Volume 10, No. 6 FEBRUARY 1999
Lee on Women Physics Majors
Vesilind on Students Evaluating Teachers
Wolbarsht on the Demise of Neandertals
Ferret's Deconstructions: Bricmont-Sokal
Possum (Passim): Readings & Culture Studies
Parrot's Recitations: William James
Editor's Correction (MLA numbers)
Why Few Women Major in Physics
--by Alfred Lee
Department of Physics
Before coming to Duke, I had assumed the people who did not major in physics were simply not very good at it. Physics courses were a "weeding" process: unworthy students were rejected, leaving only a very small percentage (e.g., 3% at Duke), the best and the brightest, as majors. This was personally satisfying, since I was one of the few. The first introductory physics course for scientists and engineers I taught at Duke was an eyeopening experience.
I was astonished to find how many students were "worthy." Having now taught introductory physics to more than 300 students, I would estimate that, at a minimum, 80% of those students would make fine physics majors. In fact, many who do not major would probably be better physics students than those, like myself, who do. Why are there not 200 physics majors per year at Duke, instead of a dozen?
Another disturbing aspect of physics is the lack of women and historically under-represented minorities in the field. In the 90's, female students have represented between 5 and 15% of the physics majors at Duke each year. For example, the majors' course I am teaching this term has 3 women and 1 African-American out of 24 students. I believe this is the highest percentage and number of either I have ever taught in a physics majors' course.
Over the last six years, I have investigated why there are not more physics majors, in particular female ones. I have talked to students about why they choose their majors, including those who start in physics and then switch to another major. I have read the physics education literature (1), taken part in a discussion group interested in the role of women in science, and contemplated the differences between people who pursue physics and those who do not.
Some of the standard reasons just don't fit. For example, in 1970, less than 10% of new MD's, law graduates, and physics Ph.D.'s were women. Today, 50% of medicine and law graduates are women, but physics is struggling to break the 15% mark. Whatever arguments can be made for historical under-representation, economic disadvantage, low parental expectations, bias of senior practitioners, etc., apply equally well to medicine and law as physics.
One cannot even claim that the "pipeline" of women is broken before college: about 35% of Duke students in introductory physics are women. The easy answer, that female students are not as "good" as the male ones, doesn't hold up. Female students in introductory physics earn the same distribution of grades as male students, or perhaps do a bit better. In my sections of introductory physics, even the gender of the top two or three students has split pretty evenly between male and female students.
In searching for a reason, I was brought back to a nearly universal characteristic of physicists: they are self-confident to the point of arrogance. The only other group I know with similar personality traits are baseball players. I believe it is for the same reason. Both groups fail constantly. In professional baseball, only one in ten swings of the bat leads to a successful hit. Similarly, an "easy" physics test will often have good students getting 50% of the questions wrong. For example, the departmental introductory physics placement exam, which is not considered to be particularly unreasonable, requires a score of only 50% to pass. Perhaps only arrogant white guys can withstand the failure?
What transpired in a recent class of introductory physics for majors makes me think I am on the right track. The students were a bit lazy, but all very good. Only one out of seventeen would perhaps not be wellserved by a major in physics. One student, an Hispanic-American women, did well, earning an A (which put her in the top five students in the class). She will probably never take another physics course again. Why?
She did well in physics (and everything) in high school. However, she doesn't believe she is good at physics now. I protested that, if she drops physics because she thinks she isn't good enough at it, two thirds of this class of budding majors would have to do the same. I was not able to convince her it was O.K. to have gotten to the end of the course not knowing 30% or more of the material.
Each term, I take a group of students who basically like physics, and have them fail at it repeatedly (2). No wonder few students want to major in physics! Why does this differentially drive women out of physics? Remember, it effectively drives everyone out of physics: 95% of male and 99% of female introductory physics students do not major. There are cultural reasons (both societal and within the physics department) why women are slightly preferentially driven away, but the experiences of women and the vast majority of men who do not continue in physics are basically similar and strongly negative.
