Event Title
Ethnic Bias within the NESARC Personality Disorders using Item Response Theory
Loading...
Location
UND Columbia Hall, 1370
Start Date
19-10-2019 8:50 AM
End Date
19-10-2019 9:05 AM
Description
Racial bias within the DSM5 has been recognized in studies of personality disorder diagnoses. Item Response Theory Differential Functioning of Items and Tests was used to examine the NIAAA's National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) Wave 1 and 2 data. A review of prevalence information by gender and ethnicity was undertaken. Hypotheses included that bias would be present in the items and total scores of personality diagnostic scales. In particular, items and scales which rely on cultural comparisons, and out of body experiences, magical thinking, hallucinations or delusions would be diagnosed more frequently in Native American participants. Results indicated test level bias within five of the scales for African Americans, two for females, one for males, and one for Native Americans. The schizotypal scales was biased for Native Americans and African American participants, as hypothesized. Only three scales contained no scale level bias: avoidant, dependent and obsessive-compulsive.
Ethnic Bias within the NESARC Personality Disorders using Item Response Theory
UND Columbia Hall, 1370
Racial bias within the DSM5 has been recognized in studies of personality disorder diagnoses. Item Response Theory Differential Functioning of Items and Tests was used to examine the NIAAA's National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) Wave 1 and 2 data. A review of prevalence information by gender and ethnicity was undertaken. Hypotheses included that bias would be present in the items and total scores of personality diagnostic scales. In particular, items and scales which rely on cultural comparisons, and out of body experiences, magical thinking, hallucinations or delusions would be diagnosed more frequently in Native American participants. Results indicated test level bias within five of the scales for African Americans, two for females, one for males, and one for Native Americans. The schizotypal scales was biased for Native Americans and African American participants, as hypothesized. Only three scales contained no scale level bias: avoidant, dependent and obsessive-compulsive.