1/n There's been some policy discussion of late in response to the fact that Hispanic and Black students are suspended at higher rates than Whites. Some have proposed a measure that would prevent such disparities, thereby enforcing an 'equality of suspension'. Before I argue...
2/n ...that such an initiative is both wrongheaded and counter-productive (to the extent that it would likely harm the learning environment of the 'behaved'), I first present some survey data of attitudes towards this issue and the policy itself.
3/n The only data I could find is limited to young (18-34) Americans, so just bear that in mind. The first thing to note are some disparities by race and party with respect to the source of the problem. Whites and Asians, for instance, are more likely to attribute the... 
4/n..problem to genuine behavioral differences than are Blacks and Hispanics. The latter, in contrast, are more likely than the former to attribute it to 'white privilege'. That said, the modal response across all groups (excepting Blacks) appears to be under-resourced schools.
Starker differences emerge when we examine by Party. Republicans are significantly more likely (45%) to attribute the disparities to behavioral differences, while nearly half of Democrats attribute it either to the insensitivity of white teachers or white privilege.
6/n As I demonstrate below, these differing attributions are (naturally) important for predicting support for 'equalized suspension rates'. But let's first look at the descriptives. Overall, a slight majority (56%) actually *opposes* such a policy, including a majority of...
7/n ..hispanics (61%), asians (55%), and whites (65%). A majority of blacks (56%), oth, appear to support it. By party, we find that the overwhelming majority of Reps and Independents oppose, while Democrats are almost evenly split.
8/n In the end, though, the strongest single predictor of support for the policy is attributing the source of the problem to insensitive White teachers and White privilege.
9/n Holding all other variables at their medians, moving from blaming poor schools/behavioral differences (33.3% probability) to blaming 'white teachers/privilege' (50.7%) predicts a roughly 17 % point increase in the likelihood of supporting enforced equalization.
10/n So why might this be a bad idea? Well, the problem is that it doesn't take into account the fact that there *are* indeed group differences in behavior. For instance, Black males score almost 0.5 of a SD higher on the NLSY's 'anti-social behavior' index than White males.
11/n Although scoring lower than their male counterparts, Black females also score over 0.4SD higher than White females. The differences are smaller when it comes to Hispanics, but they are significant nonetheless.
12/n Such may explain why a majority of Hispanics and a sizeable minority of Blacks oppose the policy: they *know* 'what the deal is' and don't want a situation where disruptive/anti-social behavior--which detracts from their learning-- gets a pass in the name of 'equality'.
13/n None of this should be taken to mean that suspension is the best way to deal with disruptive students (it may or may not be). The only point here is that there are consequences when public policies are informed by the ignorance of base-rates/group differences and the...
14/n knee-jerk inference of 'racism' from group disparities in negative outcomes.
15/n The End.
It appears that @cjprofman has actually published some research on this topic (for those that are interested in reading further) sciencedirect.com/science/articl…
@GlennLoury and @JohnHMcWhorter also addressed the topic in their latest podcast
