Multilevel Factor Structure of a Middle School Teacher Effectiveness Survey

• In Event: Roundtable Session 46
  In Roundtable Session: Applications of Factor Analysis and Structural Equation Modeling

Abstract

Although student evaluations of teachers are routinely used for measuring teacher effectiveness, most studies have focused on higher education and psychometric studies have often ignored the naturally nested data structure. However, the use of multilevel factor analyses in factor analytic studies is crucial since the constructs being investigated are at the group level and not at the individual level. The current study illustrates the use of multilevel factor analyses in investigating the factor structure of a recently developed teacher effectiveness survey using a large sample of 4,706 middle school students and 152 teachers. Results show how factor loadings are not invariant across levels and shows how ignoring the clustered nature of the data can yield misleading results.

Authors

• Christi Crosby Bergin, University of Missouri

• Francis Howard Lim Huang, University of Missouri

• Chia-Lin Tsai, University of Missouri

• Sara Bernice Chapman, University of Missouri - Columbia