Rethinking the Normality of Attrition

There are few things so beloved by the professoriate as the faculty retreat – amirite? And the highlight of every faculty retreat is surely that hour when we gather and form small groups to contemplate How Diversity Is Making Us Stronger!!1!! These are nearly always well-planned, adroitly led, and very effective. In my dreams.

At one such gathering, the first exercise our group was given consisted of a sheet of paper with four photos: a young white man in casual clothing; a middle-aged white woman in a suit; a young African-American woman in a suit; and an old, bespectacled, gray-haired, bearded eminence in tweed jacket and tie. Our task: which of these people did we think was a professor, and why? Nobody wanted to go anywhere near that booby-trap. Nobody, that is, except the old, bespectacled, gray-haired eminence in a jacket in our group. He promptly pointed to the bearded dude and said “oh, he’s the professor. He just looks like one. Don’t you think that’s how a professor is supposed to look?” The diversity workshop leader happened to be standing next to our group at the moment and the rest of us cringed. Now, this professor was a really nice guy, and he said this without any guile. In retrospect I applaud him for saying what we were all thinking but self-censoring ourselves from saying. Gray-haired bearded dude did look like what we thought a professor should look like. The question was why did we, committed as we were to diversity, still think that? How could we come to see the others – especially the women – as equally valid images of the professoriate?  And what did all this mean for our work at the university?

Well, it should be no surprise, and should not make anyone feel guilty or ashamed, to realize that we carry these internalized stereotypical images of what a professor or scientist or engineer looks like. We daily bathe in the sea of stereotypes.  We may also carry a picture in our heads of what a successful STEM student looks like, without realizing it, and may make advising decisions based on that image rather than on the student’s interests, desires, and real potential.

The first step in interrupting the circuit is to interrogate the term “successful student”. Is a successful student one who makes top grades? One who rallies after a failure? One who doesn’t have a lot of distractions to get in the way of focusing on the degree? One who learns how to manage the non-negotiable constraints of life and still continue with their studies? One who goes on to a satisfying and successful STEM career post-graduation? One who takes their STEM degree as a springboard into another career direction? Is a successful student one whom we help to succeed?

Of course, I can tell you my anecdata about getting a D in calculus and going on to a successful STEM career despite a frosh advisor who suggested I switch out of engineering, and you can counter with your scores of advisees and your, as we will see, oh-so-unfortunate example of George.  And then I’ll walk over to my bookshelf and peruse the research.

The classic reference text on students switching out of STEM majors is, of course, Seymour and Hewitt’s Talking About Leaving: Why Undergraduates Leave The Sciences. If you are a STEM professor, make yourself familiar with this book if you are not so already.  The book is an exhaustive presentation of the results of a three-year study of 335 students at seven four-year institutions of different type and location. The authors question the assumption that leaving, or switching, is natural or normal.

The revolution did not swing by anytime in the last 15 years so you can pretty much go with what the book says. Here’s the

most important single generalization arising from [the] analysis…switchers and non-switchers [were not] two different kinds of people. That is to say, [they did not] differ by individual attributes of performance, attitude, or behavior, to any degree sufficient to explain why one group left , and the other group stayed…What distinguished the survivors from those who left was the development of particular attitudes or coping strategies – both legitimate and illegitimate. Serendipity also played a part in persistence, often in the form of intervention by faculty at a critical point in the student’s academic or personal life. [emphasis mine] [p. 30]

It turns out that STEM is bleeding students, male and female, white students and students of color. Only, the bleed rates for females and students of color are slightly higher than for white males, so the overall impact of culling the herd is to reduce diversity. After all that hard work to recruit the best and brightest to your uni, and to get all those women and students of color to your doorstep! Such a shame. Well, what can you do, eh?

Seymour & Hewitt note, by the way, that inappropriate choice, underpreparedness, and overconfidence, while present for many students of color, are not sufficient factors to explain the higher switching rate of this group compared to white students. So one thing you can’t do is lay the burden for the problem on the students.  The extra difficulties that students of color face include: differences in ethnic cultural values and socialization; internalization of stereotypes; ethnic isolation and perception of racism; and inadequate program support.  It’s true. Your unis are not doing a good job of supporting students of color.

