ASTRO Goes All In with Equity and Diversity:
ASTRO new Summer 2018, Vol. 21, Number 2
I am so sad and disturbed to see ASTRO going this route, especially in a time where dissent is not permitted on this issue. I wish I didn't have to post all this non-sense, but is there anywhere else on the internet that has collated the equity & diversity push in oncology?
One paper we see cited again and again is the Penner paper (I cited even here and
now!), as well as the advocacy from Drs. Jagsi and Winkfield.
Penner etl al, note the following "The IAT is the most widely used measure of implicit bias and is extensively validated.
22,
46,
47 "
References 22 and 47 are from the makers of the IAT themselves, Banaji & Greenwald, (talk about a COI), while reference 46 is a small study (n = 61) from a German student sample. This is NOT extensive validation, but is being used as definitive evidence for oncologists' implicit bias.
Here is a meta-analysis by someone who did not invent the test:
Oswald, F. L., Mitchell, G., Blanton, H., Jaccard, J., & Tetlock, P. E. (2013). Predicting ethnic and racial discrimination: A meta-analysis of IAT criterion studies. Journal of Personality and Social Psychology, 105(2), 171-192.
and further criticism from the same group:
Oswald, F. L., Mitchell, G., Blanton, H., Jaccard, J., & Tetlock, P. E. (2015). Using the IAT to predict ethnic and racial discrimination: Small effect sizes of unknown societal significance. Journal of Personality and Social Psychology, 108(4), 562-571.
Perhaps most striking is a statement from the authors themselves in a reply to the Oswald et. al's 2013 meta-analysis:
https://faculty.washington.edu/agg/pdf/Greenwald,Banaji&Nosek.JPSP.2015.pdf
"Identifying likely perpetrators of discrimination. IAT measures have two properties that render it problematic to use them to classify persons as likely to engage in discrimination. Those two properties are modest test–retest reliability (for the IAT, typically between r = .5 and r = .6; cf. Nosek et al., 2007) and small-to-moderate predictive validity effect sizes
. Attempts to use such measures diagnostically for individuals therefore risk undesirably high rates of erroneous classifications.8 These problems of limited test–retest reliability and small effect sizes are maximal when the sample consists of a single person (
i.e., for individual diagnostic use), but diminish substantially as sample size increases. Limited reliability and small-to-moderate effect sizes are therefore not problematic in diagnosing system-level discrimination, for which analyses often involve large samples."
It is very hard to understand what is going on here, but definitively we can say that the test (on the authors' own admission) cannot identify any individual's implicit bias. So if you individually take the test, then it cannot reliable test your implicit bias.
I guess somehow it can tests groups of people (implicit bias as an emergent property of groups???), but not sure if this is true or is even logically coherent.
I believe this is simply a cop out and agree with Oswald when he notes "IATs were poor predictors of every criterion category other than brain activity, and the IATs performed no better than simple explicit measures."