Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Why Do Judgments on Different Person-Descriptive Attributes Correlate With One Another? A Conceptual Analysis With Relevance for Most Psychometric Research

Leising, Daniel ; Borgstede, Matthias ; Burger, Julian ; Zimmermann, Johannes ; Bäckström, Martin LU ; Oltmanns, Joshua ; Freyer, Nele ; Wiedenroth, Anne ; Knischewksi, Paula and Connelly, Brian (2025) In Collabra: Psychology 11(1).
Abstract

Patterns of correlations among judgments of targets on different attributes are the basis for common psychometric procedures such as factor analysis and network modeling. The outcomes of such analyses may shape the images (i.e., theories) that we as scientists have of the phenomena that we study. However, key conceptual issues tend to be overlooked in these analyses, which is especially problematic when the items are descriptions expressed in the natural language. A correlation between judgments on two such attributes may reflect the influences of (a) a common substantive cause, (b) substantive target characteristics on another, (c) semantic redundancy, (d) the perceivers' attitudes toward the targets, (e) the perceivers' formal... (More)

Patterns of correlations among judgments of targets on different attributes are the basis for common psychometric procedures such as factor analysis and network modeling. The outcomes of such analyses may shape the images (i.e., theories) that we as scientists have of the phenomena that we study. However, key conceptual issues tend to be overlooked in these analyses, which is especially problematic when the items are descriptions expressed in the natural language. A correlation between judgments on two such attributes may reflect the influences of (a) a common substantive cause, (b) substantive target characteristics on another, (c) semantic redundancy, (d) the perceivers' attitudes toward the targets, (e) the perceivers' formal response styles, or (f) any mixture of these. We present a conceptual framework integrating all of these mechanisms and use it to connect formerly unrelated strands of theorizing with one another. A lack of awareness regarding the complexity involved may compromise the validity of interpretations of psychometric analyses. We also review the effectiveness of a broad range of solutions that have been proposed for dealing with the various influences, and provide recommendations for future research.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
attitude, factor analysis, measurement, network model, person perception, psychometrics, response style, semantics
in
Collabra: Psychology
volume
11
issue
1
article number
133683
publisher
University of California Press
external identifiers
  • scopus:105004394689
ISSN
2474-7394
DOI
10.1525/collabra.133683
language
English
LU publication?
yes
additional info
Publisher Copyright: © 2025 University of California Press. All rights reserved.
id
d36ff077-a9f0-4313-95c1-7a6adbebe963
date added to LUP
2025-08-06 13:39:32
date last changed
2025-08-06 13:39:32
@article{d36ff077-a9f0-4313-95c1-7a6adbebe963,
  abstract     = {{<p>Patterns of correlations among judgments of targets on different attributes are the basis for common psychometric procedures such as factor analysis and network modeling. The outcomes of such analyses may shape the images (i.e., theories) that we as scientists have of the phenomena that we study. However, key conceptual issues tend to be overlooked in these analyses, which is especially problematic when the items are descriptions expressed in the natural language. A correlation between judgments on two such attributes may reflect the influences of (a) a common substantive cause, (b) substantive target characteristics on another, (c) semantic redundancy, (d) the perceivers' attitudes toward the targets, (e) the perceivers' formal response styles, or (f) any mixture of these. We present a conceptual framework integrating all of these mechanisms and use it to connect formerly unrelated strands of theorizing with one another. A lack of awareness regarding the complexity involved may compromise the validity of interpretations of psychometric analyses. We also review the effectiveness of a broad range of solutions that have been proposed for dealing with the various influences, and provide recommendations for future research.</p>}},
  author       = {{Leising, Daniel and Borgstede, Matthias and Burger, Julian and Zimmermann, Johannes and Bäckström, Martin and Oltmanns, Joshua and Freyer, Nele and Wiedenroth, Anne and Knischewksi, Paula and Connelly, Brian}},
  issn         = {{2474-7394}},
  keywords     = {{attitude; factor analysis; measurement; network model; person perception; psychometrics; response style; semantics}},
  language     = {{eng}},
  month        = {{04}},
  number       = {{1}},
  publisher    = {{University of California Press}},
  series       = {{Collabra: Psychology}},
  title        = {{Why Do Judgments on Different Person-Descriptive Attributes Correlate With One Another? A Conceptual Analysis With Relevance for Most Psychometric Research}},
  url          = {{http://dx.doi.org/10.1525/collabra.133683}},
  doi          = {{10.1525/collabra.133683}},
  volume       = {{11}},
  year         = {{2025}},
}