The Map Is Not the Territory: Construct Names, Items, and Responses in Measures of Mentalizing, Epistemic Trust, and Emotion Regulation
Abstract
Psychology has a measurement problem: different constructs often share the same name, and the same construct is measured under different names. This matters most for constructs theorized as interdependent. In theory, these constructs are defined by name, and their interdependence is posited at that level; in practice, each is measured by a pool of items, often from separate questionnaires, which may overlap or diverge in ways the names do not anticipate. We asked whether construct names correspond to the items that operationalize them, whether items reflect the structure of responses, and whether responses follow the boundaries posited by names or drawn by items. Using reflective functioning, epistemic trust, and difficulties in emotion regulation as a case study, we compared semantic similarities of construct names and items with response correlations in 1,167 emerging adults (aged 18–29, 72% female; item-level analyses preregistered) and 1,068 secondary school students (aged 14–20, 67% female). Construct names best matched their own items in 4 of 11 subscales. Items reflected the structure of responses (r = .62 and .58), also, less closely, across questionnaires (.42 and .37). Responses followed boundaries drawn by items more than those posited by names. A correlation between such measures, usually read as evidence of interdependence, may also arise from items that describe the same experience or from shared method variance. What theory says about a construct thus applies only in part to what its items measure.
Related
- Epistemic trust, mentalizing and extremist attitudes in a German community sample
- Oxysterols Are Not Functionally Equivalent: Cell-Type- and Territory-Specific Responses in Human Models of Atherosclerosis
- Attachment, Emerging Borderline Personality Features and Mentalizing Capacity: A Comparative Analysis in Maltreated and Community Children
- Heterogeneous and conserved radiation responses reveal FOXM1-dependent regulation of microcephaly genes in glioblastoma
- Tumor Microbial Burden Drives Immune Responses through Regulation of Interferon Signaling
BibTeX
@misc{benzi2026territory,
title = {{The Map Is Not the Territory: Construct Names, Items, and Responses in Measures of Mentalizing, Epistemic Trust, and Emotion Regulation}},
author = {Ilaria Maria Antonietta Benzi and Marco Marelli and Laura Antonia Lucia Parolin and Karin Ensink},
howpublished = {PsyArXiv},
year = {2026},
month = oct,
doi = {10.31234/osf.io/g9fhq_v1},
url = {https://osf.io/preprints/psyarxiv/g9fhq_v1/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - The Map Is Not the Territory: Construct Names, Items, and Responses in Measures of Mentalizing, Epistemic Trust, and Emotion Regulation
AU - Ilaria Maria Antonietta Benzi
AU - Marco Marelli
AU - Laura Antonia Lucia Parolin
AU - Karin Ensink
PY - 2026
DA - 2026/10/25
PB - PsyArXiv
DO - 10.31234/osf.io/g9fhq_v1
UR - https://osf.io/preprints/psyarxiv/g9fhq_v1/
AB - Psychology has a measurement problem: different constructs often share the same name, and the same construct is measured under different names. This matters most for constructs theorized as interdependent. In theory, these constructs are defined by name, and their interdependence is posited at that level; in practice, each is measured by a pool of items, often from separate questionnaires, which may overlap or diverge in ways the names do not anticipate. We asked whether construct names correspond to the items that operationalize them, whether items reflect the structure of responses, and whether responses follow the boundaries posited by names or drawn by items. Using reflective functioning, epistemic trust, and difficulties in emotion regulation as a case study, we compared semantic similarities of construct names and items with response correlations in 1,167 emerging adults (aged 18–29, 72% female; item-level analyses preregistered) and 1,068 secondary school students (aged 14–20, 67% female). Construct names best matched their own items in 4 of 11 subscales. Items reflected the structure of responses (r = .62 and .58), also, less closely, across questionnaires (.42 and .37). Responses followed boundaries drawn by items more than those posited by names. A correlation between such measures, usually read as evidence of interdependence, may also arise from items that describe the same experience or from shared method variance. What theory says about a construct thus applies only in part to what its items measure.
ER -