Perceived youth empowerment: reliability and validity of a brief scale

American Journal of Health Studies, Summer-Fall, 2005 by Raheem J. Paxton, Robert F. Valois, J. Wazner Drane

Abstract: This study determined the reliability and validity of a brief perceived youth empowerment scale in a sample of 2,482 middle school students. An alpha coefficient and confirmatory models were constructed for a five-item scale. Results suggest that the hypothesized factor structure did not fit the data. Modifications did suggest covariance between error terms [/[sup.2](3)=6.69, Root Mean Square Error Approximation = .0228, Comparative Fit Index = .9960, Non-normal Fit Index = .9868]. A valid and reliable brief scale for measuring perceived youth empowerment is advantageous to field research and evaluation of adolescent health promotion interventions.

**********

A number of authors have attempted to conceptualize empowerment. Rappaport (1981) defines empowerment as a way for individuals to gain control over their lives through active participation. Kieffer (1984) defines empowerment as a dynamic process that consists of participatory competence. The idea of participation has been demonstrated to enhance self-acceptance, self-confidence, and develop active roles in a community (Zimmerman & Rappaport, 1988). The challenge in conceptualizing youth empowerment is that it is often described as both a process and a developmental outcome. Hence, it is a factor that evolves over time, or a final attribute through relevant experience. Although the conceptual definitions differ, each shares similar features, in particular, the concept of participation.

Despite numerous conceptual definitions for empowerment, few exist for youth. A more holistic definition for youth empowerment was proposed by Chinman & Linney (1998). Their model consists of an opportunity to participate in meaningful roles, where youth are able to bond with positive adults in the environment and receive positive reinforcement and recognition. Within Chinman & Linney's model, youth who are empowered, develop appropriately during adolescence. For example, empowerment may provide youth with positive traits, positive labels, and in general, a positive identity (Bern, 1978; Rosenburg, 1979). Generally, adolescents who feel connected and are influenced by positive adult norms and expectations are expected to lead successful lives (Moody, Childs, & Sepples, 2003). Erickson (1968) believed that the reinforcement received from bonding to, and participation with, positive peers and adults may lead to positive developmental outcomes.

In addition, few conceptual definitions, empirical data, and scales that assess youth empowerment exist in the professional literature. The few studies that exist in the literature for youth empowerment are derived from the developmental asset framework conceptualized by the Search Institute (Leffert, Benson, Scales, Sharma, Drake, & Blyth, 1998). Empowerment within the developmental asset framework is one of four external assets, which refer to the positive developmental experiences of relationships and opportunities that adults provide (Benson, Leffert, Scales, & Blyth, 1998). Leffert and colleagues operationalized youth empowerment via 11 items among four subscales. The four subscales being: 1) Community Values Youth (4-items)--young persons perceive that adults in the community value youth; 2) Youth as Resources (3-items)--young people are given useful roles in the community; 3) Service to Others (1-item)--young person serves in the community one hour or more per week; and 4) Safety (3-items)--young person feels safe at home, at school, and in the neighborhood (Leffert et al., 1998).

Content and construct validity were established via an exploratory factor analysis reported by the Search Institute researchers (Leffert et al., 1998). Scales and colleagues provided evidence of both predictive and discriminant validity. Furthermore, elements of empowerment were predictive of exhibiting leadership, helping others, maintaining good health, and overcoming diversity (Scales, Benson, Leffert, & Blyth, 2000). Despite evidence for construct validity, and both predictive and discriminate validity of the 11-item youth empowerment instrument, some weaknesses were observed. Primarily, the internal consistencies on two of the instruments' subscales (i.e., youth as resources and safety) were quite low (.48 and .50, respectively). A widely accepted rule of thumb proposed by Nunnally (1978) suggests that alpha coefficients that don't reach at least a = .70 are inadequate. Additionally, the Service to Others subscale only consisted of one item.

A systematic literature review suggests that only two other published studies propose a youth empowerment scale, or a scale conceptually similar to empowerment in youth. Each was based on data from high school age adolescents. The Reininger, Evans, Griffin, Valois, Vincent, et al. (2003) scale assessed the validity via exploratory factor analysis, as well as the reliability of a youth empowerment scale. The scale consisted of a total of 5 items and appeared to have adequate validity (i.e., loading > .40) and reliability or internal consistency (a = .78). Oman et al. (2002) proposed a 6-item community involvement scale in a high school population. Validity and reliability was similar to that of Reininger, Evans, Griffin, Valois, Vincent et al. (e.g. loading > .40, a = .78). Construct validity was established by the presence of simple structure revealed through exploratory factor analysis in the Reininger, Evans, Griffin, Valois, Vincent et al. study and principal component analysis in the Oman et al. (2002) scale. Despite reliability and validity of these two instruments, no conceptual definitions were provided by the authors of these studies.

Constructing reliable and valid scales for adolescent empowerment research and evaluation can be difficult. In order to conduct reliability and validity investigations, consistent conceptual frameworks and standard instruments are needed. In support of Chinman & Linney's (1998) theoretical hypotheses regarding empowerment, we support the notion that if youth are participating in meaningful activities within the community and bonded to positive peers and adults in these activities positive youth development may transpire. Empowerment is potentially an important aspect in understanding the developmental nature of adolescents. This idea is often imperative to health promotion program intervention. Few short scales exist which access youth empowerment. A burden is often placed on adolescents to complete long assessment questionnaires, often with multiple subscales with a large number of items. Hence, the development of short psychometric scales could be advantageous to school or community-based interventions. Based on previous evidence, the five-item scale proposed by the Reininger, Evans, Griffin, Valois, Vincent, et al. (2003) study appeared to have adequate face validity. In turn, the purpose of this study was to extend the work by Reininger, Evans, Griffin, Valois, Vincent, et al. (2003) via confirmatory factor analysis (CFA) of a modified version of their scale in a middle school population.