Type of Document	Dissertation
Author	Moore, Trent Wade
Author's Email Address	twm8357@garnet.acns.fsu.edu
URN	etd-04102006-173259
Title	A Construct Validation of Partner Awareness Risk among Seronegative Heterosexual Couples
Degree	Doctor of Philosophy
Department	Sociology, Department of
Advisory Committee	Advisor Name Title David F. Sly Committee Chair Dianne F. Harrison Committee Member Elwood Carlson Committee Member John R. Tayler Committee Member
Keywords	Couples Partner Risk Heterosexual HIV Sexually Transmitted Diseases
Date of Defense	2006-03-15
Availability	unrestricted
Abstract Heterosexual transmission of HIV is contingent not only one’s own but one’s partner’s behavioral risk levels. Public health advocates have long promoted “knowing” one’s partner as a means for preventing the spread of STDs. Perceptions of partner risk levels, however, are often inaccurate and incomplete. Published research on this issue is limited due to an over-reliance on partner-specific research designs and overly rational assumptions of health-protective behavior. Derived from couple-level data, partner awareness risk (PAR) is based on respondent perceptions of partner HIV-related behavioral risks and partner self-reports of incurring those risks. Using data from the pretest phase of a randomized trial to reduce the HIV risks of at-risk heterosexual couples in South Florida, the internal and cross-structural validity of PAR is assessed. Confirmatory factor analysis results suggest PAR is composed to two distinct dimensions i.e., PAR attributable to both ignorance (PARig) and misperception (PARmp). A second-order analysis further suggests these two dimensions load significantly onto a single higher-order factor. Cross-structural analyses suggest PAR is associated with a variety of sociodemographic, behavioral risk, relational, and psychosocial variables. Unmarried cohabiting and non-cohabiting respondents incur significantly higher PAR levels than those who are married. While the harmful effect of being unmarried is greater for men, being married is more hazardous for women than being unmarried. Those in close, affectionate, and largely democratic relationships incur lower PAR levels. Both respondent and partner behavioral risk levels are positively associated with PAR levels. When both partners incur elevated sex history risk, however, respondents incur relatively lower PAR levels. Women incur significantly higher PAR levels than men. Many correlates of PAR are more harmful for women (e.g., partner sex history risk, being black or Hispanic, intimate partner violence, masculinity attitude, relational power, partner having greater educational status, etc.). A variety of other correlated serve as more effective buffers for the men (e.g., general partner knowledge, condom attitude, partner perception of respondent self-efficacy, etc.). PAR level is positively associated with some health-protective outcomes (e.g., perceived HIV vulnerability, preparedness for behavioral change, and HIV testing). Negative influences, however, are indicated through interactions. For example, substance use risk is negatively associated with HIV testing only in the presence of elevated PAR levels. Also, for the women only, elevated PAR is associated with lower odds of reported condom use with a main partner.
Files	Filename       Size       Approximate Download Time (Hours:Minutes:Seconds)     28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access    trentwmoore.pdf 3.69 Mb 00:17:03 00:08:46 00:07:40 00:03:50 00:00:19