Any plan that encourages women to stay in physics is likely to increase the number of men in physics too. The answer to increasing the number of women in physics will also answer how to have more majors in general. It probably would increase the number of under-represented minorities as well. There are not several problems to be solved, but one.
I have tried a few approaches to improve the situation. One technique that helps is to make it very clear to the students that everyone gets questions wrong; 50% wrong is to be expected. I explain the grading scheme carefully, and write letter grades as well as point scores on tests; the 50% and B are listed side by side. Grading the introductory courses on an absolute scale also helps to lessen the competitive tendencies of first year students.
These small changes have not been enough. When I first wrote this essay, I concluded with an optimistic hope that things in physics might yet change. However, a collection of events (including incidents where the physics faculty preferentially made life worse for female students) has made me pessimistic.
It is not 1980; it is 1999. Too much time has passed with too little progress. Despite the University's best attempts at getting grant money to address "gender equity," the fundamental interaction of faculty with students, and students with their own success or failure, will remain extremely difficult to transform. Without a transformation, physics will remain a male-dominated field.
1 There is a very interesting book called, "They're not Dumb, They're Different: Stalking the Second Tier," by Shiela Tobias (1990 Research Corporation). It involved a study where literate adult students (non-scientists) enrolled as "stand-in undergraduates" in introductory science classes and recorded their impressions.
2 Why do physics courses have to involve so much failure? Nature doesn't follow our naïve expectations, and physics makes that very clear. The mantra of physics education currently is "elicit, confront, resolve." Find out what the students are thinking, confront them with evidence that it is not how nature really works (or is logically inconsistent, or is incomplete), then help them resolve their misconceptions. This is not a plan likely to build self-confidence.
ARE STUDENT EVALUATIONS
OF TEACHING EFFECTIVENESS?
--by P. Aarne Vesilind
Dept of Civil and Env. Engineering
Someone keeps putting copies of a Chronicle of Higher Education article entitled "New Research Casts Doubt on Value of Student Evaluations of Professors" into mailboxes in the College of Engineering at Bucknell University, where I am spending a sabbatical leave. The article, discussing a study conducted by two professors at the University of Washington, concludes that the way to get high student evaluations is to be an easy grader, in effect casting doubt on the validity of student evaluations. The reappearance of this article in our mailboxes seems to be a shadow campaign against student evaluations, at a university that values teaching and uses teaching as the primary criterion for granting tenure. Apparently this individual feels strongly that student evaluations are unfair and wants to convince others by redistributing the Chronicle article.
I must admit that I am a firm believer in student evaluations and that I had difficulty accepting the conclusions as reported in the Chronicle article. So I decided to read the original studies to see if the reporting was accurate. The papers referred to in the Chronicle article are part of a compendium of papers in which the question of student rating validity is discussed.2 One of the papers (Greenwold and Gillmore) does indeed conclude that higher grades result in higher evaluations but that it is possible to normalize the ratings if this is thought to be a contaminant. All other authors disagree with the conclusion that leniency is a serious contaminant, and subsequent discussion papers also argue for the accuracy of student evaluations. The author of the Chronicle article apparently chose to quote only the one paper casting doubt on the validity of student evaluations and ignored all the others.
The authors of one of the papers in
the original compendium (Marsh and Roche) conducted a thorough review of
student evaluation literature and presented their conclusions as shown
in the table below. I have made some comments in brackets.
Relationships Found Between Student's Rating and Background Characteristics
Background Characteristic Summary of Findings
Prior subject interest
A higher interest rate results in more favorable evaluations.
[Comment: This is news? If students are interested in a course they
work harder and get more out of it. Besides, how can we tell if the
interest existed before the start of the course or was generated by the
course or the instructor?]
Expected grade/actual grade Class-average
grades are correlated with class-average students.
[Comment: How can we ever know if the higher grades were because
of grading leniency, superior leaning, or preexisting differences? If I
get all of my students so excited about a course that they work very hard
and learn a lot, why should I not give them all high grades?]
