dry, narrow, borring, and irrelevant to your aspirations.

And of course you will end up dropping out, especially if you face competition from classmates who got their K12 science training overseas, where rigorous math and science classes still abound. Indeed, this explains why, as the article notes, "the attrition rate can be higher at the most selective schools, where...the competition overwhelms even well-qualified students." The most selective schools, after all, attract the highest numbers of better-trained STEM students from overseas.

Ironically, Drew begins his article by alluding to "test scores showing American students falling behind their counterparts in Slovenia and Singapore." Isn't the first step, then, to look inside Slovenian and Singaporean classrooms and see what sorts of curricula and pedagogy these countries are using? I could be wrong, but I'm guessing that one would find many more hours of rigorous study of core content, and many fewer hours of leadership, inspiration, and project-based learning.


gasstationwithoutpumps said...

I had a different critique of the article—he starts from an urban legend. See my response at

John said...

This student-centered touchy-feely not-teaching-anything movement is really starting to feel less like a mistake, and more like giving plants gatorade because we think they crave electrolytes. (ie, this is going into "not even wrong" territory)

Dead plants? More Gatorade!

Katharine Beals said...

Interesting critique, gwp. I'm curious how today's situation compare with a generation ago--i.e., more or fewer STEM majors.

John, I can't help thinking of the movie Idiocracy...

ChemProf said...

I wonder about any calculation that including incoming "premed" majors. In my experience, many of these students aren't actually interested in STEM fields, but were told by parents that being a doctor was a good goal. They tend to fall away more often than students who want to study science or math.

Also, for a contrary view, see

Of course, some of this is an oversupply of bioish majors, where they do seem to have trouble working (and where some students are apparently paying for grad school, which is always a danger sign).

TerriW said...

My local district (which we don't attend, we homeschool) has as their big draw a huge STEM focus (Project Lead the Way, a Fab Lab) ... and they still use Everyday Mathematics.

So. You've got very hands-on-ish engineering lite stuff to make it all seem so fun to gin up interest in the non-hardcore math/science folks coupled with a leaves-something-to-be-desired math curriculum.

What is this supposed to *do* exactly? Say they get some kids to go down the engineering path in college who wouldn't otherwise do so. Can I suppose that they wouldn't otherwise do so because they, perhaps, aren't particularly strong in math? And EM sure isn't giving them the rock solid foundation they need to make it through the wash out courses.

I can't help but think it's a well-intentioned recipe for failure.

ChemProf said...

That fits with what I see a lot, TerriW. We get a lot of students who are excited about science, but when you talk to them about what they like, it is "docent science" like you'd see at the zoo or aquarium. Which is fine, and a great volunteer opportunity, not but a great career path. It seems to be a result of lots of science appreciation but not so much real science (which can be abstract a lot of the time).

Katharine Beals said...

"docent science"--love it!
Has anyone seen the Brian Greene "Fabric of the Cosmos" Nova episode? What I've seen of it strikes me as right up there with the rest of "docent science."

Anonymous said...

Unfortunately, there are many people who are more interested in good intentions than actual results; hence the old saying "The road to Hell is paved with good intentions." They're likely to be the ones uninterested in the unintended consequences of their good intentions.

Anonymous said...

I just read through the Scientific American article and I have read something along these lines before. What's often ignored is that the scientists and engineers we produce often aren't as good as those in other countries. They are simply far less qualified due to a poorer education at the K-12 level.

Most American companies have canceled projects or plans for new divisions due to an inability to find enough qualified people to fill required jobs. If you need 50 scientists with specific skills and you only can find 20, you will either not go ahead with a project or you will go overseas. So, even those 20 qualified people have lost out on a potential job.

We need to produce more highly qualified scientists and engineers precisely so companies can go ahead with new projects and new divisions that will create full employment for STEM graduates. But that isn't going to happen with the continual dumbing down of education.

GPC said...

A couple of problems jumped out at me from the Scientific American article. First the suggestion that American students don't do as badly on the PISA test as we think. There is no doubt that maybe the top 10-15% of American students are getting a world-class education. The point is that this isn't nearly enough. So, how the top 5% compare on PISA isn't really all that important. How the top 40% or 60% compare is what's important. The top 5% to 10% can't provide all of the qualified graduates that the largest economy in the world needs.

