Anchoring vignette: the good, the bad and the ugly

Self-report Likert-type personality measures have long co-existed with response styles and reference group biases. These issues, when unaddressed, can influence the reliability, validity, and cross-cultural comparability of personality scales. A posterior bias-correction method promised to address these challenges called anchoring vignette was tested in the current study. Using three samples collected in China and the US, the author examined the bias-correction power of anchoring vignette in the following scenarios: 1) when vignettes are written in a domain-specific vs. domain-general fashion. 2) when vignette assumptions are violated vs. not violated. 3) when vignettes are compared to forced choice measures, -- an a priori bias-prevention method. Results showed that domain specificity influenced the validity of vignette-recoded data, with domain-specific vignette being the more desirable approach. Measurement properties of the vignette-recoded data were negatively influenced by the violation of vignette equivalence assumption. And the overall bias-correction power of anchoring vignette is inferior to forced choice.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Psychology

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2023

Author dc:creator

• Chen, Yueyang

Contributors dc:contributor

• Drasgow, Fritz
• Rounds, James

• Personality Measure
• Response Bias
• Anchoring Vignette
• Forced Choice

Statement dc:rights

• Copyright 2023 Yueyang Chen

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/120476

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F120476