Scheduled for Research Consortium Measurement, Exercise Physiology and Motor Behavior Poster Session, Friday, April 28, 2006, 8:45 AM - 10:15 AM, Convention Center: Exhibit Hall Poster Area I


Measuring Student Perception of School Connectedness

Nicole J. Smith, Monica Lounsbery, Kevin Crehan and Michelle Weibel, University of Nevada, Las Vegas, Las Vegas, NV

The relationship between health risk behaviors and student perception of school connectedness has been established (Resnick et al., 1997). In addition, the need to develop valid and reliable instruments has been identified (Lounsbery, Gast, & Smith, 2005). The purpose of this study was to design and validate an instrument to measure student perception of school connectedness. Student perception of school connectedness has been defined as students' belief that adults in the school care about them as individuals in addition to caring about their learning (Blum & Libbey, 2004, Wingspread Declaration, 2004). Pilot results from Planned Approach to Healthier Schools (PATHS; Lounsbery et al., 2005) showed that five qualities might impact students' connectedness including perceptions of (1) teachers' caring, (2) teachers' fairness, (3) teachers' school belonging, (4) students' acceptance, and (5) students' school belonging. A 43-item instrument comprised of items designed to tap these five qualities was developed and administered to a sample of 363 students in grades 6-12. An item analysis was conducted to identify items with possible faults. Based on this analysis, two items were dropped and a principal components analysis (PCA) was conducted for the remaining 41 items. Kaiser Meyer-Olkin (.923) and Bartlett's Test of Sphericity (p < .05) indicated the data was suitable for PCA. Initial results indicated 10 characteristics with eigenvalues > 1.0 that explained 60.03% of the variance. The scree test showed two components above the “elbow.” Items with loadings less than .5 were eliminated and a second PCA was conducted using a two-component model. The resulting 21-item solution explained 47.2% of the variance. Although the two components explain less total variance than the initial 10-component model, the simpler model is judged to be worth the cost. Based on inspection of item content the two components were identified as (1) teacher acceptance and caring and (2) student acceptance and school belonging.
Keyword(s): measurement/evaluation

Back to the 2006 AAHPERD National Convention and Exposition