Seymour & Hewitt speak in their conclusion of a desire to marginalize the issue of wastage of students, given the consequences of taking seriously the loss of 40 to 60 percent of a group of students with above average ability.

Switching is not defined as a problem when it is believed to be caused, on the one hand, by wrong choices, underpreparation, lack of sufficient interest, ability, or hard work, or on the other, by the discovery of a passion for another discipline. Either way, there is little that faculty feel they can, or should, do about people who leave for such reasons. The difficulty about our data is that they support neither type of explanation for switching. We find no support for the hypothesis that switchers and non-switchers can be sufficiently distinguished in terms of high school preparation, performance scores, or effort expended...Nor do switchers neatly divide into those who are pushed out (by inappropriate choice of major, lower ability, poorer preparation, lower levels of interest, or unwillingness to work), and those who are pulled out (because they discover a vocation elsewhere)...[W]e posit that problems which arise from the structure of the educational experience and the culture of the discipline (as reflected in the attitudes and practices of S.M.E. faculty) make a much greater contribution to S.M.E. attrition than the individual inadequacies of students or the appeal of other majors. [p. 392]

Ouch. That hurts.

Students who wash up on your advising shores performing poorly in their major classes may be doing so for any number of reasons. In my opinion, if you let them get to their junior year and flunk a major course three times without an intervention, your uni is failing that student, and not by giving them a failing grade, if you follow me. Read the conclusions chapter of Seymour and Hewitt if you read no other part of it. There's more in there about the groups of students that are being lost from STEM, groups that faculty members might very much want to retain. And rethink your notions of the successful student and beneficial advice to switch majors. Even if you think you're doing the student a favor, is it really a good thing for your uni to continue recruiting, but not retaining, STEM students?

Share

22 responses so far

  • LM says:

    What, then, would your advice to George be?

  • Anecdata: I was a George. I almost left my major. A professor overheard me talking about this and invited me to his office to be a place to bounce my thoughts off of him. IT MEANT SO MUCH. So did the regular heart to hearts with a female-of-my-general-ethnicity TA who guessed that I might be struggling with depression. I didn't change majors. I realized that I needed to spend more time with female grad students. They provided a major source of comfort, encouragement and mentoring to get me to the decision to continue on to grad school.

    I don't know that PLS is the correct person to intervene for George. As a faculty adviser, meeting with student 2-4 times a year, he may not be able to build up enough of a relationship with George to discern if George is having a hard time, or if this really is a bad major for him. But someone needs to step up to the plate for George, perhaps a professor he's seen in multiple classes over his time at the university, or a staff member in the department office, or a TA who recognizes a younger version of themselves.

    I wholeheartedly agree with you that it is incumbent on university faculty to re-examine and broaden our idea of what a successful STEM student looks like. They may not look like younger versions of ourselves.

  • proflikesubstance says:

    I'll admit that this statement bothers me, but not for the reasons you think:

    [W]e posit that problems which arise from the structure of the educational experience and the culture of the discipline (as reflected in the attitudes and practices of S.M.E. faculty) make a much greater contribution to S.M.E. attrition than the individual inadequacies of students or the appeal of other majors.

    It bothers me because they are making an inference based on not seeing a pattern. I haven't read the whole study so it may be better in context, but "it's not A, B or C, so it must be X." is not a very strong position to be in. Would be great if I could publish that way, but I can't. These students presumably still exist, so talking to them might be a nice way to determine what is actually going on. I'm not saying I don't believe their posituition, but we have the technology (phone) to get better data.

    I don't know that PLS is the correct person to intervene for George. As a faculty adviser, meeting with student 2-4 times a year, he may not be able to build up enough of a relationship with George to discern if George is having a hard time, or if this really is a bad major for him. But someone needs to step up to the plate for George, perhaps a professor he's seen in multiple classes over his time at the university, or a staff member in the department office, or a TA who recognizes a younger version of themselves.

