The Psychometric Properties of the Harter Self-Perception Profile for Children with At-Risk African American Females

Abstract

The Harter Self-Perception Profile for Children (SPPC) is one of the most commonly used measures of childhood self-esteem, yet there is little research assessing the psychometric properties of the SPPC for use with an African American population. A sample of 92 African American adolescent females (M age = 12.33) was administered the SPPC in order to assess its suitability for this population in three ways. First, an exploratory factor analysis demonstrated complex components without any factors being identical to the normative factors. The greatest differences were within the behavior and scholastic subscales which had items cross loading on three different factors. Second, the SPPC demonstrated only moderate internal reliability with subscale alpha coefficients ranging from .71 to .82. Third, a comparison with the Rosenberg Self-esteem Scale provided evidence of poor convergent validity. These results raise questions about the validity of the SPPC for use with African American adolescent females.

Improved Representation of the Self-Perception Profile for Children Through Bifactor Exploratory Structural Equation Modeling

“…First, the superiority of ESEM over ICM-CFA suggests that cross-loadings are needed to reflect the fact that the SPPC assesses conceptually related constructs. The observation of multiple cross-loadings remaining within an acceptably low level (i.e., most of them below .200) corresponds to the findings from previous EFA investigations of the SPPC (Chan, 2002;Miller, 2000;Stewart et al, 2010). The strongest cross-loadings emerged between GSC and physical appearance, in line with previous research (Frost & McKelvie, 2004;Harter, 1999;.…”

“…Prior CFA research on the factor structure of the SPPC and other SPPbased instruments has led to some surprising results (e.g., high factor Arens, Morin correlations; Boivin et al, 1992;Cole et al, 2001;Maïano et al, 2008;Stewart et al, 2010) that have often been attributed to its structured alternative response format (Eiser et al, 1995;Marsh et al, 1994;Wichstrøm, 1995;Wylie, 1989). The present study shows that this might not be the case.…”

Examination of the Structure and Grade-Related Differentiation of Multidimensional Self-Concept Instruments for Children Using ESEM

“…A series of CFA studies have generally tended to support the factor validity of the SPPC, as well as its invariance across gender and grade levels, within samples of Dutch (e.g., Muris, Meesters, & Fijen, 2000;Van den Bergh, & Van Rats, 1998), French-Canadian (e.g., Boivin, Vitaro, & Gagnon, 1992;Gavin, & Herry, 1996), and Irish (Shelvin, Adamson, & Collins, 2003) children. Using EFA, Miller (2000) managed to replicate the original factor structure of the Finnish SPPC, while Stewart, Roberts, and Kim (2010) failed to do so with a sample of African American females. Similarly, Schumann et al (1999) showed that whereas the a priori factor structure of the SPPC was well supported among white American girls, a more complex factor structure including cross-loadings was required for African American girls.…”

A Child Self-Report Exemplar: Adapting Administration Procedures of the Self-Perception Profile for Children to Reduce Measurement Error

“…Although it is beyond the scope of this cross-sectional study to explore differences between cultural groups, several studies (Schumann et al, 1999;Stewart, Roberts, & Kim, 2010) have noted component structure differences between Harter's normative sample and African American preadolescent girls. In a comparison study between African American and European American girls, Schumann and colleagues (1999) used principal factor analysis and found that, over time, items in the behavior and academic domains began to merge in the African American group only, suggesting culture related shifts with maturity.…”