Tuesday, October 18, 2011

No one would talk much in society if they knew how often they misunderstood others (Goethe)

Anonymous asks here about potential interview bias/hiring against non-native English speakers:

What are the biases against those candidates for whom English is a second language? How much does one's accent influence her/his epistemic reliability as a scholar, as well as a potential teacher, in the context of the job interviews? Do some accents (e.g., of French speakers) make one sound more sophisticated and knowledgeable than others (e.g., east asian)? What kind of measures, if any, are taken to prevent this sort of bias?
I've known a number of philosophers on the market for whom English was a second language. Some were extremely and comfortably fluent, and it did not seem to affect their prospects (although they faced immigration/visa hurdles instead). So, I would imagine that fluency and how easily you are understood (or perceived to be understood) matter here. Just having a non-American accent (British or Australian, say) does not seem to be an impediment.

(As an aside, I spoke to an insurance agent on the phone the other day, and she had a very, very pronounced Southern accent, and I really struggled to understand her. I wonder if there may be biases against certain regional American accents?)

As American universities admit growing numbers of international students, this becomes a relevant issue for those students if they hope to pursue an academic job in the US. Is the deck stacked against them if they have an accent?

~zombie

26 comments:

Christopher Hitchcock said...

I wouldn't be surprised if there were conscious or unconscious biases against Southern (USA) and Appalachian accents. It still seems to be considered acceptable to adopt such an accent if one wants to feign being stupid or ignorant (or make jokes about marrying one's sister). One sees this trope regularly, e.g. on Jay Leno or Saturday Night Live.

I also predict that a Minnesota accent will quickly acquire a similar status.

Anonymous said...

I think your example of the insurance agent with the southern accent is telling. On the one hand, there is the problem with accents that they can be hard for many to understand, and foreign applicants can often be penalized for that (I have seen my department count this against applicants on the assumption that it would be an impediment in the classroom). Then there are biases against certain accents because they are associated with lack of intelligence or sophistication. Sometimes, though, these two issues might be hard to pull apart.

With the latter case, though, I wonder if certain American candidates aren't at a potentially greater disadvantage than foreigners, because members of search committees wouldn't have the relevant cultural knowledge about various accents. For example, someone with a really strong southern accent is likely to be implicitly taken as a hick by many Americans, but those same Americans would not be able to tell the difference between, say, a Swabian accent and a high German accent, and thus take the Swabian to be the hick...

In any event, I think non-native speakers should probably worry much more about being understood than they should worry about cultural biases.

Anonymous said...

I've worked with a couple of search committees where it was pretty much a given that women and minorities would not be hired unless trying to meet a quota. "Lack of communication skills" sometimes came up as a reason to reject such applicants

Anonymous said...

I have a very pronounced great Lakes/Rust Belt accent (think the old "Superfans" sketch from SNL) with a slight dusting of "New York Jew." Clearly any job is mine for the taking...

Anonymous said...

This seems like a question that can be answered via empirical research. While I don't know of any that has been done on interviews, there's a wealth of literature in pedagogy. (A Google search will turn up many results; see references contained in http://repository.upenn.edu/cgi/viewcontent.cgi?article=1192&context=gse_pubs&sei-redir=1 )

Just having a non-American accent (British or Australian, say) does not seem to be an impediment.

I take it that British (and British-sounding-to-American-ears) accents are actually a plus to many. So the original questioner's concern with different kinds of accents having different kinds of influence is certainly salient here.

Another thing worth noting, in the context of the interview, is that perceptions of accents can be affected by more than the sounds. Rubin (1992) found that even with the same audio track, an East Asian-looking instructor is judged to have a more foreign (presumably in a bad way) accent.

WV: "cries"

Anonymous said...

In any event, I think non-native speakers should probably worry much more about being understood than they should worry about cultural biases.

Sure, but given that we have good reasons to think that these cultural biases exist, shouldn't search committees try to be more aware of them and try to mitigate their own implicit biases?

Anonymous said...

Anon 12:59 PM: Haven't you paid attention? According to ZERO biases are *just* part of being human (hint, hint there's nothing we can do about them). How naive of you to expect philosophers to at least try to be epistemically responsible.

YFNA

Mr. Zero said...

According to ZERO biases are *just* part of being human (hint, hint there's nothing we can do about them).

