BEGIN:VCALENDAR VERSION:2.0 PRODID:-//UBC Department of Psychology//NONSGML Events//EN CALSCALE:GREGORIAN X-ORIGINAL-URL:https://psych.ubc.ca/events/event/ X-WR-CALDESC:UBC Department of Psychology - Events BEGIN:VEVENT UID:20181205T0926Z-1544002002.9474-EO-17033-2@137.82.45.12 STATUS:CONFIRMED DTSTAMP:20240328T083538Z CREATED:20181105T233002Z LAST-MODIFIED:20181121T000525Z DTSTART;TZID=America/Vancouver:20181115T123000 DTEND;TZID=America/Vancouver:20181115T143000 SUMMARY: Invited Talk: Philip Chalmers | Quantitative DESCRIPTION: Title: Model-based Measures for Detecting and Quantifying Resp onse Bias Abstract: An important research area in psychometrics is the iden tification and quantification of measurement bias. Measurement bias occurs when one or more items on a psychological test\, survey\, rating scale\, an d so on\, demonstrate favoritism towards at least one group of individuals\ , resulting in composite test […] X-ALT-DESC;FMTTYPE=text/html:
Title: Model-based Measur
es for Detecting and Quantifying Response Bias
Abstract:
An important research area in psychometrics is the identification
and quantification of measurement bias. Measurement bias occurs when one or
more items on a psychological test\, survey\, rating scale\, and so on\, d
emonstrate favoritism towards at least one group of individuals\, resulting
in composite test scores that will ultimately favour one group over anothe
r. However\, while the identification of measurement bias has been studied
using many statistical approaches\, particularly under the topic of differe
ntial item functioning (DIF)\, obtaining optimal quantifications of measure
ment bias in the form of effect sizes has yet to be resolved in the literat
ure. In this talk\, I will discuss a set of model-based effect size measure
s for response bias that (a) boast optimal large-sample statistical propert
ies in terms of efficiency and bias\, (b) do not require ad-hoc assumptions
after models have been fitted\, (c) are applicable to any select item resp
onse model in current use\, and (d) are capable of quantifying response bia
s in item bundles of any size.