Poster #98 - Person-Centered Profiles of Risk and Protective Factors in College Students

• In Event: 3-206 - Poster Session 13
  In Poster Session: PS 13 Section - Family Context & Processes

Integrative Statement

According to the cumulative theory of risk, more adversity during childhood increases the risk for problems in adulthood, such as depression and suicide, academic failure, and physical disorders (Felitti et al., 1998; Romano et al., 2015; Rutter, 1979). Importantly, experiences of risk can be mitigated by protective factors that foster resilience and promote positive outcomes (Masten, 2001). Person-centered approaches can enhance our understanding of resilience by describing how different risk and protective factors tend to aggregate within individuals (Bauer & Shanahan, 2007). In this study, we utilized latent class analysis to identify latent profiles of childhood risk factors and protective factors within a sample of young adults. We then examined whether outcomes of post-traumatic stress symptoms, cross-domain competence, and academic functioning would differ based on these latent classes.
Participants were 549 college undergraduates from two different universities (one public, one private) who completed an online survey for class credit. The sample were 67% female, 54% Caucasian, 17% African American, 15% Hispanic/Latino, and 12% Asian. Measures included the Adverse Childhood Experiences survey (Felitti et al., 1998) to assess risk and the Protective and Compensatory Experiences Survey (Morris et al., 2016) to assess protective factors. Outcome measures included the Post-traumatic Stress Disorder Checklist (Weathers et al., 2013), a competence score based on six items assessing competence in areas of work, family, health, and wellbeing, and self-reported high school GPA.
Using a model comparison approach, we determined that the best fitting and most parsimonious model of latent risk involved two classes, one with low rates of all risk factors and one with higher rates. Similarly, the best-fitting, most parsimonious model of latent protection involved two classes, one with high rates of protective factors and the other with lower rates. Indicators and probabilities by class are presented in Table 1. We then estimated a model correlating the latent variable of risk and the latent variable of protection to produce joint probabilities. This model had good fit with χ2= 241.9, p < .999 and adequate class separation with entropy = .74. Joint probabilities of the two categorical latent variables were as follows: the largest class representing 63% of the sample were low risk, high protection while 21% were high risk, high protection, 12% were high risk, low protection, and 4% were low risk, low protection.
One-way ANOVAs revealed significant differences in all three outcome measures based on most likely class membership (ps < .05, see Table 2). Post-hoc Tukey HSD comparisons identified significant differences between the low-high and the high-low group in terms of PCL scores, competence scores, and grades, and the high-high group in regards to trauma symptoms and competence (ps < .05. There were also significant differences in competence between the high-low group and both the low-low and high-high groups (ps < .05). The difference between the high-low and high-high groups suggests that having high levels of protective factors might afford better outcomes to an individual exposed to high levels of risk compared to those who do not have many protective factors.

Authors

• Raechel Martin, Villanova University
  Presenting Author

• Joanna Nicole Keane, Villanova University
  Non-Presenting Author

• J. J. Cutuli, Rutgers University - Camden
  Non-Presenting Author

• Janette E. Herbers, Villanova University
  Non-Presenting Author