The Snowball Effect: The Calculation of a Cumulative Risk Index for Multiethnic Sexual Minority Youth

Purpose: Sexual minority youth (SMY) are considered a population at risk. Studies have identified factors such as discrimination from peers and familial rejection (D'Augelli, Grossman, & Starks, 2008), to be associated with risks for low self-esteem (Elze, 2002) and depression (Bontempo & D'Augelli, 2002), yet little research has focused on the cumulative risks that could exacerbate such outcomes for multiethnic sexual minority youth (MSMY). Risk has been defined as an accumulation of stressors that increase the likelihood of negative developmental effects (Hughes, 2007). Studies have focused on the influence of individual risk factors, yet they rarely occur in isolation or are solely responsible for negative outcomes (Seifer, Sameroff & Baldwin, 1992). The quantity of experienced risk factors can be more influential than any particular risk (Goldman, 2010), despite the assumption of equal weight for risks that may be experienced disproportionally for some adolescents. Cumulative risk indexes (CRI) have been found to predict a multitude of negative developmental outcomes (Bean, Gil-Rivas, Greenberger & Chen, 2002) and capture the covariation of risk factors (Luthar, 1993). This paper will discuss the construction of the CRI and the advantages and disadvantages of using this type of risk analysis with MSMY. Methods: Data were collected from a counseling program for SMY (n=163) situated in a resource depleted urban environment. Based on ecological systems theory (Bronfenbrenner, 1989), and previous studies of SMY (Elze, 2003), risk factors were categorized into family, sociodemographic, personal health, and school risk variables. The CRI was constructed as follows: 1) Categorical measures of risk were dichotomized indicating the absence or presence of risk factors; 2) continuous responses were scored using a 75th percentile cut point (Gerrard & Butler, 2004), and 3) total scores were added to create a CRI. Self-esteem was assessed using the Self Esteem Scale (a=.77), (Rosenberg,1994). Utilizing multivariate structural equation modeling (SEM) with AMOS 18.0 software, this study explored the relationship of a latent variable of cumulative risk to self esteem. Results: The majority of participants identified as female (60%), Latina (55%), and bisexual (43%), with a mean age of 16. Family (86%) and sociodemographic (63%) risk domains were the most common. The mean CRI score was 9 (SD=3.4; range 1-13). SEM analysis produced good fit statistics x2 (12, p = > .853) = 12.21; CFI = .997; GFI =.995; and examination of the model path coefficients revealed that significant negative relationships were found between the CRI and the self-esteem of MSMY above and beyond particularly influential domains such as family risk. Conclusions and Implications: This study attempts to illuminate the role of cumulative risk factors for MSMY through the utilization of a CRI. This study extends our knowledge by providing evidence that individual risk factors may not be sufficient to fully explain outcomes. Specifically, it appears that greater cumulative risk negatively impacts the self-esteem of MSMY. A CRI based on an ecological systems approach will enhance the development of more effective interventions and policies that capture emerging needs of MSMY exposed to myriad risk factors.