Using Multilevel Factor Analysis With Clustered Data

Abstract

Advances in multilevel modeling techniques now make it possible to investigate the psychometric properties of instruments using clustered data. Factor models that overlook the clustering effect can lead to underestimated standard errors, incorrect parameter estimates, and model fit indices. In addition, factor structures may differ depending on the level of analysis. The current study illustrates the application of multilevel factor analytic techniques using a large statewide sample of middle school students (n = 39,364) from 423 schools. Both multilevel exploratory and confirmatory factor analyses were used to investigate the factor structure of the Positive Values Scale (PVS) as part of a school climate survey. Results showed that for the PVS, a two-correlated factor model at Level 1 and a one-factor model at Level 2 best fit the data. Implications and guidance for applied researchers are discussed.

The Moderating Role of Ethnic Identity on the Relationship Between School Climate and Self-Esteem for African American Adolescents

“…Some scholars purport that individual perceptions of school climate should be analyzed only at the individual level (Dixon et al, 1991;Mok & McDonald, 1994;Raudenbush et al, 1991). Conversely, other scholars argue that the focus of analysis for school climate should only be examined at the school level or above (Huang & Cornell, 2016;Konold et al, 2014). Given this lack of consistency in definition and analysis, the current study is limited in the interpretation of the results as representative of perceptions of overall school climate as a construct.…”

Exploration of Factor Structure and Measurement Invariance by Gender for a Modified Shortened Adapted Social Capital Assessment Tool in India

“…Exploratory factor analysis (EFA) was implemented using the first random subset (n = 6,207) to identify a tenable factor structure, which also has the necessary goodness of fit indices. Next, using the second subset (n = 6,266), a confirmatory factor analysis (CFA) was implemented to assess the generalizability of the possible factor structure (Huang and Cornell, 2015) to confirm if social capital measured by SASCAT-I conformed the theoretical constructs of social capital (Stafford et al, 2008;Elgar et al, 2011).…”

Development and validation of the School Climate Questionnaire for Secondary and High School Teachers (SCQ-SHST).

“…This questionnaire does not evaluate those aspect related to Safety and Institutional Environment because these dimensions are less consistently associated with academic outcomes and their effects often disappear when other factors of school climate are controlled (Wang & Degol, 2015). In addition, recently some authors have developed school climate measures that address these aspects from an authoritative approach to school discipline and childrearing (e.g., Bear et al, 2014;Huang et al, 2015), which makes these questionnaires useful for those schools interested in implementing bullying prevention programs. In contrast, the SCQ-SHST provides a perspective of school climate especially useful for those schools interested in articulating interventions aimed at improving the learning environment through the seven dimensions addressed.…”

“…Future studies should evaluate the differences between schools in terms of school climate, for which it is necessary to consider the clustering nature of individual participants within schools. Recent studies have shown that multilevel modeling techniques are useful and necessary for this purpose and that, in fact, there seem to be such differences between schools (e.g., Huang et al, 2015).…”