Integrated analysis of content and construct validity of psychometric instruments

Establishing adequacy of psychometric properties of an instrument involves acquisition and evaluation of evidence based on item content and internal structure. Content validity evidence consists of subject matter experts providing quantitative ratings of the extent to which items are a representativ...

Full description

Saved in:
Bibliographic Details
Published in:Quality & quantity 2013, Vol.47 (1), p.57-78
Main Authors: Gajewski, Byron J., Price, Larry R., Coffland, Valorie, Boyle, Diane K., Bott, Marjorie J.
Format: Article
Language:English
Subjects:
Anniversaries
Bayesian analysis
Bayesian method
Classical test theory
Cultural Change
Data analysis
Discriminant analysis
Distribution
Evidence
History, theory and methodology
Item response theory
Methodology
Methodology of the Social Sciences
Methods
Nursing Homes
Probability distribution
Psychological factors
Psychometrics
Quantitative analysis
Quantitative psychology
Ratings & rankings
Social Sciences
Sociology
Sociology of health and medicine
Statistical inference
Studies
Validation studies
Validity
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Establishing adequacy of psychometric properties of an instrument involves acquisition and evaluation of evidence based on item content and internal structure. Content validity evidence consists of subject matter experts providing quantitative ratings of the extent to which items are a representative sample of targeted domain. Evidence of internal structure includes factor analytic studies and examination of item interrelationships based on item responses from participants. Although subject matter expert ratings and participant response data are traditionally analyzed separately, each serves to inform the other in important ways. We propose integrating subject matter experts’ and participants’ data seamlessly to establish a unified model of validity evidence. The approach is applied to an instrument designed to measure nursing home culture change (i.e., resident-centered care). The proposed method has been demonstrated to be useful with a posterior distribution resulting in stable estimates of psychometric parameters superior to traditional analytic approaches. To illustrate the efficacy of the methodology, we present a simulation study and discuss its place in psychometric methods.
ISSN:0033-5177
1573-7845
DOI:10.1007/s11135-011-9503-4