The Development of the Cayci Parent/Caregiver Overall School Perceptions Scale

Background & Purpose: Parent/caregiver involvement in school is among the most important influences and determinants of children’s learning, healthy development, and success in school.  Still, a growing body of research indicates that school practices may deter meaningful parent/caregiver engagement in schools. There are few sound tools, however, to assess parent/caregiver overall perceptions of the school and its practices. This study explores the initial psychometric properties associated with the Community and Youth Collaborative Institute (CAYCI) Parent/Caregiver Overall Perceptions Scale.

Methods: As part of a school community needs assessment process, 1395 parents/caregivers of elementary school students from Salt Lake County, Utah completed the paper-pencil version of either the Spanish-language or English-language CAYCI Parent/Caregiver Overall Perceptions Scale. The respondents identified themselves as White/Non-Hispanic (49.4%), Latino/Latina (40.0%), Mixed/Multi-Racial (5.0%), African American (1.7%), or Asian (1.2%). This included 1100 mothers, 221 fathers, 22 grandmothers, 8 grandfathers, 13 legal guardians (not parents), 2 foster parents, and 10 who identified their relationship as “other.”

The scale was comprised of seven contrasting items, such as “My child’s school is dangerous” versus “My child’s school is safe.” Each contrasting statement was positioned at the opposite end of a 7-point continuum, with a score of 1 associated with the negative evaluative statement (e.g., is dangerous) and a score of 7 associated with the positive evaluative statement about the school (e.g., is safe). A confirmatory factor analysis (CFA) was conducted using robust maximum likelihood estimation procedures.

Results: The CFA model specified that the items loaded on a single latent factor, termed Overall School Perception. The factor variance was freely estimated, as was the uniqueness for each item. No covariances between uniquenesses were modeled and data were input using the asymptotic covariance matrix.  The overall fit of the model to the data was reasonably good based on commonly recommended cut off values for evaluating model fit (S-B c² = 33.77, df = 14, p = .00; RMSEA = .032 (90% CI = .018-.046), SRMR = .01; CFI = 1.00, TLI = 1.00). Squared multiple correlations for each item averaged .72.  We also tested factorial invariance of the scale across language version using multigroup CFA procedures. We first tested a baseline model with the seven items loaded on the same latent factor across groups to test configural invariance. Next, we tested for metric invariance by constraining the factor loading to be invariant across groups. Both models fit the data reasonably well and there was a non-significant (p >.05) difference in the Satorra-Bentler Scaled Difference in c² Test.

Conclusions and Implications: Overall, the results of the psychometric testing indicate initial support for the reliability and factorial validity of the CAYCI Parent/Caregiver Overall School Perception Scale. The use of this measure could provide valuable information about parent/caregivers’ beliefs and attitudes about schools; and, in turn, document specific school characteristics, as perceived by parents, as in need of improvement. Future research should examine the use of this measure in multiple settings, and in relation to other variables that may be related to parent/caregiver perceptions of schools.