Individual and Contextual Effects on Science Identity Among American Ninth-Grade Students (High School Longitudinal Study 2009): Hierarchical Linear Modeling

• In Event: Marriott Roundtable Session 11
  In Roundtable Session: Multilevel Modeling Applications

Abstract

This study explored the variability in science identity that is accounted by student-level (i.e., gender, science self-efficacy, science utility, and science interest) and school-level predictors (i.e., school type and locale). Participants were 9th grade students (N = 5,027). Data was obtained from large-scale database (HSLS:09). Six models were tested using HLM. Findings showed significant variability in science identity that was mainly explained by student-level predictors (i.e., 40% of variance). Only 5% of variance was accounted by school differences. This identity was bolstered by higher level of self-efficacy, utility, interest and SES. Rural and towns schools had lower average identity relative to city schools. Private schools had higher identity compared with public schools. These findings implied practical suggestions, nourishing science identity.

Author

• Amal Alhadabi, Kent State University - Kent