2

This article claims that a study researching the effect prejudice/racism has on psychiatric treatment found that working class individuals were 70-80% less likely to be accepted as patients by a therapist if they were black then if they were white.

Are these numbers accurate? Did the study find a 70-80% difference based off of race?

Sklivvz
  • 78,578
  • 29
  • 321
  • 428
dsollen
  • 10,062
  • 12
  • 43
  • 71
  • That may depend on the country/region. Does your claim relate to any country in particular? – gerrit Nov 21 '16 at 15:05
  • @gerrit the article refers to therapists in New York, so lets say new york ;) – dsollen Nov 21 '16 at 15:13
  • The [study that is referenced](http://hsb.sagepub.com/content/57/2/168) was published in a peer reviewed journal. I don't have access to it right now to check if the New York Times summarized it correctly. – tim Nov 21 '16 at 15:30
  • Your paraphrase is wrong; that article does not say what you say it says. – Dan Getz Nov 21 '16 at 16:31
  • It might be relevant to ask whether a therapist can effectively treat someone from a much different culture. If the therapist doesn't believe that meaningful interaction is possible because of cultural differences, then wouldn't accepting such patients constitute fraud? – jamesqf Nov 21 '16 at 18:35
  • 2
    Let's try to avoid US-centric bias here: the way blacks are seen in the US is not the same as in Kenya, China, India or Europe. I've added a tag to this purpose, but let's remember in the future to specify. – Sklivvz Nov 21 '16 at 21:05

1 Answers1

4

In the comments, Tim helpfully links the original study (pdf).

Are these numbers accurate?

I went quickly through it and the answer is not quite.

Did the study find a [...] difference based off of race?

According to the paper, yes when considering callbacks.

enter image description here

No when considering overall chances of getting any appointment. (there is no difference among the working class, there is a difference within the middle class)

enter image description here

Images added according to fair use, as suggested by Oddthinking in the comments.

To answer the questions in the comment about "actual income":

Voice-over artists recorded scripted messages using racially distinctive names and adopting specified race- and class-based speech patterns.

To answer the question about insurance (emphasis mine):

On the recordings, the help seekers each mentioned symptoms of depression or anxiety, named the same health insurance plan, requested an appointment, and indicated a preference for a week-day evening.

Federico
  • 1,810
  • 3
  • 18
  • 25
  • 1
    Did the study control for (1) **actual** income and (b) more importantly, for health insurance coverage? Not all insurance covers mental health same way. – user5341 Nov 21 '16 at 15:52
  • Also, i never tried to get an appointment, but is it actually typical for a receptionist of a mental health professional to ask for the caller's race when appointment is requested? And if not, how would the proposed discrepancy even work? – user5341 Nov 21 '16 at 15:54
  • that addresses my second question (I have quibbles with such methodology, but they are irrelevant to your answer). However, my (b) question from the first comment is the really important one. Have they controlled for insurance when evaluating responses? I have a strong suspicion that therapists want to get paid. – user5341 Nov 21 '16 at 16:08
  • 34% of 45% is 75.6%, which is 70-80%. Changing "70-80%" to "70-80%" less can be easily achieved with bad reading comprehension.. – Peter Nov 21 '16 at 19:15
  • 1
    "not included a copy of the figure for copyright reasons." This falls under [Fair Use](http://meta.skeptics.stackexchange.com/questions/358/copyrighted-images). Go ahead and post a fragment, with a link. – Oddthinking Nov 22 '16 at 00:52
  • 1
    It is incorrect to say there's a difference unless we look at the uncertainties on the numbers. With only 160 people per group the experiment will have a very very low resolution. In fact, running the numbers gives a p-value of 0.043 which is barely significant (just one different response would make the p-value not statistically significant). I call BS on this. – Sklivvz Nov 22 '16 at 02:51
  • @user5341 answer edited – Federico Nov 22 '16 at 09:04
  • 1
    @Federico The answer is getting downvoted because it does not provide a fair representation of the paper. The author themselves note that figure 6 is weak evidence, _"The differences in callback rates were more muted than the effects presented in the accessibility analyses, suggesting that the discriminatory effect is not driven entirely by a lack of response"_, the paper is focused on other biases. – Sklivvz Nov 22 '16 at 09:08
  • @Federico if you are citing numbers, cite the relevant ones? :-) – Sklivvz Nov 22 '16 at 09:28
  • The answer is significantly improved, thanks! Who decides on the appointments? receptionists or therapists? (assuming the paper answered that). – user5341 Nov 22 '16 at 16:29
  • @user5341 you can now access the paper, see the pdf link – Federico Nov 22 '16 at 16:54