Building: Pinnacle
Room: 3F-Port of Hong Kong
Date: 2016-07-03 11:00 AM – 12:30 PM
Last modified: 2016-05-31
Abstract
Introduction & Objectives
Critical consciousness (CC) represents marginalized people’s critical analysis of their social conditions (i.e., critical reflection) and individual/collective action to change perceived inequities (i.e., critical action; Freire, 1973). CC is considered an ‘antidote’ to structural oppression by unlocking human agency despite societal constraints (Watts et al., 1999), and has been associated with a host of positive outcomes among marginalized youth (e.g., Author Citation, 2011; Campbell & MacPhail, 2002; Ramos-Zayas, 2003).
Despite these scholarly advances, one vexing problem in CC scholarship is the inconsistent conceptualization and measurement of CC (Watts et al., 2011). The Critical Consciousness Scale (CCS) was developed to address this gap.
Methodology
Participants (N = 326; 56.9% female; M age = 15.47) were predominantly poor and working class youth of color (92.3%) from the U.S., who completed an initial 46-item measure. Data were subjected to EFA and CFA with two randomly generated and independent datasets (each containing N = 163; DeVellis, 2012).
Results
EFA in Study One suggested three factors: (1) Critical Reflection: Perceived Inequality, (2) Critical Action: Sociopolitical Participation (3) Critical Reflection: Egalitarianism. CFA in Study Two confirmed the suggested factor structure, resulting in a final three-factor and 22-item ‘Critical Consciousness Scale’ (CCS). [Items and a description of each factor detailed in embedded image, below.] The EFA and CFA exhibited strong model fit (i.e., RMSEA = .05; SRMR = .06; CFI & TLI = .98) and internal consistency estimates (α range = .85 - .90).
Conclusions
Substantively, the CCS has the potential to unite and advance fragmented conceptualization and measurement of CC in diverse populations, by providing a theoretically informed, reliable, and valid CC measure. Methodologically, this research provides further evidence that strong item loadings attenuate the need for large sample sizes in factor analyses (e.g,. Gagné & Hancock, 2006; Marsh et al., 1998).
