Search News


Browse Archives

News

Too Nice to Land a Job

November 10, 2010

Share This Story

FREE Daily News Alerts

Advertisement

You are reading a letter of recommendation that praises a candidate for a faculty job as being "caring," "sensitive," "compassionate," or a "supportive colleague." Whom do you picture?

New research suggests that to faculty search committees, such words probably conjure up a woman -- and probably a candidate who doesn't get the job. The scholars who conducted the research believe they may have pinpointed one reason for the "leaky pipeline" that frustrates so many academics, who see that the percentage of women in senior faculty jobs continues to lag the percentage of those in junior positions and that the share in junior positions continues to lag those earning doctorates.

The research is based on a content analysis of 624 letters of recommendation submitted on behalf of 194 applicants for eight junior faculty positions at an unidentified research university. The study found patterns in which different kinds of words were more likely to be used to describe women, while other words were more often used to describe men.

In theory, both sets of words were positive. There's nothing wrong, one might hope, with being a supportive colleague. But the researchers then took the letters, removed identifying information, and controlled for such factors as number of papers published, number of honors received, and various other objective criteria. When search committee members were asked to compare candidates of comparable objective criteria, those whose letters praised them for "communal" or "emotive" qualities (those associated with women) were ranked lower than others.

The research found no difference between men and women as letter writers -- both are more likely to describe women with communal words than they are to describe men that way. And the bias appears to act against male candidates who are praised for traits people associate with women. But a much higher proportion of female candidates -- regardless of their overall qualifications -- are praised with these words that appear to hurt their chances of being hired for faculty jobs.

"When you use communal terminology, it is linking people to a feminine type, and they are not seen as credible and they don't get hired," said Michelle Hebl, a professor of psychology at Rice University and one of the authors of the study, along with Randi Martin, also a psychology professor at Rice, and Juan Madera, assistant professor at the University of Houston. "It's not just men doing this to women, and it's not just women being hurt, but it hurts women more."

The research was supported by the National Science Foundation and published in the Journal of Applied Psychology. The National Institutes of Health is now supporting a follow-up study looking at letters of recommendation for medical faculty positions.

In the scholars' analysis of the words that appeared in the letters of recommendation, they found clear patterns of word use for women's and men's letters. Women were more likely to be described with words such as those cited above, as well as "nurturing," "kind," "agreeable" and "warm." Men, in contrast, were much more likely to be described in words classified as "agentive" -- words such as "assertive," "confident," "aggressive," "ambitious," "independent" and "daring."

What the analysis showed is that letter writers didn't need to use words like "feminine" to create female stereotypes -- and that they did so, time and again, with women who had the same intellectual achievements as their male counterparts.

Hebl said that women in academe face a dilemma. Hiring committees appear to devalue women who are identified as people who would be nice or supportive colleagues. But women who aren't seen as nice and supportive "get called bitches," she said. So the solution for women is "to have both sets of qualities" -- the communal and the agentive. But when it comes to getting letters of recommendation, she said, women need to be sure their letter writers focus on the agentive qualities.

"Communal might be nice, but agentive is what's really important," she said. Women perceived as too communal "are seen as being pushovers, not somebody to run a program."

Asked if she believes she would find similar results in faculty searches at liberal arts colleges or community colleges -- institutions that tend to value teaching more than research and that place an emphasis on close ties to students -- Hebl said she guessed there would be only a slight variation. She said that even in stereotypically female fields like nursing, research has shown that many place more of a value on qualities associated with men than those associated with women (even if they also want the latter qualities).

Hebl said that the implications of the research for those writing letters of recommendation are clear: stay away from communal words, whether writing on behalf of men or women.

Given how subtle the issue may seem, and that letter writers may not be conscious of what they are doing, Hebl urged those seeking letters of recommendation to not be afraid of talking about the issue with their letter writers. "Given them a copy of the research," she said.

Advertisement
Advertisement

Matching Jobs

Comments on Too Nice to Land a Job

  • What needs changing
  • Posted by Grace , ret. prof/english at Butler univ on November 10, 2010 at 7:15am EST
  • "Hebl said that the implications of the research for those writing letters of recommendation are clear: stay away from communal words, whether writing on behalf of men or women."

    The central implication--more clear and more urgently needing to be addressed--is the pervasive sexist stereotype that still operates in the academy.

