Abstract
ABSTRACTA first application of multilevel latent class analysis (MLCA) to educational large-scale assessment data is demonstrated. This statistical technique addresses several of the challenges that assessment data offers. Importantly, MLCA allows modeling of the often ignored teacher effects and of the joint influence of teacher and student variables. Using data from the 2011 assessment of Dutch primary schools’ mathematics, this study explores the relation between the curriculum as reported by 107 teachers and the strategy choices of their 1,619 students, while controlling for student characteristics. Considerable teacher effects are demonstrated, as well as significant relations between the intended as well as enacted curriculum and students’ strategy use. Implications of these results for both more theoretical and practical educational research are discussed, as are several issues in applying MLCA and possibilities for applying MLCA to different types of educational data.
Highlights
Recently, the technique of Latent class analysis (LCA) has been extended to accommodate an additional hierarchical level (Vermunt, 2003): the nesting of variables within individuals is included in the model, and the nesting of individuals in some higher-level group
We describe a first application of multilevel LCA (MLCA) to educational large-scale assessment data
The third challenge that MLCA addresses is the inherent multilevel structure of educational data
Summary
Recently, the technique of LCA has been extended to accommodate an additional hierarchical level (Vermunt, 2003): the nesting of variables within individuals is included in the model, and the nesting of individuals in some higher-level group (e.g., students within school classes). A first challenge that many large-scale assessments offer is that they employ so-called incomplete designs: the complete item set is too large to be administered in full to students, and is decomposed into smaller subsets Relating these subsets to each other is difficult using traditional techniques, but is possible using a latent variable to which all items are related (Embretson & Reise, 2000; Hickendorff, Heiser, Van Putten, & Verhelst, 2009), such as the latent class variable in LCA. Zumbo et al (2015) recently proposed an ecological model of item responding where responses are influenced by contextual variables at various levels: characteristics of the test, of the individual, of the teacher and school, of the family and ecology outside of school, and of the larger community Based on this model, the authors demonstrate ecologically moderated differential item functioning (DIF) where different factors in this broader context play a role. We investigate the relation between the curriculum on the one hand and students’ use of solution strategies on the other (while controlling for student characteristics), and describe the technique of MLCA and some of the challenges in its application in more detail
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