Reason for taking a course Elective
courses and those with a higher percentage of students
taking the course for general interest tend to be rated higher.
Workload-difficulty Harder, more difficult
courses requiring more effort and time are rated
somewhat more favorably.
studies show smaller classes are rated somewhat more favorably. Some research
shows that large classes are also rated more favorably.
[Comment: Small classes tend to be upper level seminars so that makes sense, and
very large classes are often taught by master teachers who are teaching the course
because they are such good teachers.]
Level of course or year Graduate-level
courses are rated somewhat more favorably; upper
division courses are rated somewhat higher than lower division courses.
Instructor's rank Little or no effect.
Sex of instructor or student Little or no effect.
Academic discipline Somewhat higher rating in the humanities and lower ratings in the sciences.
Purpose of ratings Somewhat higher if students know they are to be used for tenure or promotion.
Administrative conditions Somewhat
higher if ratings are not anonymous and if professor is present
when ratings are being completed.
Student' s personality Little or no effect.
Their overall conclusions, matching the conclusions of many others,3 is that student ratings provide a valid and reproducible measure of teaching effectiveness, but that this should be only one measure and other data such as observation should be used.
Often the argument used to discredit student ratings (especially at research universities) is that student evaluations are meaningless because the only true measure of teaching skill is to ask for evaluations from students who have graduated some years back. The hypothesis is that only later in life will students realize from whom they learned the most and who the best teachers really were, and that asking students immediately after a course is finished favors professors who by virtue of their pleasing personality or lack of course rigor will receive high ratings from students.
But research has shown this hypothesis to be false. The data indicate that there is no significant change in teacher ratings with time. Students asked 10 and 20 years after graduation to name their best instructors will name the same instructors whom they rated highly while they were students. The tough instructors who had poor teaching skills (regardless of how difficult their courses) are still rated poorly.5
Another indication that the opinion of teaching effectiveness does not change with time is the stories alumni tell of their best teachers. According to a recent research study (6), the instructors rated highly by students are still rated highly when the students become alumni. No difference was found for humanities or the sciences. When alumni were asked to describe their former professors they told stories that illustrated the positive effect the teachers had on their lives. One student, finishing his favorite story about his former professor, ended reflectively -- "I miss him" he said -- thirty years after graduation!
1. Robin Wilson "New Research Casts Doubt on Value of Student Evaluations of Professors,"
The Chronicle of Higher Education 16 January 1998, p. A13
2. Herbert W. Marsh and Lawrence A. Roche, "Making Students'
Evaluations of Teaching
Effectiveness Effective," American Psychologist v.52, n. 11, pp 1187-1197; Sylvia d'Apollonia
and Philip Abrami, "Navigating Student Ratings of Instructors" American Psychologist v.52, n.
11, pp 1198-1208; Anthony Greenwald and Gerald M. Gillmore, "Grading Lenience Is a
Removable Contaminant of Student Ratings" American Psychologist v.52, n. 11, pp 1209-1217;
Wilbert J. McKeachie "Student Ratings," American Psychologist v.52, n. 11, pp 1218-1225.
3. Cashin, W. E. "Student Ratings of Teaching: A Summary
of the Research," IDEA Paper No.
20, Center for Faculty Evaluation and Development, Kansas State University, 1988
4. L. M. Aleamoni and M. Yimer, "An Investigation of the
Relationship Between Colleague
Rating, Student Rating, Research Productivity, and Academic Rank in Rating Instructional
Effectiveness," Journal of Educational Psychology, v.64, pp 274-277, 1973
5. J. Drucker and H. H. Remmers, "Do Alumni and
Students Differ in Their Attitudes Toward
Instructors?" Journal of Educational Psychology, v.42, pp 129-143, 1951
6. Barbara Harrell Carson, "Thirty Years of Stories," Change v.28, n. 6, pp 10-17, 1996
FROM THE ARCHIVES
EdNote: Our polyglot colleague Myron
Wolbarsht, then at the Duke Eye Center, published
the following analysis of an ancient mystery in Science
magazine a full quarter century ago (186: 1974, 18-619)--well ahead of
the recent outbreak of speculation about Neandertal culture.
