Search Committee Process

[dismalist]

Reducing certain information about candidates so as to decrease bias is really a mechanism to increase efficiency, and if workable, would not increase diversity.

That circle cannot be squared.

It would reduce implicit bias, but I suspect that that's not the only impediment to diversity in the academic workforce.

Seriously, what then is?

[dismalist]

Unfortunately I don't think you could hide the gender of the applicant given the gendered language in recommendation letters.  Beyond the "she, his" etc language there is huge bias in how recommendation letters are written.  I've served on search committees and seen this for a high percentage of female applicants at all career stages.  I think the committee would be better off knowing the gender of the applicant and putting the rec letter language in context than ignoring this.  If you read one package about how brilliant the applicant is and other about how hard working/kind the person, wouldn't you select the brilliant scholar (all other aspects of the applications being constant)?  I would rather be in a position to put the rec letter language in context than try to be blind to this.

For what it's worth, I ran the gender bias calculator,

https://www.tomforth.co.uk/genderbias/

on two letters of recommendation that I wrote and the amount of gendered language I used did not correlate with the gender of the person I was writing the letter for, because the female candidate was a postdoc I worked closely on in terms of research, and the male candidate was a postdoc with whom my main interaction related to his teaching responsibilities, and the calculator was essentially picking up on whether I was focusing on research or teaching in the letter.

Clearly, the gender bias calculator is -- biased! :-)

[the_geneticist]

So what type of diversity question(s) you 'd suggest to be included in the rubric?

Thanks,

You could have a "statement of commitment to diversity" included in the application package.  Specifically, how will the candidate uphold the mission of the university/department in supporting students/removing barriers/creating new opportunities for historically marginalized groups?  How will their research program address these issues?  Most NSF grants have a specific "outreach" component and just saying "'I'll hire some undergrads to help in the lab" doesn't cut it anymore. 
On my most recent hiring committee for a teaching job at our HSI (also high 1st to college, high financial need), we could easily eliminate some candidates right away due to their abysmally bad diversity statements.  Things like "all students are the same" or "I don't worry about skin color, I just teach my class to everyone" or "I think girls can learn too" are archaic at best.  The best candidates gave specific examples of how their teaching practices will support our very diverse student body (like making sure they use examples by female authors, Black scientists, issues of local interest, etc.).

[mleok]

This is an example rubric from Berkeley for assessing diversity statements,

https://ofew.berkeley.edu/recruitment/contributions-diversity/rubric-assessing-candidate-contributions-diversity-equity

[spork]

[. . . ]

Sure. Bias against anybody could be a result of animus. That's pretty much gone, certainly in our circles.

[. . . ]

We are in different circles.

More thoughts:

• Recommendation letters should not be required as part of the initial application. In many cases they don't provide unique, useful information anyway. If they aren't required up front, no need to blind them when screening a large applicant pool.
• At best, "diversity statements" simply duplicate information that can be found in a well-written "teaching philosophy statement." At worst, they are empty exercises in virtue signaling.
• If a university or department is operating on the opinion that "diversity" is good, then less visible attributes of diversity (economic class, religious belief) ought to be weighted just as heavily as the more visible attributes (gender, race, ethnicity, nationality). And if a university or department is truly interested in diversity, then that means looking at the structure and delivery of the curriculum, regardless of what kind of person is hired to teach it.

[Langue_doc]

[. . . ]

Sure. Bias against anybody could be a result of animus. That's pretty much gone, certainly in our circles.

[. . . ]

We are in different circles.

More thoughts:

• Recommendation letters should not be required as part of the initial application. In many cases they don't provide unique, useful information anyway. If they aren't required up front, no need to blind them when screening a large applicant pool.
• At best, "diversity statements" simply duplicate information that can be found in a well-written "teaching philosophy statement." At worst, they are empty exercises in virtue signaling.
• If a university or department is operating on the opinion that "diversity" is good, then less visible attributes of diversity (economic class, religious belief) ought to be weighted just as heavily as the more visible attributes (gender, race, ethnicity, nationality). And if a university or department is truly interested in diversity, then that means looking at the structure and delivery of the curriculum, regardless of what kind of person is hired to teach it.

+1

"Diversity" and "inclusion" are merely buzzwords. Just because an applicant can discuss these concepts in their application materials and in the interviews doesn't necessarily mean that the person can teach a class that includes students from a variety of backgrounds, ethnicity, and religious beliefs.

[marshwiggle]

• Recommendation letters should not be required as part of the initial application. In many cases they don't provide unique, useful information anyway. If they aren't required up front, no need to blind them when screening a large applicant pool.

Good point. Isn't part of the point of recommendation letters the subtle value of who writes them? If so, that's precisely the kind of subtle bias that could be eliminated.

• At best, "diversity statements" simply duplicate information that can be found in a well-written "teaching philosophy statement." At worst, they are empty exercises in virtue signaling.

It has occurred to me that ideas like being "anti-racist" sound really bizarre if you talk about people being "anti-murder" and "anti-child-pornography" with the implication that not being in some way "actively" against those things makes you for them. Virtue signaling implicitly indicates that there are few enough situations in ordinary life which will demonstrate one's virtue organically that they have to be manufactured.