"On tests comparing the U.S., Japan and five Western European countries, for example, white Americans on average substantially outscored the Europeans in math and science and came second to the Japanese. American whites came first in reading by a wide margin."

There is far smaller disparity between the performance of privileged and underprivileged students in Western Europe than in America. We may be producing more elites at the very top but Western Europe is producing far larger numbers of educated people overall. This explains why there is less social mobility in America than in Western Europe. Again, this elite cannot provide all the qualified workers that our economy needs. Of course, there also have been studies done that compared higher performing American students to high performing students in other countries. American students did pretty badly. I would be curious to know more about this particular study because it seems to be at odds with many other comparisons.

American companies are panicking about where their future workforces will come from. They wouldn't be doing this if there was an oversupply of qualified candidates coming out of high school and college. I used to interview people for jobs. I met so many young, white, middle class college graduates who had terrible writing and math skills. They lacked the most basic knowledge of their field. I recently met a middle class white college student doing volunteer work at a library book sale. He had to use a calculator to figure out $3.50 from $10 to give me my change. He told me that he is terrible at math. Yes, there are highly educated Americans. But not nearly enough. This is why the company I worked for had such a hard time filling jobs with great pay and benefits (including 4 weeks of vacation because it was a European company).

Secondly, the article focused a lot on the competition for jobs in academia and research. Are these the absolute only places science graduates can work? There are thousands of middle and high schools in this country that badly need people with science degrees to teach. I have read that more than half of American students are taught science by teachers who don't have degrees in the Sciences. Sure, if science graduates are looking for jobs in a limited number of places, there will be an employment problem. Finance graduates would also have an employment problem if they limited their job options to Wall Street firms only.

Anonymous said...

The Global Report Card (GRC) is a project that uses PISA data to determine how individual school districts perform on an international level. The Pelham School District in Massachussetts ranks in the 95th percentile. So, if you do a comparison using Pelham students, they will obviously outperform the competition. The Beverly Hills School District ranks at the 53rd percetile. So, if you use those students as the basis of a comparison, they would underperform.

According to the GRC, of the 50 richest school districts with populations of 50,000 residents (not students), almost half perform at the 50th percentile or below. Newton, Mass comes in at number 1 at the 80th percentile. I suppose you can break PISA results down in different ways. But the GRC does indicate that we have a crisis even with our more privileged students.

I briefly scanned the SA article. It did seem to focus a lot of Ph.Ds. I assume most Science majors don't actually earn Ph.Ds. I remember an article from a few years back. If I am correct, it was the American Academy of Sciences issuing a warning that America will face a serious shortage of scientists when the baby boomers leave the workforce. According to the article, most of this loss would be in public health, like food safety, water safety, etc. Science majors also go into various healthcare fields. Maybe there is an oversupply of Ph.Ds but I think the point about a serious shortage of science teachers is a good one. If we were overproducing science graduates, you would think there would be no shortage of science teachers. But there is. I could be wrong, but doesn't this suggest that science graduates have many other options available to them?

ChemProf said...

I think some of the problem in any of these discussions is grouping everything under "STEM." And some of it is the peculiar world of academia. The best science/math undergrads are often encouraged to go to grad school whether it makes sense or not, and whether it meets their goals or not. I've encouraged some students to consider high school teaching (and there are some tremendous scholarships out there for science folks; google Noyes Scholarship) but I'm in the minority.

Once in grad school, there is a lot of pressure to only consider jobs as research institutions. I entered grad school wanting to teach at a liberal arts college, but I remember applying to a UC for a job and having my advisor write to me "oh good, I'm so pleased to see this application because I want to see you live up to your potential." Thanks, Dad. Jobs at national labs or in industry are seen as inferior.

And we are actually seeing signs of oversupply in some biology fields (and if you see the most plaintive complaints, you'll see they are from bio people). Saw a former student complaining about her tuition costs for her PhD in plant biology, which is just wrong. Never had a chem student pay for the PhD.