Introduction
Introduction Statistics Contact Development Disclaimer Help
ERIC EJ1098651: Better Quality in Assessments: Consideration of Con...
by ERIC
Thumbnail
Download
Web page
This study considers the importance of contextual effects
on the quality of assessments on item bias and
differential item functioning (DIF) in measurement.
Often, in educational studies, students are clustered in
teachers or schools, and the clusters could impact
psychometric issues yet are largely ignored by
traditional item analyses. A statistical model for
incorporating these cluster effects will be presented. By
considering DIF from this perspective, it is possible for
DIF to be consistent or vary across clusters (e.g.,
teachers or schools). By using this model, researchers
can be shown much more detail into the nature and source
of DIF and provide greater quality in the analysis and
use of their assessments. This can be of benefit in
determining whether the nature of DIF is exclusively due
to student attributes, or a particular combination of
student and school attributes. In addition to this, DIF
may exist only among clusters, but not for students. This
is an extra detection not possible with traditional DIF
analysis. For example, in some educational situations,
DIF may not exist among the subjects of interest, but in
certain type of schools. The use of this procedure will
be demonstrated using real assessment data.
Date Published: 2018-10-06 00:12:02
Identifier: ERIC_EJ1098651
Item Size: 9516637
Language: english
Media Type: texts
# Topics
ERIC Archive; Test Bias; Educational ...
# Collections
ericarchive
additional_collections
# Uploaded by
@chris85
# Similar Items
View similar items
PHAROS
You are viewing proxied material from tilde.pink. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.