THE DEMISE OF THE NEANDERTALS:
WAS LANGUAGE A FACTOR?
Kolata reports on a possible link between the disappearance of the Neandertals and their inability to use an easily understood language. However, there may be an organic cause for the disappearance of the Neandertal in addition to any language difficulties.
The Neandertals had a well-developed burial ritual. In many of the later burials it is evident that the skull had been opened to allow removal of the brain (1). Often the long bones are also cracked, apparently for removal of the marrow. The treatment of the skull and bones suggests the use of rituals similar to those practiced by contemporary cannibalistic tribes in New Guinea and other places as documented by modern anthropologists.
Such practices, in one place at least, have led to the development of a slow viral infection, kuru, in the population (2). The kuru virus is neurotropic and fatal. The infection leades to a complete degeneration of the nervous system due to the extensive viral multiplication within it. The virus may develop a long time after infection-10 to 30 years. It presumably is spread by the women and children (mainly female) eating the brains of a deceased member of the tribe or by allowing infected material to enter the bloodstream through cuts in the hands when the skull and bones are opened. Gadjusek (2), among others, has hypothesized that any tribe practicing cannibalism will develop a similar type of virus.
The virus may arise from some animal reservoir but, in general, dies out in humans. However, cannibalism permits serial passage of the virus from human to human, thus increasing its virulence and infecting the population. The prevalence of this disease combined with increased devotion to cannibalism would lead to the annihilation of a tribe, similar to that once facing the Fore in New Guinea. It has been suggested that the virus may survive in the earth surrounding the bones. Perhaps incubation of this earth would indicate its continued virulence (3).
Retardation in the development of a language on an anatomical basis could, of course, further decrease the chances of survival of the few remaining members of the tribe. A language problem would also isolate them from a larger competing population. We should include the possibility that, if the Neandertals were warring with the Cro-Magnons, their lack of ability to pronounce the vowels [a], [i], and [u] would perhaps have effected a "shibboleth." Thus "bat" would have been pronounced "bot" by Neandertals, who failed the test (and failed to live).
1. K. Gorjanovic-Kramberger, Mitt. Anthrop. Ges. Wein 29, 1 (1889), and many others. See
C. B. Courville, Bull. Los Ang. Neurol. Soc. 15, 1 (1950) for a review.
2. D. C. Gadjusek, Trans. R. Soc. Trop. Med. Hyg.
57, 151 (1963). See reviews in J. D. Matthews,
R. Glasse, S. Lindenbaum, Lances 1968-II, 444 (1968); Am. J. Pathol., 68, 626 (1972).
3. D. Regan, personal communication.
4. Judges 12:6.
5. I thank Susan C. Smith and B. Anderson, Jr., for their assistance.
--Book Review by Thomas Lewis
In a memorable scene from "Annie Hall," Woody Allen stands fuming in line ahead of a man pontificating on Marshall McLuhan's place in popular culture. When Allen contradicts him, the man pulls rank by flashing his credentials as a Columbia professor. Allen retaliates by dragging McLuhan himself into the scene as the ultimate rebuttal witness. "You know nothing of my work," McLuhan snaps at the sputtering academic. "How you ever got to teach a course in anything is totally amazing."
The same spirit of expertly delivered comeuppance inhabits "Fashionable Nonsense," in which two scientists debunk postmodern philosophers' pernicious misuse of science and math. Assuming the reins are Alan Sokal, a physics professor at NYU, and Jean Bricmont, a theoretical physicist at the University of Louvain in Belgium. Their mission, they write, is "to 'deconstruct' the reputation that certain texts have of being difficult because the ideas in them are so profound. In many cases we shall demonstrate that if the texts seem incomprehensible, it is for the excellent reason that they mean precisely nothing."