In a war zone or a natural disaster, heroism doesn't need to be "signaled".

• If a university or department is operating on the opinion that "diversity" is good, then less visible attributes of diversity (economic class, religious belief) ought to be weighted just as heavily as the more visible attributes (gender, race, ethnicity, nationality). And if a university or department is truly interested in diversity, then that means looking at the structure and delivery of the curriculum, regardless of what kind of person is hired to teach it.

Absolutely! Looking down my short street, with several different ethnicities/nationalities represented, I probably have much more in common with my neighbours than with other people who are native born like me but live in small rural areas, or in in public housing in my city.

I'd guess I'm not unusual in that.

[Ruralguy]

I'm inclined to agree that diverse means (or should mean in this context) more than just being of certain under-represented minorities. Now if a school just comes out and says "we need more African-American faculty because that will help us retain African American students" then I'd have respect for that goal, especially if there is information suggesting that the goal (hiring) is likely to lead to the desired outcomes for students (i.e., staying and graduating). But stating a specific goal is different than just using general diversity buzzwords.

[marshwiggle]

I'm inclined to agree that diverse means (or should mean in this context) more than just being of certain under-represented minorities. Now if a school just comes out and says "we need more African-American faculty because that will help us retain African American students" then I'd have respect for that goal, especially if there is information suggesting that the goal (hiring) is likely to lead to the desired outcomes for students (i.e., staying and graduating). But stating a specific goal is different than just using general diversity buzzwords.

Question: Does anyone know of studies providing evidence of faculty makeup affecting retention of minority/underrepresented  students?  I could see it potentially affecting recruiting, but it's hard to imagine the performance of struggling students being improved significantly by prof identity *alone.


(Obviously, if department faculty changes are accompanied by changes to curriculum, etc., then the effects could be significant.)

[Ruralguy]

I don't have such data, but the claim has been made. I can look for it, but others might have better access, assuming these data exist.
Its plausible due to the effect culture can have on whether you seek help from the professor or even select that professor in the first place. But I can also see why it might have minimal effect.

Of course, you'd need baseline data on students broken down by race (or other factors, such a first gen student, etc.). We've seen the evidence for positive effect on recruiting at my school , but that's more about having the right staff or faculty (mostly staff for us) getting out there to the high schools or speaking with them on visits and follow up calls.

[ciao_yall]

I'm inclined to agree that diverse means (or should mean in this context) more than just being of certain under-represented minorities. Now if a school just comes out and says "we need more African-American faculty because that will help us retain African American students" then I'd have respect for that goal, especially if there is information suggesting that the goal (hiring) is likely to lead to the desired outcomes for students (i.e., staying and graduating). But stating a specific goal is different than just using general diversity buzzwords.

Question: Does anyone know of studies providing evidence of faculty makeup affecting retention of minority/underrepresented  students?  I could see it potentially affecting recruiting, but it's hard to imagine the performance of struggling students being improved significantly by prof identity *alone.


(Obviously, if department faculty changes are accompanied by changes to curriculum, etc., then the effects could be significant.)

You could search the ERIC database. Lots of research about the importance of representation in student retention.

[mleok]

I'm inclined to agree that diverse means (or should mean in this context) more than just being of certain under-represented minorities. Now if a school just comes out and says "we need more African-American faculty because that will help us retain African American students" then I'd have respect for that goal, especially if there is information suggesting that the goal (hiring) is likely to lead to the desired outcomes for students (i.e., staying and graduating). But stating a specific goal is different than just using general diversity buzzwords.

Yes, if you look at the rubric for evaluating diversity statements I posted, the emphasis is on activities that have the potential to impact student outcomes, as opposed to the candidate being of a certain under-represented minority. As I've said before, focusing on potential benefits to student outcomes is more likely to yield broader support for increasing the diversity of the faculty.

[pgher]

I've been watching this thread with some interest. Hiring is such a difficult task. I think it's important to remember that a job is not a reward for previous good work but an opportunity to do future good work.

[research_prof]

I've been watching this thread with some interest. Hiring is such a difficult task. I think it's important to remember that a job is not a reward for previous good work but an opportunity to do future good work.

How can you evaluate that without considering past work and track record?

[dismalist]

I'm inclined to agree that diverse means (or should mean in this context) more than just being of certain under-represented minorities. Now if a school just comes out and says "we need more African-American faculty because that will help us retain African American students" then I'd have respect for that goal, especially if there is information suggesting that the goal (hiring) is likely to lead to the desired outcomes for students (i.e., staying and graduating). But stating a specific goal is different than just using general diversity buzzwords.

Yes, if you look at the rubric for evaluating diversity statements I posted, the emphasis is on activities that have the potential to impact student outcomes, as opposed to the candidate being of a certain under-represented minority. As I've said before, focusing on potential benefits to student outcomes is more likely to yield broader support for increasing the diversity of the faculty.

Yeah, the rubric pushes toward having diverse kids graduate. Save a lot of work and just graduate the kids!

Hell, none of this could make sense if a lot of higher education weren't merely signalling.