    I don't know either, but I'm the one sitting at the desk. Whereas it is a nice thought exercise to figure out exactly what each student would like so that they can be supported to their needs, some of us actually have to do this job. Similar to Z's statement above about the university failing a student who hasn't passed a class after three attempts, these are all good sentiments, but it is the current reality. We can strive to improve the model, but we still have to exist today. I can't be the perfect advisor to every student I meet, be that because of time or ability to empathize. I also can't pull the budgetary purse strings to bring in individuals to tackle the advising burden of our department in a way that best meets the needs of the students. Undergraduate advising is not my primary, secondary or even tertiary job, so I do what I can with the resources I have to help as many students as possible. If I met with every advisee 2-4 times a year some of them would be left in the lurch when I had to leave after not getting tenure.

    WRT the system leaking those we most want to keep, I am fully on-board there. Everyone seems to be inferring that "George" is a somehow disadvantaged student (interesting in its own right), but I see plenty of Georges who fall smack dab in the center of the most privileged groups. I think we (both my university and STEM in general) have some major minority retention issues (women are well represented from undergrad through PhD in my field) to deal with and Seymour and Hewitt's article should provide food for thought. I'm not going to list the steps I take to combat this IRL because I don't need the internet affirmation, but Z is right that combatting internal biases among faculty is a major hurdle. Getting more people to acknowledge that would be a great first step.

    • Zuska says:

      You can't dismiss Seymour & Hewitt that easily...it's not like they extensively studied the reasons students were leaving, determined it wasn't x, and then posited y by pulling it out of their asses. There are reasons why they come to that supposition and the reasons are in the interviews they did with 335 students. Btw, they emphasize that current STEM practices are causing the bleeding of ALL students, not just women & students of color. While the bleed rates are higher for those groups, they are dismal overall. 40-60% is nothing to be proud of.

      This is an institutional problem & requires an institutional reform - you on your on can't cure it tho you may save a life here or there. But I do think a necessary step in that transformation is professors at least becoming aware of the relevant research, and then not dismissing it as irrelevant because "there's nothing I can do" which is precisely the sort of defeatist, accept the status quo as natural attitude S&H critique as helping to perpetuate the problem.

      Maybe it's not your job to save George. But now you know what the research says about how institutions are crushing whole crops of kids, so you can't rely on the old myths about why students leave STEM anymore. What you do with that information is your call and depends on how you are positioned.

  • DJMH says:

    I think a huge amount of attrition is probably attributable to the fact that in the sciences, you can do all the reading and turn in all the problem sets and still fail, whereas in pretty much any other discipline on campus, as long as you do all of the reading and turn in some sort of essay or equivalent without plagiarizing, you will pass.

    It is possible that grappling with failure, or hard right/wrong answers, is harder on students from certain backgrounds (which could include *both* backgrounds where it is strongly expected the student will succeed in college, and backgrounds where the student is the first to go to college....).

    Unfortunately the answer to *those* issues probably starts in pre-K, teaching kids "resilience" or "toughness" or whatever the buzzwords are right now.

  • becca says:

    PLS- keep in mind, you may have more retention issues with women than you think. In the PISA tests, girls out perform boys in science everywhere except the US, Canada and Britain. If I recall correctly, ~55% of 15 year old girls are interested in a STEM career, compared to ~45% of 15 year old boys (not sure of the exact numbers, but it was ~10% more girls). Maybe the natural state of the world, before our culturally specific flavor of misogyny, is for women to be the vast majority of scientists. We've never had a world where it could happen before, but that's no reason to assume it's not possible.

    I don't know what your specific field looks like, but I see again and again men and women in biomedical science thinking "oh, women are 50% of the grad students, problem solved!" and I want to give them a good thwap, so I figure it's worth bringing up.

    • antistokes (allison l. stelling) says:

      "I don't know what your specific field looks like, but I see again and again men and women in biomedical science thinking "oh, women are 50% of the grad students, problem solved!" and I want to give them a good thwap, so I figure it's worth bringing up."

      As someone who does biomedical research but did her doctoral work in analytical/physical chemistry.....yeahhhh, I feel that way sometimes. There were tons of fellow chicks in my genetics courses. The complete reverse could be seen in my upper division physics courses-- all math dudes who liked to yell at each other. Now I've got neuroscience collaborators who don't know a lick of physics and physics dudes who've never taken biochem, and they are all trying to get me to "translate" what's being said. One of my co-authors, a female neuroscientist, said to me a few summers ago that "we have to get the biology girls talking to the physics boys...."