    This is important research. To conclude that what is needed is the mere tweaking of letters is less than "agentive."
  • Posted by Richard on November 10, 2010 at 7:45am EST
  • What is more interesting to me is that the words used more often to describe women contain the qualities many if us believe we embody as an academic community. Yet the words more often used to describe men, who we hire more frequently, reflect a corporate model which many academics object to and refute when such comparisons are made.
  • Posted by Milton on November 10, 2010 at 8:15am EST
  • Here is what is sexist: assuming that words such "caring," "sensitive," "compassionate, or a "supportive colleague" convey a gendered view of a given job candidate. I have had female students and colleagues who are "caring," "sensitive," etc., but I have also had male students and colleagues have the same attributes. I have also had female and male students and colleagues who were very far removed from these charactersitics. Why assume that when one sees these words in a recommendation letter that the reader is, consciously or subconsiously, feminizing the candidate about whom the letter is being written?

    This research is no doubt helpful in showing letter writers what kind of words might turn off potential academic employers, but it is important to consider other, nongendered interpretations of these data.
  • Posted by Wendy on November 10, 2010 at 8:15am EST
  • So interesting, though not surprising. I can see the next recommendation letter.

    She is an aggressive supporter of colleagues, a confident, daring team player, and a nurturing bitch.

    Another example of how much harder women have to work to navigate a career.
  • more importantly...
  • Posted by Mike on November 10, 2010 at 8:15am EST
  • Even more crucial than the gender bias is the bias against hiring ANYONE, at least if your field is in the humanities. No matter what the rec letter says, the vast majority of candidates will not get a job. So if the vast vast vast majority of men don't get a job, but the vast vast vast vast majority of women don't get a job, that's actually, IMHO, a lesser concern than the fact that most men, women, Martians, Earthlings, etc., won't get a job.
  • Everything is Gendered
  • Posted by Curro Romero on November 10, 2010 at 9:00am EST
  • Gender is hierarchical, ever since patriarchy began.

    Derrida: Think of word pairings. Most of the time the privileged term will be on the lefthand side of the slash mark (in Western languages): hot/cold, day/night, open/shut, on/off, good/evil, male/female and so on. The term on the right is regarded as a mere supplement to the defining term on the left.

    These cultural prejudices run deep and have profound effects on society whose female members have paid a heavy price throughout the ages, notably by utter obscurity.

    To respond to the dismal hiring in the Humanities, gendering again: At my community college we have a math/science-heavy Liberal Arts Associates degree (L.A.S)and a humanities-heavy L.A.A. The attributes tracking sheet for the former is blue, the one for the latter is pink. Sends a message. I have complained about this to no avail.

    I think one reason the humanities, like women, are devalued is that they are regarded not just as a supplement, but as what Derrida called "a dangerous supplement," to be kept at bay, prevented from transgressing the boundary. For it is all too insightful. Agentive.
  • Let's talk about the causal agent...
  • Posted by Hoosier Prof on November 10, 2010 at 9:00am EST
  • This is a fascinating and worthwhile study, and the implications are alarming. But I have real issues with the conclusions drawn by the researchers as to WHY women whose rec letters are more emotive are ranked lower.

    First, from a methodological point of view, researchers who interpret on their own why a respondent acts as it does only impose their own disciplinary frameworks and personal biases. They don't necessarily answer the question correctly. So I do wonder why they didn't turn over the results to the respondents to seek help with the interpretation.

    I suspect that the SCs would have explained that many set their screening mechanisms to favor those oriented toward research and scholarship over teaching. In other words, they could worry that the women they hire are going to be spending too much time on teaching (I cringe as I write this, but you all know someone like it) where these emotive words would be perfectly appropriate.

    I also must observe that I rarely see those particular adjectives used in ACADEMIC rec letters regardless of gender. They seem more appropriate to the business sector. So, again, who picked out the words to use in this experiment?

    And just for the record, I'm not sanctioning or excusing any search committee gender bias, just pointing out that the interpretation seems rather stale.
  • Letters of rec matter little in real decision making
  • Posted by skeptical , professor of social sciences at R1 university on November 10, 2010 at 9:30am EST
  • I think this has the makings of a tempest in a teapot. I've been on three search committees in the past 5 years, and letters of recommendation are just about useless in making a hiring decision. They universally rave about the candidate; I'm still trying to figure out the statistical odds of every single candidate being "one of the best graduate students I've had the pleasure of mentoring in my career." Perhaps it is different across other disciplines or at other universities, but I can state with confidence that no candidate in any search I've been involved with has ever been demoted from or promoted to a search list on the basis of the adjectives included in a letter of recommendation.

