The Impact of Setting Accommodation on Large-scale Assessment of English Language Learners with and without Learning Disabilities: Balanced vs. Unbalanced Data in Latent Class Analyses

Pei-Ying Lin, Yu-Cheng Lin

Abstract


This exploratory study examined the effect of setting accommodation on Grade 6 students’ number sense and numeration skill in relation to whether they speak a language other than English at home, have a learning disability, and received setting accommodation for a large-scale math assessment. A set of latent class analyses was conducted to investigate students’ response patterns and latent class memberships in both balanced and unbalanced data; three- and two-way ANOVA were also performed to examine the effect. The results suggest that a 4-class model with a covariate and indirect effects have better model fit to the balanced and unbalanced data. Percentage correct and missing response rate in two datasets were also compared and discussed. The results indicate that number sense and numeration skill of accommodated Grade 6 students with LD was comparable to their non-accommodated peers with LD when their response patterns were taken into account. For non-disabled groups, non-accommodated Grade 6 students outperformed their accommodated counterparts. The language spoken at home by students was not a significant indicator of the effect. The implication for data design of latent class analysis and parametric statistics for large-scale data was also discussed in this paper.

Full Text:

PDF


DOI: http://dx.doi.org/10.5296/jse.v3i2.3371

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.

To make sure that you can receive messages from us, please add the 'macrothink.org' domain to your e-mail 'safe list'. If you do not receive e-mail in your 'inbox', check your 'bulk mail' or 'junk mail' folders.Copyright © Macrothink Institute ISSN 2162-6952'Macrothink Institute' is a trademark of Macrothink Institute, Inc.