  • Jim Thomerson says:

    I've been a Deans' College Advisor for a number of undergraduate students. This is a fairly close association. Some have gone on to get a PhD and study fishes. However, a couple I recall decided to take another path, one to a French major, the other to Journalism. I thought I had done well by those students.

    I read something years ago which I think has some verisimilitude. Male students seek fair treatment. Female students want to be seen as individuals. I have, over the years, had five minute, shoot the bull, conversations with female students who were not doing well in class, and seen an immediate improvement in their work.

    I think the general problem is our adoption of mass education, starting with the consolidation of school districts, in hopes of achieving economy of scale. People have different temperaments, and a Procrustean approach is going to fail for a fair percentage.

  • proflikesubstance says:

    Becca, I specifically stated "undergrad through PhD" because we DO have a retention beyond grad school. However, there were only so many parenthetical statements I could justify in one sentence and this discussion is about undergrad advising.

  • drugmonkey says:

    in the sciences, you can do all the reading and turn in all the problem sets and still fail, whereas in pretty much any other discipline on campus, as long as you do all of the reading and turn in some sort of essay or equivalent without plagiarizing, you will pass.

    Did you want to walk that one back a little bit DJMH?

  • Jim Thomerson says:

    I have been in a position to see campus wide grade reports. All students in education courses received an A. This is, of course, what happens when you have professors who really know how to teach. My requests to have the education faculty train us science folks, who generate C+ class averages, as to how to do it right, were never taken seriously.

    DJMH is being kind.

  • antistokes (allison l. stelling) says:

    Yeah, gonna side with Jim Thomerson on this one. In fact, a lot of the faculty at Reed left fairly prestigious universities (my physical and quantum chem prof taught at Harvard for years) to be at Reed, since the college de-emphasizes GPAs and gives the profs a lot of freedom in how they come up with grades. My undergrad thesis adviser (an analytical chemist) once told me about how, at the college he was at formerly, he had a student threaten to sue him over a bad letter of recommendation-- which was part of the reason he left. Nowadays, he does not allow any of his students to see his letters of rec, and Reed sides with him.

    On another note, going back to the girls vs boys thing: a lot of my female lab partners insisted that I ask their questions on the problem sets because they didn't want to look stupid in front of the prof. (I never minded, I like badgering profs until I have the concept clear in my mind; esp. with physics problems that involve generating a physical model in your head and translating this into math.) Whereas the boys usually thought they were god's gift to clever and had to be taken down a peg--- lest they think too quickly in the wrong direction. (One of the string theory professors that taught my partner's theoretical physics course one called his audience of mostly male math and physics students idiots for not being able to to multivariate calculus in their heads. He was an old Russian guy, and most of the boys thought he was hilarious and took it as a challenge rather than as a verdict.)

  • Alex says:

    antistokes, I disagree with Zuska on a lot of things here, but "This way of doing things worked out so well at Reed!" is not terribly useful in getting a wider perspective on higher education issues in the US. I know a few people at Reed, and I like them, but it's a small place, a unique place, a selective place, and not a cheap place. The vast majority of the action in US higher education is in large state universities (and I don't mean R1 flagships and runners-up, I mean Compass Direction State University) and community colleges. Even among private schools, Reed is less typical than, say, Random Private Urban Non-R1 With A Few Grad Programs. (A surprisingly common genre of school that serves a larger population than Reed and usually offers a wider range of degrees.)

  • antistokes (allison l. stelling) says:

    .....um. I was a fin aid kid (we referred to the kids whose parents where shelling out for tuition as "trust fund babies"-- they were nice folk, but a bit....well, if you know Reedies you know what I am talking about, it's like they don't understand what money is). I have federal loans to pay off (tho the college gave me full tuition, need based), which is why I went to Stony Brook-- a huge public research university-- for grad school. So I could get a real job afterwards and pay back my debt. (A lot of my friends went to CalTech, MIT, BU, UCSF, etc-- basically everyone wanted to live in either Boston or SF.) And lord knows the Reed way would not work for the majority of people, not everyone wants or needs to be a prof when they grow up. You only need so many of those per unit humans. I saw an interesting solution in German unis, which involved people actually paying their taxes and offering the education for free. They got of lot (some might say, too many) of close ties with industry too, so trained kids are funneled straight into high tech jobs. There's a few problems with the German system (too many engineers with OCD-- it's a bit of a problem since they run the country), but overall I think it works out for the majority of their population.

