Chris Uggen's Blog: sentencing project report on racial disparity

Wednesday, July 18, 2007

sentencing project report on racial disparity

updated: 7/19

the sentencing project sent word today of their new report, Uneven Justice: State Rates of Incarceration By Race and Ethnicity. taking data from their tables, i graphed the state ratios of black-to-white incarceration shown below (note: this figure was revised 7/19 to correct a mislabeled state):

the disparity seems to be lowest in hawaii, though -- let's be clear about this -- a ratio of 1.9 still means that african americans are almost twice as likely as whites to be incarcerated in that state. southern states also have relatively low disproportionality ratios, partly due to their higher-than-average incarceration of whites. things are most disparate in iowa, vermont, new jersey, connecticut, wisconsin, and the dakotas, with african americans getting locked up at a rate 10 times that of whites. there is no state in which african american incarceration rates are anywhere near parity with white rates.

the report also computes ratios for hispanics versus non-hispanic whites, though i suspect that data quality varies considerably among the states on this indicator. nevertheless, i graphed these data as well:

comparing the two charts, the first thing i notice is the difference in scale on the y-axes: from 1.9 to 19 for the african american-to-white chart and from .4 to 6.6 on the hispanic-to-white chart. only connecticut, massachusetts, and pennsylvania had hispanic-to-white ratios of greater than 5. moreover, two states reached parity -- a ratio of 1.0 -- and five states had ratios indicating lower incarceration among hispanics than among non-hispanic whites: georgia, alaska, florida, arkansas, west virginia, louisiana, and hawaii. again, such ratios should probably be interpreted with a bit more caution than those presented in the first figure, since ethnicity is inconsistently reported in the criminal justice system.

the state-to-state differences are instructive and sobering, especially for northerners who might be smug or complacent about racial inequality. criminal punishment represents one area in which racial disparity appears far worse in the north than in the south, with mostly-white states such as connecticut leading the way in racial inequality. still, the overall disparities remain the big story: nationally, african american incarceration rates are 5.6 times as high as white rates, while hispanic rates are 1.8 times those of non-hispanic whites.


At 7:31 PM, Blogger Woz said...

Phew, quite a bit to digest here. But if I could point out that D.C. is a district and not a recognized state, Iowa is the nation's leader in incarcerating African-Americans.

Hey, at least we're on the map.

I know the Hispanic measure is probably highly inconsistent across geography, but is it possible that states such as Conneticut have such high ratios because of the low numbers of Hispanics? I thought for sure that places with heavy Latin@ immigration (southwest, Midwest) would have the higher rates.

But then again, I also thought the enlightened North would be faring far better...

At 7:57 PM, Blogger christopher uggen said...

right, woz. at least part of the story is about low population bases (e.g., african americans in iowa) and part of the story is about exceptionally low white incarceration rates (e.g., whites in dee cee and connecticut). the putatively enlightened north generally incarcerates less overall, but the social distribution of punishment appears far more uneven in that region.

At 12:17 AM, Blogger eszter said...

Why does Alabama show up twice on the first one?

At 10:47 AM, Blogger christopher uggen said...

nice catch, eszter! the mistake was mine. the data in the report are correct, but i had mislabeled alabama in constructing the figure. i checked and made a new figure this morning, which should accurately reflect the sentencing project's report.

At 11:40 PM, Blogger eszter said...

Thanks.:) I had also meant to mention that it's very interesting to see this, thanks for graphing it.

On a different note: "this figure was revised 9/19" is meant to say 7/19?:)

At 12:11 AM, Blogger christopher uggen said...

thanks, eszter, i figured a picture might be useful here. what month is it anyway? :)


Post a Comment

<< Home