    If I were Czar of Academia, in fact, one of my first decrees would be to require letters only for those individuals on the short list, and then just as the t-crossing and i-dotting formality that they largely are.
  • Posted by S on November 10, 2010 at 10:30am EST
  • Well. I guess this is why so many of the faculty I knew in grad school lacked compassion and an awareness of community needs, and why they usually couldn't get along even with each other.

    I thought those attributes were coincidental; I didn't know they were specific job requirements.
  • sorry, just one more point.
  • Posted by Hoosier Prof on November 10, 2010 at 10:30am EST
  • Skeptical, do spend just a little more time in the trenches before you decide rec letters don't matter based on a sample of three search committees.

    As you develop more knowledge of their variety, you would be amazed at how well rec letters inform the process. Every SC member has his/her own favorite choice of "telling factor" in the screening. For me and many of my colleagues (we are hiring for social science positions at an R1 school), it's the job talk. But second, I place the rec letters. Faculty who are good at writing them can convey in subtle ways how they rank a candidate among others.
  • Hmmmm
  • Posted by Male Prof on November 10, 2010 at 10:30am EST
  • Having written many such letters, and having read many more, this really surprises me. We typically skim these letters and don't give them much weight. I am also in a college where male professors are in the minority. But this research will no doubt make me more conscious of the adjectives I use.
  • Letter of Rec tend to be "Pass/Fail"
  • Posted by James Morgan , Assistant Professor at Illinois on November 10, 2010 at 11:00am EST
  • Study makes some good points re: language selection. As noted by others, if you are looking for research productivity, tounchy-feely descriptions are not a positive.

    More to the point, at every search committee I'm been on, the leters of Rec tend to be used as "Pass/Fail" in terms of who wrote them, rather than an intensive comparision of what is written. Who is going to have a BAD Letter of Rec.?
  • huh?
  • Posted by Howard on November 10, 2010 at 11:00am EST
  • Here's another explanation: women really are more caring, sensitive, helpful, supportive. Letters of recommendation reflect this. And readers of such letters, looking exclusively or almost exclusively for people who will privilege research over teaching, especially time-consuming nurturing teaching, downgrade the candidate.
    Since letters of recommendation are filled with praise and enthusiasm, readers are looking for subtle cues. These are it.
  • Letters of recommendation do matter
  • Posted by Al Beitz , Professor/Dept Veterinary and Biomedical Sciences at University of Minnesota on November 10, 2010 at 11:45am EST
  • As a department Chair, I agree with the Hoosier Prof that letters of recommendation do matter. Sure most letters provide glowing assessments of the candidates, but adjectives do matter. We want to hire candidates with exceptional research backgrounds and teaching skills, but we also want candidates that are independent, collaborative, trustworthy, ambitious, enthusiastic, dependable, etc. It is very difficult to evaluate these characteristics from CVs and thus letters of recommendations play an important role. I would argue that letters of recommendation as well as the candidates own letter of intent augment the CV and enter into the decision making process that narrows the applicant pool. We typically do a video conference with our top 6-10 candidates and this allows us to narrow down our selection to our top 3 or 4 whom are invited into interview.
  • More opposites
  • Posted by RW at Big10U on November 10, 2010 at 11:45am EST
  • "...the privileged term will be on the lefthand side of the slash mark (in Western languages): hot/cold, day/night, open/shut, on/off, good/evil, male/female and so on."

    You mean like black/white, ladies/gentlemen, old/new, wet/dry, aunt/uncle, cat/dog and so on?
  • Gendered Terminology?
  • Posted by Dr. F. Gump on November 10, 2010 at 2:45pm EST
  • Are the terms really gendered or as some have suggested, some attributes really are preferred? I'm amazed at the number of students who do not know the difference between assertive and aggressive; faculty too.

