Beliefs and Attitudes about Mental Health and Illness
of White Middle-Class Americans
Claire M. Cassidy
KEY WORDS: mental health conditions, question comprehension, health reporting, National Health Interview Survey, questionnaire pretest, cross-cultural reporting differences
This 1992 report is one of five exploratory ethnomedical studies sponsored by the Census Bureau and funded by National Center for Health Statistics in anticipation of adding mental health measurements to the National Health Interview Survey (NHIS). To examine possible cross-cultural differences in reporting of mental health conditions, the Census Bureau solicited proposals from ethnographers and medical anthropologists to explore peoples' language and conceptual organization of the mental health domain, specifically depression, anxiety, phobia, and panic. Each study was conducted in a different group, following a similar research protocol. In-depth ethnographic interviews explored health and mental health concepts in an open-ended fashion, then a second set of focused interviews elicited reactions to a set of proposed survey measures of symptoms of anxiety, depression, panic, and phobia. The intent was identify culture-bound disorders which may affect survey reports of mental health conditions, and to comment on the cultural appropriateness and validity of the questions and terms proposed to elicit reports of mental health symptoms. The studies attempted a different approach to questionnaire pretesting by using ethnographic research to inform, refine, and test survey question wordings that have greater cross-cultural validity.
Cassidy reports the results of 17 interviews conducted in the first phase, and 15 in the second, all with white middle class Americans living in the Baltimore-Washington DC area. She found that her respondents tended to “normalize” names of symptoms that may be treated as abnormal by mental health professionals. “Mental illness” was regarded as a very strong and forbidding term whose use was reserved for the most extreme symptoms. Respondents’ reactions to questions about mental illness were varied, with majorities perceiving the questionnaire as very negative in its tone. Questions were seen by most respondents as ambiguous in that they did not allow for an assessment of the context in which a symptom occurred. Based on respondents’ assessments, Cassidy offers general recommendations for designing a more user-friendly questionnaire to measure mental health conditions.
CITATION: Claire M. Cassidy. 1992. Beliefs and Attitudes about Mental Health and Illness of White Middle-Class Americans. Study Report Series (Survey Methodology #2007-18).