  • Alex says:

    I'm not questioning your personal financial situation, antistokes. For the record, I also got a full ride to an expensive private school, and then went to a large public university for my PhD. For that matter, the author of this blog got a Master's at MIT, another place that is very unique (and proudly so). However, all of that is irrelevant if the question is about systems rather than individuals. My point is that there's only so much we can learn about What's Going On In Higher Ed These Days from looking at a place like Reed, which is proudly (and rightly!) unique in many ways. It's not about you or me or Zuska as individuals and how typical or atypical we might be. It's about whether the unique institutions that provided parts of our training are good places to look for insights into higher education in general.

    There are things to admire about higher ed systems in Germany and many other European countries. However, they usually specialize in high school, so by the time they come to college they are (usually) better-prepared in their major. There are pros and cons for that kind of specialization, and those systems would be well worth examining (for pros and cons) if the question is "Do we want to have a system that specializes people sooner?" However, if the question is "Given that the colleges and universities in the US will be training the products of the high school system that we have, how should they do it?", then Europe seems to mostly be a poor place to look for lessons.

  • antistokes (allison l. stelling) says:

    Also, I had a serious advantage in that while my parents were "poor" (according the the fin aid people at least) they are both college educated. My mom was educated on the GI bill (it's a military family), and my dad worked his way through CalPoly.

    Whereas my partner's parents (he's Irish) have the equivalent of high school degrees and both work in government jobs. But, the Irish educational system gave him free problem sets, and when he started going over the teacher's head he was sent off to the professors at Trinity College. So, the Irish system was able to solve its "poor but smart" Lisa Simpson Problem without loading debt onto its students. But, it's a small country and easier to run. Also, they can't afford him anymore- which is why he's in the States now doing math for a materials science department.

  • antistokes (allison l. stelling) says:

    Well, if Americans would pay their taxes and educated their citizens- which we used to do, if I recall my history correctly- perhaps our unis wouldn't have such issue with incoming students. It's a mess, really. (And like I said, I have some issues with the German system. Wrote a whole essay about this, actually, including a flowchart. Lemme know if you're curious, I'm at antistokes@gmail.com, I can send it to you and I keep all personal correspondence private.)

  • antistokes (allison l. stelling) says:

    heh, i mean educate. ahh, typos. seen a lot of those in bio papers recently.....

  • Zuska says:

    Well, as time passes, nothing changes really. As the NYT will inform you in this article which is basically a redux of everything Seymour and Hewitt did 15 years ago, minus the nuanced attention to gender and race and the substantive conclusions and recommendations. http://www.nytimes.com/2011/11/06/education/edlife/why-science-majors-change-their-mind-its-just-so-darn-hard.html Very interesting and timely read, tho.

  • Jim Thomerson says:

    I changed majors twice, once as an undergraduate, and then between MS and PhD. So maybe I should not be held up as a role model. :-)

  • Selina Lee says:

    Hey Zuzska,

    My name is Selina, and I’m with the Andrew Goodman Foundation, a non-profit organization that seeks to recognize, create, and encourage effective action towards social justice and civil rights. You can read more about us here: http://www.andrewgoodman.org

    In honor of Women’s History Month and this year’s theme (Women in STEM), we’re publishing some content on our blog and social media pages relating to women who’ve broken into these fields, and young girls who are still seeking an entry point. Considering your background and success in your field, we’re wondering if you would like to write a guest post for our blog this month. We’re mostly focusing on the difficulties of entering STEM careers for women, particularly with regards to the local and/or federal government (funding, opportunity, educational resources, etc). If you’re willing to share your insights, advice, and experiences with our readers, we’d really appreciate it. Here’s a link to our blog: http://www.blog.andrewgoodman.org

    Thanks so much, and please let me know if you’re interested. You can contact me at andy@littlebiggirl.co.

    Thanks again,
    Selina