There's a very simple procedure you could perform that would confirm or disconfirm this contention. You could look it up. Psychology done from the armchair is psychology done poorly.

How naive of you to expect philosophers to at least try to be epistemically responsible.

I didn't suggest that people should not try to be epistemically irresponsible. I suggested that people, when they tried, would be at least moderately unsuccessful. This is because people are animals with imperfectly designed intellectual organs. Sorry to have to break it to you.

Anonymous said...

Alright Zero, yeah, what scientists says reigns even when they can't even remain consistent in their definitions. Whatever you say, bud. Put your faith in that, if you like. I'll stand by the powers of rational inquiry and analysis any day. Christ. I don't *really* want to get into with you, since I notice you persist in interpreting not only me but others in the most uncharitable way possible. Charity is a principle that helps you avoid arguing against a straw man -- just in case you were wondering, which from your prior comments, you seem to have no grasp on.

YFNA

Mr. Zero said...

yeah, what scientists says [sic] reigns even when they can't even [?] remain consistent in their definitions.

I've seen people dismiss science on the basis of some really flimsy excuses before, but I can't remember when I've seen one that flimsy. My hat's off to you, sir.

...since I notice you persist in interpreting not only me but others in the most uncharitable way possible.

Wait, I'm confused.

I took you to be making reference to this comment, in which I said, "I took Zombie's point to be that the biases exist and that being affected by them doesn't make you a bad person, it makes you a human being... ...even if it is in fact true that the biases should not exist. ...the fact that they shouldn't do this [i.e. make unwarranted inferences as a result of latent biases] has no bearing on whether they will. And that even a highly conscientious person who consciously disavows these inferences and takes care to avoid making them is going to be at least moderately unsuccessful. And so a wise applicant takes care to present herself in a way that acknowledges and reflects this reality."

I take this stuff seriously, so I want to be sure we're clear for the people following along at home. There are a couple of important points here worth emphasizing. One is that I am pointing out that latent biases exist and are characteristic of human thinking. These biases are latent, so they're not available to introspection, and it is not possible for a person to tell just by thinking about it whether she is influenced by them or not.

Another is that people cannot stop themselves from being influenced by them simply by making a conscious effort. When we are affected by these latent biases we don't know it and we cannot help it. Employing some version of the "ought implies can" principle, I infer that being affected by such biases is compatible with not being a bad person.

Another is that I clearly make no claim to the effect that people should not bother trying to avoid unreliable inferences. I point out that these biases are bad, not ideal, and that people shouldn't make unreliable inferences. I did not encourage complacency.

My point was, where the rubber hits the road, that stuff doesn't matter. People have the biases whether they know it or not, and they can't turn them off. So, I said, you should wear clothes to your interviews that won't make your interviewers be biased against you.

You paraphrased these remarks here as "According to ZERO biases are *just* part of being human (hint, hint there's nothing we can do about them). How naive of you to expect philosophers to at least try to be epistemically responsible."

Which one of us interpreted the other uncharitably, again? Because it seems like you're saying that I interpreted you uncharitably, and that's obviously total bullshit.

Anonymous said...

Leave Minnesota Alone. Michele Bachmann was born in Iowa. I find the Minnesota accent to have a charmingly rustic quality.

Anonymous said...

"I wouldn't be surprised if there were conscious or unconscious biases against Southern (USA) and Appalachian accents."

Looking over the Leiter rankings, I notice that only one program in the top 20 (UT Austin) is in what is commonly known as "the south" (east of Texas, as the southwest and CA coast are a different beast).

So any philosopher worth his/her salt should have ample time to drop a southern accent during his/her years in a top PhD program. ;)

Anonymous said...

No one claims you can just turn biases off at will. What people are saying is that there are institutional procedures that could compensate for the fact that, try as they might, people continue to have biases. Affirmative action is one.

Also, stereotypes and schemas can be changed--not at will, but gradually. So, perhaps people on search committees should casually watch more videos of East Asians who speak English quite well.

Anonymous said...

Hmmm...equivocation is a flimsy reason to reject an argument. OK. Thanks. I'll remember that.

I agree with much of what you are now saying Zero. But I disagree that we cannot do anything about our biases. I think you are confusing the tendency to have biases with the content of those particular biases. We may not be able to do anything about the former, but certainly we can do something about the latter. Well, unless you believe that the particular biases we do have are hard-wired.