Sokal and Bricmont step directly into the natural hostility that sizzles between postmodernist philosophy and hard science. Postmodernists have often endorsed an aggressive relativism that denies the existence of "truth" in favor of an egalitarian, multiple-points-of-view approach. That ethos helps sociologists avoid chauvinism when comparing cultures, but the freewheeling, nobody's-wrong-if- everybody's-right spirit doesn't mix well with a scientist's conviction that immutable laws define the universe, and that reality is, well, real.
Making matters worse, many prominent post-modernists have borrowed concepts, language and even equations from science and applied them to politics, gender issues and sociology. That practice yields balderdash, Sokal and Bricmont maintain, and they've combed the literature for the most blatant infractions of logic.
Famed French psychoanalyst Jacques Lacan is the book's first target. When Lacan expounds that "the erectile organ is equivalent to the square root of 1" or that a "torus is exactly the structure of the neurotic," many assume his assertions have some basis other than lunacy. Perhaps only a mathematical genius could propose these insights? Don't buy into the con, Sokal and Bricmont advise. When he tosses tori and square roots in with penises and neuroses, Lacan is betting that a math-induced brain freeze will suspend the critical faculties of his audience. Not so readily bamboozled, the authors of "Fashionable Nonsense" expose Lacan's mathematical antics as "showing off a superficial erudition and manipulating meaningless sentences."
Alas, the authors demonstrate, Lacan is not the only one bluffing his readers when he name-drops impressive-sounding scientific concepts into flowery discourse. For the French philosopher Luce Irigaray, the main issue Albert Einstein raises is, she writes, "his interest in accelerations without electromagnetic reequilibrations." Unfortunately, there is no such thing.
When Jean Baudrillard says that the Gulf War took place in "hyperspace with multiple refractivity," is he just making that stuff up? Indeed he is, and his text continues in what Sokal and Bricmont, connoisseurs of the absurd, charitably term "a gradual crescendo of nonsense." One can almost hear the air whistling out of acclaimed thinkers Julia Kristeva, JeanFrancois Lyotard, Bruno Latour, Paul Virilio, Giles Deleuze and Felix Guattari as Sokal and Bricmont cite and deflate their pretensions.
The germ of "Fashionable Nonsense" began in 1995, when Sokal submitted an article to the au courant cultural studies journal Social Text. In a parody of postmodernist pomposity, Sokal purposely fashioned his piece to be "a melange of truths, half-truths, quarter-truths, falsehoods, non sequiturs and syntactically correct sentences that have no meaning whatsoever." The editors of Social Text swallowed Sokal's bait and, mistaking his lampoon for serious scholarship, published it. . . .
Just as debate over the Sokal Affair was dying down, he teamed up with Bricmont and published the first version of their book in France (under the title "Intellectual Impostors"), where it garnered extensive media attention and a blistering reception from French intellectuals. That continental firestorm may help explain the defensive tone permeating this revised American edition of Sokal and Bricmont's work. Barely five pages into the introduction, they are busy refuting the attacks European critics leveled at them. Apparently stung by allegations that they overstepped their bounds, they begin this book declaring themselves competent to comment only on the small fraction of postmodernist philosophy that runs afoul of scientific rigor. . . .
So the next time you're stuck in line with a philosopher opining that E=mc2 is a gender-biased equation (Irigaray), or that Einstein was a closet dictator because his thought experiments sent exploitable nobodies hurling around the galaxy at the speed of light just on his whim (Latour), don't despair. You may not be able to drag in Sokal and Bricmont from off-camera but at least you can brandish their book.
Random Readings & Culture Studies
THE MALTHUSIAN SYNDROME:
"In absolute numbers, putting the first billion people on Earth took from the beginning of time to about 1830. Adding the latest billion took twelve years."
