Racial Identification Confounds the Assessment of Obsessive-Compulsive Symptoms in African-Americans
Download data and study materials
Field period: 12/16/2004 - 12/27/2004
African-Americans show unusually high endorsement rates on self-report measures of contamination anxiety. Using a nationally representative sample of African-Americans supplemented with a smaller national sample of European-Americans (N=258), we compare black-white responses on contamination scales of two OCD measures. This study confirms the pattern of over endorsement is a national phenomenon rather than an occurrence restricted to certain geographical areas. We also document the salience of participants' racial / ethnic identity as a factor in this bias. Blacks who complete a measure of ethnic identification prior to completing anxiety items score significantly higher than blacks given the ethnic identity measure last. We examine the hypothesis that non-pathological unconscious feelings of threat by African-Americans are contributing to over-endorsement of anxiety items.Hypotheses:
Hypothesis 1: Blacks will report significantly more contamination anxiety than whites.
Hypothesis 2: Blacks will report significantly more anxiety when race is made salient.
We introduced thoughts about race in black and white participants by administering a measure of ethnic identity immediately before anxiety items (MEIM: Multi-Ethnic Identity Measure). Questions about ethnic identity were intended to prime the subject with thoughts about his or her own ethnic group, thereby activating racial stereotypes. Half the subjects were randomly chosen to the MEIM before contamination measures, and half the subjects were given the MEIM after the contamination measures.Key Dependent Variables:
The experiment takes the form of a 2 (Black and White participant) X 2 (ethnicity salient and ethnicity non-salient) design. The primary dependent variables are scores on the following measures of contamination anxiety: the Padua Inventory contamination scale and Obsessive-Compulsive Inventory (OCI) washing scale.Additional Information:
Data were collected by TESS via Internet using web TV. The sample consisted of 298 black and white participants from across the US. Excluded were Hispanic participants, those living in the US for less than ten years, and those who reported a race other than white or black. After data was collected, any participant reporting a mixed racial background or a history of OCD was removed from the analysis (40 participants in all). This resulted in a final sample of 208 blacks and 50 whites. The mean age of participants was 28.6, SD 5.1. Subjects were 35% female and 65% male.Summary of Findings:
T-tests were used for comparisons between blacks and whites on the items and contamination scales. Blacks scored higher on all items, and were significantly higher on 90% of the items. On the Padua contamination scale, whites had a mean of 9.06 (SD 7.00) and blacks had a mean of 14.65 (SD 8.33). This difference was significant t(256)=4.39, p<0.001. On the OCI washing scale, the same pattern was observed, with a mean of 2.28 (SD 2.56) for whites and 3.69 (SD 2.87) for blacks, where t(256)=3.19, p<0.001.Figures/Tables: Conclusion:
This study confirms that in a non-clinical sample, blacks significantly outscore whites on the contamination and washing scales on the Padua Inventory and Obsessive-Compulsive Inventory. Over endorsement occurs consistently across all washing and contamination items, and is greater when awareness of ethnic and racial identification is increased. Clinicians and researchers should use such measures with caution in African-Americans. Evidence suggests over endorsements are due in part to the salience of ethnic and racial information about the participant, which could result in a phenomenon similar to stereotype threat. Further study is needed to identify additional factors contributing to these differences
Williams, M. T., E. Turkheimer, E. Magee, T. Guterbock. 2008. The Effects of Race and Racial Priming on Self-Report of Contamination Anxiety, Personality and Individual Differences. 44: 744-755.