Sunday, Mar 22, 2015 06:00 AM CST
“Race is being used to wreck the middle class”: The silent bigotry of America’s poverty politics
The rightwing media's true influence: Racial resentment is the "strongest predictor" of welfare spending in America
Progressives
have long argued that conservatives play up racial resentments to
undermine the welfare state. Conservatives tend to respond to such
accusations with their own charges of liberal “race baiting.” But
whatever right-wing voices might say on the matter, a new analysis of
the data suggests that racism does, in fact, strongly predict welfare
spending.
Using two surveys from 2008, Douglas Spencer and Christopher Elmendorf estimated racial resentment across the states by measuring responses to questions about the work ethic, intelligence and trustworthiness of blacks Americans. They estimate the portion of whites in each state that are in the top quartile of racial resentment (so a state with large amounts of racism would have more than 25 percent of whites espousing racially biased views).
I combined this with data
from the Center for Budget and Policy Priorities showing Temporary
Assistance for Needy Families benefits for a single-parent household of
three. As the scatterplot below shows, these two variables are strongly
correlated:
I asked Eric Stone, the lead data scientist at Smarterer and a former hedge fund manager and USDA statistician, to help me see if there was another factor driving this trend. He ran the numbers again, controlling for state revenues per capita (based on Tax Foundation data, averaged from 2008-2010), as well as the average ideology of citizens and the average ideology of the state government (based on data from Richard Fording, again averaged over the period of 2008-2010).
Stone tells me that after controlling for these factors, “racial resentment is indeed the strongest predictor of TANF, at least in the context of this model.” He reports a p-value of <.001 (indicating there is a less than .1 percent chance that the relationship was observed by chance) and an R2 of .556 for the model (meaning that these variables collectively explain more than half of the variation in TANF spending).
Using two surveys from 2008, Douglas Spencer and Christopher Elmendorf estimated racial resentment across the states by measuring responses to questions about the work ethic, intelligence and trustworthiness of blacks Americans. They estimate the portion of whites in each state that are in the top quartile of racial resentment (so a state with large amounts of racism would have more than 25 percent of whites espousing racially biased views).
Other stories recommended for you
I asked Eric Stone, the lead data scientist at Smarterer and a former hedge fund manager and USDA statistician, to help me see if there was another factor driving this trend. He ran the numbers again, controlling for state revenues per capita (based on Tax Foundation data, averaged from 2008-2010), as well as the average ideology of citizens and the average ideology of the state government (based on data from Richard Fording, again averaged over the period of 2008-2010).
Stone tells me that after controlling for these factors, “racial resentment is indeed the strongest predictor of TANF, at least in the context of this model.” He reports a p-value of <.001 (indicating there is a less than .1 percent chance that the relationship was observed by chance) and an R2 of .556 for the model (meaning that these variables collectively explain more than half of the variation in TANF spending).
advertisement
Sean McElwee is a research assistant at Demos. His writing may be viewed at seanamcelwee.com. Follow him on Twitter at @seanmcelwee.
More Sean McElwee.
More Sean McElwee.
No comments:
Post a Comment