-Joel E. Cohen, The New York Review of Books (10/8/98, 29)
ECOFAILURE IN CHINA:
"Air pollution in Chinese cities has been estimated to be five to ten times that of US cities. . . . In north China, where rainfall is scant, diversion of water to agriculture and industry has caused streams and small reservoirs to dry up. During the first six years of the 1990s the Yellow River, China's second largest river, was dry for a total of 333 days. In the 1950s the water table in Beijing was about sixteen feet below the surface; today it is more than 150 feet down. . . . China now has two thirds of the arable land it had four decades ago, and 2.3 times as many people. . . , and, nationwide, about one third of China's topsoil has washed into the Pacific Ocean."
--Liu Binyan and Perry Link, The New York Review of Books (10/8/98, 21)
A MODEL DIASPORA:
"The 30 million overseas Chinese account for more goods and services than the entire mainland China economy."
--Martin Peretz, The New Republic 6/15, 42
NEXT WE'LL TRY COW GELDERS:
"[Benjamin Franklin suggested] in May of 1774 that a hundred sow gelders be sent to accompany the British army to the colonies, where they could systematically castrate all the American males and thus put a stop to the dangerous increase of the rebellious population."
-Edmund S. Morgan, The New York Review of Books (10/8/98, 42)
INQUEST OF THE CONQUEST:
"Cortez and his men comprehended little of what they saw, and were not especially curious. These crude zealots massacred Indians, built Christian altars where they had smashed idols, and went mad at the sight of gold. . . . Unlike the children of the Enlightenment and the Protestant Reformation who settled the east coast of what became the United States, Cortez and his men came to steal, not to work or build cities. Religious dogmatists who combined the worst of Spanish and Moorish culture, they lacked the habit of process, of investing years of labor to achieve material gain--the bourgeois mentality, in other words."
--Robert D. Kaplan, The Atlantic Monthly 8/98 (47)
Adventures in Noble Thinking
"The belief in free will is not in the least incompatible with the belief in Providence, if you allow him to provide possibilities as well as actualities to the universe. . . . Suppose that the world's author put the case to you before creation, saying, 'I am going to make a world not certain to be saved, a world the perfection of which shall be conditional merely, the condition being that each several agent does its own level best. I offer you the chance of taking part in such a world. Its safety, you see, is unwarranted. It is a real adventure, with real danger. . . .'
"I find myself willing to take the universe to be really dangerous and adventurous. . . . I am willing that there should be real losses and real losers, and no total preservation of all that is."
--William James, "Pragmatism and Religion"
EdNote: CORRECTION--In the January FF, I introduced Mark Kelleyís "Report from the Job Market" with the remark that ìthe Modern Languages Association is the leading professional organization of about 30,000 professors of languages and literature.î A recent Chronicle of Higher Education, however, corrected that figure when it described ìthe Graduate Student Caucus, a 5,000-member affiliateî as ìrepresenting nearly a third of the MLAís total membership.î As a member of the MLA since 1962, when it did include about 30,000 members, I apologize for my careless reliance on ancient memory.
The Faculty Forum is published monthly by the Academic Council of Duke University. Editor: Victor Strandberg (English). Editorial Advisory Board: Matt Cartmill (Biological Anthropology & Anatomy), Howard Clark (Biomedical Engineering), Buford Jones (English), Seymour Mauskopf (History--on leave), and Kathy Rudy (Women's Studies).
The purpose of the Faculty Forum is to provide an opportunity for faculty members to discuss topics of common interest. The FF therefore welcomes, and indeed depends on, contributions by faculty members. Contributions may be published anonymously, but the editor will need to know the identity of all contributors so as to authenticate the contribution and permit appropriate editing procedures. We welcome fact, opinion, and creative whimsy from across the wide range of Arts & Sciences and Professional Schools--book reviews (new or reprinted), short essays, serendipitous items from random reading, and voices from the eternal kulturkampf.
Contributions of every kind should be sent, in both print and (if lengthy) computer disk, to the editor,Victor Strandberg, in 315 Allen Building. Voicemail is 684-3976; no objection to calls at home, 489-5531. E-mail is firstname.lastname@example.org FAX (English department) is 919-684-4871. The deadline for submissions is about the 25th of each month.