I certainly don't think that people who have latent biases are bad. I have them myself. I do think that they can be influenced however supposing their content was influenced and shaped by something beyond hard-wired biology in the first place. Sure, it may be a long-term project, and sure it might take time.

You seem to be taking out of both sides of your mouth. You say ought implies can which suggests that you think we cannot do anything about these things. On the other hand, you insist that we should not be complacent. Which is it?

YFNA

Anonymous said...

A couple of other things: Zero, I won't search through the posts to find your question. However, at one point you asked why I was insisting on the application of a principle of charity, so I answered you, finally.

Also, about the science stuff: It's not like I don't think science is important, but I don't think that careful analysis and argument isn't part of being a good scientist. I guess my position on scientific inquiry is best expressed by Williamson's piece in the NYT on naturalism. So I guess I look at theorizing at an abstract level regardless of your topic of inquiry as similar or should be.

I see that kind of required abstract thinking as lacking in *some* of the scientific studies I read where the concept/s they are studying are not even clearly set out. Making sense of the results or drawing any conclusions from such work is rather difficult.

Question: I admit I haven't read the studies on latent biases, but it seems to me quick to conclude anything about the content of what those biases could be given different environmental factors impinging upon us at formative moments. Of course, ideally, if biases are tendencies to make judgments without sufficient evidence, then none of them would be epistemically justified, but there may be some that are more or less pernicious.

carp said...

Looking over the Leiter rankings, I notice that only one program in the top 20 (UT Austin) is in what is commonly known as "the south"

Chapel Hill.

Also, stereotypes and schemas can be changed--not at will, but gradually. So, perhaps people on search committees should casually watch more videos of East Asians who speak English quite well.

Perhaps people not on search committees should too. Tell you what, come on by and I'll interview you for an hour. At the end of the interview I will analyze your character deficiencies and give you twelve weekly exercises to make you a better person!

Mr. Zero said...

Hmmm...equivocation is a flimsy reason to reject an argument. OK. Thanks. I'll remember that.

You didn't reject "an argument" on the basis of an equivocation. In order to do that, you have to identify an argument and pinpoint an equivocation. You didn't do that. You rejected science on the basis of an alleged tendency of scientists to equivocate. Not the same thing. (Was that your principle of charity in action?)

I think you are confusing the tendency to have biases with the content of those particular biases. We may not be able to do anything about the former, but certainly we can do something about the latter. Well, unless you believe that the particular biases we do have are hard-wired.

I'm just looking at the research, which suggests that the biases are acquired early on and remain stable over time. If there was evidence other than your faith in human reasoning capabilities to think otherwise, I'd be happy to consider it.

And maybe I'm misunderstanding you, but in order to do something about the particular content of some latent bias, you would have to know which beliefs were the deliverances of this bias, and this information is not readily available. It is not possible to determine when a bias is affecting you, which bias is affecting you, or with respect to which belief, via introspection. They're latent. You can't tell.

You seem to be taking out of both sides of your mouth. You say ought implies can which suggests that you think we cannot do anything about these things. On the other hand, you insist that we should not be complacent. Which is it?

Again. The research I am familiar with suggests that the biases are stable over time and are basically unaffected by the acquisition of new beliefs or our attempts to compensate for them.

That doesn't mean I think we should just go along with them. There are things we can do to cope with the biases, even if we can't get rid of them. The most effective way to cope with them is to remove the information that triggers them. This is why reputable journals and conferences insist on anonymized submissions.

However, at one point you asked why I was insisting on the application of a principle of charity, so I answered you, finally.

No. I was not asking you why the PoC is important. I was accusing you of hypocrisy in that you insist that others employ charity when interpreting your remarks but you employ no such charity when interpreting the remarks of others. Like that remark right there.

It's not like I don't think science is important, but I don't think that careful analysis and argument isn't part of being a good scientist.

Who would deny that careful analysis and argument is an important part of being a good scientist? But you don't know what the argument is. You haven't engaged with the research at all. I said, there's research; look it up and you said oh yeah, scientists can't even keep their definitions straight. The idea that your position here is carefully analyzed and argued is nuts.