    People who aren't able to be forceful and task-oriented without stepping over the line to aggressive behaviors may be those who immediately cry foul and look for a non-related reason they weren't hired?
  • Posted by Mordaunt , Professor at UNO-Nebraska on November 10, 2010 at 2:45pm EST
  • . . . an issue to float around in my Sociolinguistics class but one I do not think has any effect on hiring in my department where there are more women faculty members than men and only 3-4(?) are so-called minority faculty. What I heard from time to time was whether the person being considered for the job would be a person who would work well with faculty--a case in point was a male applicant who was never hired. Perhaps what is happening a
  • Hiring Practices
  • Posted by Mordaunt , Professor/English at University of Nebraska at Omaha on November 10, 2010 at 2:45pm EST
  • ... an issue to float in my sociolinguistics class but one that I believe has no effect on hiring a department where there are more women faculty than there are men and where there are roughly 3-4 (?) so-called minority faculty. Maybe there has been a dramatic change in hiring pratices from what applied some 20 years ago. What I have heard all the time I have been here is whether the person being considered for the job would work well with the faculty--a case in point was a male applicant. Perhaps the difference lies in the fact that we are not considered "flagship" or whatever. Should we then be concerned for our MA graduates who are moving on to work on Ph.Ds. in the most powerful institutions in the country and perhaps hope to find employment there?
  • bad choice of letter writer
  • Posted by MTEH , Modern Languages and Literatures at University of Texas, San Antonio on November 10, 2010 at 2:45pm EST
  • As a woman trained at a research institution who has written many letters of recommendation for women, I would never ever talk about someone's ability to be supportive or nurturing. In this cut-throat academic market, talking about supportive qualities is about as relevant as talking about someone's physical appearance. If women are getting described in such a way more so than men, I would suspect that women are choosing bad letter writers more often than men are. Anyone keeping up in academia knows that you should not waste time talking about supportive-ness. Someone once told me that praising a student for his or her "efforts" was code for saying that the student was incapable of accomplishing. I would have to suspect that a letter writer talking about supportive qualities is either somewhat out of touch with academia or else is writing code to cover up the fact that he or she really does not want recommend the candidate. Either way, to me, it comes down to these so-termed "supportive" candidates having chosen poorly their letter writers.
  • RW
  • Posted by Curro Romero on November 10, 2010 at 3:00pm EST
  • Exactly. As Derrida stresses, and as I was careful to include, "most of the time" the privileged, defining term is spoken or written first, and as such is acted out within the culture.
  • Sometimes it happens
  • Posted by Brooke , Postdoc at R1 on November 10, 2010 at 3:45pm EST
  • I applied for a position once and asked a female colleague to write a letter for me. After she sent the letter to the search committee chair, she sent me a copy for "my files." I was shocked to discover that she had made reference to my amazing ability to multitask, attend conferences, and publish, etc. despite having a baby to look after. Needless to say, I never heard from that particular search committee again.
  • Posted by RJ on November 11, 2010 at 7:30am EST

  • I think of myself as supportive, nurturing, etc. and I think that these are strengths of mine and have made a tremendous difference to my students.

    I am at an R1 and these qualities are in short supply among my colleagues. The culture definitely devalues them.

    I am not hiding them, I think the world would be a better place if more people in academia had and used these qualities - even if that meant more time with students and gasp - fewer submissions to top tier journals.

    I won't last here, but I am not changing my fundamental values to fit into a culture I don't believe in.

    RJ
  • I’m Surprised That You’re Surprised
  • Posted by Frizbane Manley on November 11, 2010 at 11:15am EST
  • The findings of the Hebl/Martin/Madera research seems so “old hat,” I’m wondering, “who didn’t know this already.”

    I’m sorry I can’t show you the entire letter (too long), but here are a few excerpts from a letter I fired off for a student of mine who recently applied for (and got) a mathematics position at an R1 ...

    “Dear Jack:

    I’m writing this letter of recommendation on behalf of Bartholomew Musselman -- we call him Big Bart around the Department -- who is clearly one of the strongest graduates we have had in some time. His self-confidence and energy are astounding ... confront him with a challenge and he’ll knock it on its ass in no time flat. And respect? ... everywhere he sets his spurs he commands respect ... and without even trying. He presented a revolutionary, never-seen-before, eight-line proof of the Brouwer Fixed-point Theorem at our graduate seminar; and although the proof contained a significant error, the audience rose in unison with shouts of ‘Bravo!, Bravo!!’ Some even conjectured that he included the error to keep our junior faculty on their toes. ...

    As a teacher, he takes no prisoners. I once watched him knock off a proof of the Heine-Borel Theorem in a manner that had the guys cowering under their desks and had the girls weeping in their hands. Upon completing the proof (which took more than 30 minutes), he required the students in the class to open their blue books and repeat all of his steps from memory.

