Measuring racism is difficult for researchers. This is because of the problem of social desirability. There are various ways in which researchers have become more comfortable measuring it. The method that Tesler (2012) uses to measure racism is the Racial Resentment Scale (RRS) which is derived from Kinder and Sanders (1996). I argue that this scale is not measuring what Tesler (2012) wants it to measure for the purposes of his research.
The components that make up the RRS include questions of work ethic of African Americans, respondents’ perceptions of the advancement of discrimination, and questions about black people receiving more than they deserve. Two of the prompts included in the RRS that I see red flags with are “Over the past few years blacks have gotten less than they deserve,” and “generations of slavery and discrimination have created conditions that make it difficult for blacks to work their way out of the lower class” (Kinder and Sanders, 1996). These prompts, in my opinion, are set up in such a way that they cannot capture racial resentment in the way that Tesler would like it to be used. Tesler is looking to see if racial resentment has a bigger impact on healthcare attitudes during Obama than they did during Clinton. Since Tesler is using the RRS to measure if racism was present in a significant way during the Obamacare campaign, it is implied that the RRS is a measure racist attitudes rather than racial resentment, which I believe are different. I argue that the prompts within the RRS are not measuring racist attitudes, but rather measures from commiseration to resentment.
The racial component of the opinions on Obama’s healthcare plan being causally due to his race was debated during the promotion his healthcare plan. Obama didn’t think that attitudes were due to race, but Tesler thought there was a causal role in race and subsequently uses the RRS to see if people resent blacks during this Obama healthcare campaign more than they did in the 90s. He uses an experiment to establish causation, but the measurement is still questionable. Take the prompt, “Over the past few years blacks have gotten less than they deserve” for example. If the respondent doesn’t agree with this statement, it would signify racial resentment when in reality, they may not quite understand whether the statement is true. What “they deserve” is a bit ambiguous as well.
127 our experts are availableright now
The response to this prompt may not represent racism but rather, confusion. For the prompt “generations of slavery and discrimination have created conditions that make it difficult for blacks to work their way out of the lower class” the assumption is that they haven’t worked their way out of the lower class (Kinder and Sanders, 1996). This wording is a bit concerning considering that the president at the time was black and had clearly worked his way out of the lower class. Even if the respondent were to disagree with this prompt, and the disagreement was understood as racial “resentment”, I don’t think that resentment particularly translates to racist attitudes, which is what Tesler is trying to measure. Rather, resentment may simply be a matter of irritation (Kinder and Sanders, 1996).
Tesler’s findings allegedly support his hypothesis by showing that as whites become more racially resentful, their views grow more unfavorable towards government health insurance. As I said, I believe that the RRS measures levels of sympathy and resentment instead of racist attitudes. Even if a white respondent was less sympathetic towards blacks that doesn’t neccesarily translate to attitudes about healthcare policy. I don’t think that lack of sympathy towards blacks specifically would lead to racialized healthcare policy sentiment because the white respondent could be an unsympathetic person in general.
Tesler’s measure of stereotypes was an attempt at a more direct way of measuring racist attitudes I believe. Problematically, stereotype recognition also does not necessarily translate to racism. Especially with a black president, negative black stereotypes will have questionable credibility. Further, one may agree with certain racial stereotypes in a more matter-of-fact way without being racist at all. One more note about Tesler’s stereotype measure is that there is also statistical significance that shows anti-white stereotypes alongside of anti-black stereotypes in 2009 in the unweighted sample, in is arguably close to the .05 level on the weighted sample. Tesler does not discuss this in his results at all. My argument as to why there is an increase in the appearance of alleged racial resentment and stereotypes in 2009 as opposed to the Clinton healthcare era is because of the 2008 housing recession.
This sudden hit to the economic security of many Americans likely translated into selfishness. This self-regard is likely due to an “every man for himself” policy that many white Americans likely experienced during the recession. Self-interestedness would explain the statistically significant stereotypes for both blacks and whites. The relation of this heightened self-concern to sentiment on Obama’s healthcare plan is the assumption that its implementation would harm voters with private health insurance because of increased costs with the introduction of public health insurance. Controlling for ideology was helpful in this analysis, but it still falls short in explaining that the economic insecurity induced by the recession was likely to have an impact on all respondents as they may be afraid of the economic precarious situation publicized healthcare may put them in.
I propose two remedies for the reliability of this article. The first is a more appropriate measure of racial attitudes. I believe that implicit methods of interpreting racism are too relaxed. They may get mixed up with irritation as opposed to true racism. I don’t believe irritation will consistently be grounds for suppositions of inferiority. Explicit measurement of racial attitudes may yield some social desirability biases, but the spectrum would be more applicable since what Tesler is trying to measure is racist attitudes rather than irritation or lack of sympathy. Second, a control for self-perceived economic stability would likely explain some of the spikes in stereotyping in 2009. The stereotype measure used measured whether racial groups were work ethic, proneness to violence, temperament, self-sufficient, and demeanor. It is likely that the work ethic and the self-sufficiency indicators were heightened because of the recession. Including this control variable would remedy this confound within the measure of stereotypes.
In sum, Tesler’s (2012) question about whether Obama’s healthcare plan was divided based on racialization is an important one. The RRS fails to capture racial attitudes in a trustworthy way. Future research should use a more direct method of measuring racial resentment and it should control for self-perceived economic stability in order to appropriately draw conclusions regarding the racialization of healthcare in 2009.
