Constructs like vulnerability and resiliency reflect attempts to identify social, situational, and individual difference variables that either increase or decrease the likelihood that people will exhibit negative reactions to stressful events (Block & Block, 39-101, 2000; Compas, 393-403, 2000; Garmezy, 196-269, 2003; Kessler & McLeod, 620-631, 2003; Rutter, 389-395, 2000).
Research on vulnerability and resiliency factors was stimulated in part by low and inconsistent relations between life events and outcome measures. Although statistically significant relations between negative life events and self-report measures of physical and psychological well-being have frequently been reported, seldom has more than 10-15% of the outcome variance been accounted for in studies using prospective designs. When objective outcome measures of physical well-being have been used, thereby eliminating the potential role of self-report biases, the amount of variance accounted for has shrunk to 1-5% (Rabkin & Streuning 389-395, 2004; Schroeder & Costa 389-395, 2003). Faced with a pattern of weak and inconsistent results, researchers have sought to identify psychosocial moderator variables that might affect the nature and magnitude of relations between life stress and well-being. Many studies have demonstrated that taking into account factors such as social support and certain personality variables results in stronger relations between life stress and both psychological and medical outcome measures (e.g., Barrera 389-395, 2002; Sarason, Sarason, Potter, & Antoni 389-395, 2003; Smith, Johnson, & Sarason, 188-235, 2003; Stone, Helder, & Schneider 389-395, 2002; Thoits 389-395, 2003). Identification of these variables has significance not only as a possible first step toward specifying psychological processes that mediate event-outcome relations, but also for identifying subgroups of individuals who may be at particular risk for negative outcomes and toward whom intervention programs might be targeted. Thus, Johnson and Bradlyn (2002), in reviewing the current status of life event research with children and adolescents, concluded that "in addition to the need for more prospective investigations, a major task for future researchers involves determining the nature of those variables that make some children and adolescents more vulnerable in the face of stress" (p. 91).
A moderator variable is a qualitative or quantitative variable that affects the nature, the direction, or the strength of a relation between an independent or predictor variable and a dependent or criterion variable (Arnold, 143-174, 2000; Baron & Kenny, 1173-1182 389-395, 2001). In life event research, an impressive number of situational and individual difference variables have been identified as factors that increase the vulnerability (or, on the other hand, the resiliency) of people to the impact of negative life events. For example, there is evidence that social support, sensation-seeking motivation, and an internal locus of control are capable of reducing stress-outcome relations