    The first sentence of his syllabus for Introductory Topology states, “By next Wednesday, find a copy of Nicolas Bourbaki’s ‘Topologie Générale,’ read it from cover to cover, and find three errors never previously discovered. When I asked if that wasn’t expecting a little much he responded, ‘Monday is Labor Day, so they have a long weekend that should give them enough time to complete the task.’ Wow ... what a man! He’s the kind of guy you hope your daughter will bring home for Thanksgiving break in order to even up the sides on the tag football teams. ...

    Big Bart is not intimidated by instructional technology. In fact he conducts out-of-class, not-for-credit workshops on Saturday and Sunday mornings in which he teaches his students everything there is to know about LaTex and Mathematica ... and all on his own time. He’s a hoss! ...

    His research will bowl you over ... I mean literally knock you off your feet ... I mean blister your butt. His latest publication was a 20-page tour de force in ‘The American Political Science Review’ that reportedly had the highest ratio of equations to words with political content of any article in the history of that journal. He’s the kind of guy about whom the students and his colleagues exclaim, ‘He’s brilliant! In fact his mental powers are so great he has difficulty bringing them down to a level we can understand.’ ...

    While BB refuses to perform service activities for the Department and the University, no one cares. Here at FU, we have always respected his genius, and we have thought it would be non-optimal to distract him from his hard-hitting, intellectually dynamic, research blockbusters. ...”

    Well, my letter goes on for almost three pages. As is true of every graduate advisor or distinguished professor, I write a variation of this letter for each of my students, changing a word here or there, and inserting different events, illustrations, and adjectives. It’s those minor changes, of course, that enable me to distinguish between my brilliant and incredibly talented students on the one hand and those who are merely exceptional on the other.

    For those of you who struggle with writing effective letters of recommendation, you can find the complete letter on my web-site. Feel free to use it to your heart’s content as your boilerplate letter of recommendation.
  • I am what I am
  • Posted by Ken , Physics Professor on November 11, 2010 at 11:45am EST
  • I have been called many things. Supportive is quite common. I work better as part of a large team than as an island. "Supportive" is a good thing when you want everyone to help everyone develop. It is bad when you want extreme competition between everyone. Each area has its own preference. I tend to see more of the former in science and more of the latter in liberal arts. Perhaps physical science research requires more mutual support than is needed in liberal arts careers. Compare based on what the job involves before making general conclusions. Different instructor positions involve different responsibilities. Different departments have different "environments".
  • Women don't deserve it
  • Posted by Beth on November 11, 2010 at 4:00pm EST
  • Time and time again, women have proven they are unqualified to meet the demands of the male working environment. Feminism has seduced women into thinking they are smarter than they actually are. Finally we can see F U to feminist nonsense: http://manhood101.com
  • Is this research relevant to college admissions?
  • Posted by TP , Director of College Counseling at private girls' high school on November 11, 2010 at 4:00pm EST
  • At a time when the freshman applicant pool at most liberal arts colleges is heavily female, I am very aware that the girls applying for admission are at a distinct disadvantage in the selection process. Reading this article made me wonder if our letters of recommendation should emphasize the applicants' "male" qualities over the caring and consensus- building characteristics that would help to create a campus climate of cooperation rather than competition.
  • Thank you, Frizbane Manley!
  • Posted by JOX on November 12, 2010 at 12:45pm EST
  • LMAO. Whew! That was funny.
  • Posted by MR on November 12, 2010 at 12:45pm EST
  • I agree that the research university culture devalues these qualities. I wouldn't mention them in a letter, but I am now having second thoughts about a letter I recently wrote for a candidate who had been involved in many, many collaborative research projects, which is unusual in the humanities. This is in part because she is one of the few experts in the use of technology in this humanities field, so she is regularly consulted and invited to be part of larger research teams. I pointed out that these projects attested to her "strength as a team player" in the field. At least the comment was about her professional work rather than her personality, but I now wish I hadn't have said that. Ugh.
  • Reference words
  • Posted by newEdD , Instructor at MCC on November 13, 2010 at 7:00pm EST
  • With the large push towards civility in academia, one might think communal attributes might be valued more than agentive descriptors of our faculty. Perhaps the pendulum will swing back towards a more collegial environment as we carefully choose descriptors in our everyday conversations and in how we assist our colleagues to find the best fit in the academic environment or elsewhere. Let's not only hope so-let's do so!
  • A Closer Look at the Evidence
  • Posted by William Clark , Professor/ Political Science at University of Michigan on November 13, 2010 at 8:30pm EST
  • A closer look at this paper suggests that the evidence for bias is far from clear cut. See my post <ahref="http://www.themonkeycage.org/">here</a> for details.