I admit I haven't read the studies on latent biases, but it seems to me quick to conclude anything about the content of what those biases could be given different environmental factors impinging upon us at formative moments.

This culture has been pervasively abusive toward women and minorities for a long time, though.

Anonymous said...

Perhaps people not on search committees should too.

For sure. Everyone watch videos of East Asians who speak English well!

(And people on the search committees: just because we think the procedure can be better doesn't mean you have a "character defect" or should take it personally.)

I'm just looking at the research, which suggests that the biases are acquired early on and remain stable over time.

That's just not true. See, for example, http://faculty.washington.edu/agg/pdf/Dasgupta_Gwald._JPSP_2001.OCR.pdf

Big D said...

Instead of disputing who's on the side of "science" here, maybe someone could cite/link to research bolstering their case?

Mr. Zero said...

Hi anon 9:09,

Thanks for the link. I haven't had time to read the article carefully, but my initial impression is that their results seem very similar to the framing effect. It seems like this study shows that the effects of having been "framed" linger in diminished form 24 hours post-framing. It does not show, for example, that coming to disavow an explicit bias reduces latent bias, or that latent biases are responsive to explicit reasoning about them, or even what happens on the third day.

So, my question is, does this study show that there is instability in (e.g.) the latent pro-white bias, or does it show that it is possible to manufacture situations in which two latent biases are made to act in opposition to one another in such a way that one (the framing effect) temporarily weakens the other (pro-white bias)?

Anonymous said...

Mr. Zero,

As I noted earlier, no one (I think) is claiming that explicit reasoning can reduce bias. What you call "framing" is important because this is where institutional and procedural changes can be most powerful. We want to manufacture interview situations (or whatever) where known biases are weakened.

Mr. Zero said...

As I noted earlier, no one (I think) is claiming that explicit reasoning can reduce bias.

I realize I'm not having this out with you, but I did take YFNA to be making a claim in this neighborhood. In an earlier thread I said something about how a conscientious person who consciously disavows biased attitudes and who has taken steps designed to minimize her latent biases would nevertheless be at least somewhat unsuccessful; YNFA disputed this, saying that the suggestion that we cannot do anything consciously to alter our biases was questionable and specious. YNFA brought this comment of mine up again in this thread, claiming that I had counseled against even trying to be epistemically responsible.

So although I guess I don't agree that nobody is making that claim, I definitely agree with you that a discussion about what procedures might be employed to reduce the action of latent biases in hiring would be more fruitful. And if it could be shown that interviewers could neutralize their pro-white, pro-male, pro-tall, pro-handsome, etc, biases by properly "priming" themselves before conducting the interviews, that would be both pretty interesting and worth promoting.

Christopher Hitchcock said...

I volunteer to have my picture used for priming against the 'pro-handsome' bias. I can't help with tall, white or male.

Anonymous said...

Nope. I wasn't making any such claim Zero. Just that they could be changed. I didn't say how.

YFNA

Anonymous said...

Zero,

I will admit that I was not explicit enough in what I meant, but I really did mean that we could likely influence these biases in indirect ways, and that we could likely influence their content. I still wonder about the study's implications -- is it that we are biased or that we must have these particular biases? As I indicated before, I can buy the latter but still not buy the former. I may, as a matter of empirical fact be wrong, but I think the contents of these biases are probably responsive to cultural and social influences and that if we could change those, as I said, slowly over time, then we could change our biases (I'm still not even sure we agree on the definition of a bias). This doesn't seem crazy to me. At least it seems less crazy than the idea that the contents of our biases are hard-wired or innate. But, Zero, you know more about the studies than I do. Perhaps this has just been a big misunderstanding.

YFNA

zombie said...

Possibly some sort of behavioral therapy involving aversive stimuli could create a pro-short, plain, non-white, female disposition. We might as well add pro-fat to that list too. And pro-over-the-hill. I volunteer to administer the electric shocks.

Here at my university, members of SCs are required to be "diversity literacy" certified. There is a published online list of who has and who has not been trained. I don't know how widespread such practices are. Hopefully, the training is more robust than the sexual harassment training I was required to undergo, which involved watching a badly acted video about what is and is not sexual harassment (not sexy at all!), followed by an online quiz. I did have to send a copy of my (online) certificate of completion to someone in HR.