Properties (15)
Id d366885b-c580-4f2b-8d03-d33cb7269cc2
Year 2025
Abstract ABSTRACT Background The Duke Misophonia Questionnaire (DMQ) is one of few psychometrically constructed scales designed to measure misophonia, a condition characterized by heightened sensitivity to specific auditory stimuli, resulting in strong emotional, physiological, and behavioral responses. While prior research has evaluated the dimensionality of individual subscales, a comprehensive psychometric analysis of the entire DMQ has not been conducted. Objectives This study aimed to extend the psychometric evaluation of the DMQ using contemporary psychometric methodologies to enhance its consistency, stability, and applicability in future research. Methods Two samples of university students were recruited online via a research participation pool (Sample 1: N = 318; Sample 2: N = 424; both predominantly female and White, aged 18–30). A novel, empirically derived five‐factor model (beliefs, coping, impairment, aversion, and anxiety) and the theoretically defined eight‐factor model, developed from clinical and theoretical foundations, were assessed. The five‐factor model was further examined using multidimensional item response theory (MIRT) to gain insights into item‐ and test‐level functioning. Results Both models demonstrated adequate fit, though the five‐factor model was more parsimonious and fit slightly better. Neither model contained cross‐loadings. The five‐factor model was further evaluated through MIRT which found that the DMQ provides the most information at moderate to high levels of misophonia severity. This analysis provided valuable information regarding the functioning and efficiency of the DMQ at both item and test levels. Conclusion The results contribute to the refinement of the DMQ by identifying a more parsimonious psychometric structure, promoting greater consistency and stability in its use for future misophonia research.
Cssclasses
- purpleRed
Aliases
- (Bain et al., 2025)
Import Date 06-06-2026
Template Version Notes •corrected error in rendering the linked PDF in metadata section • added the tag “noted” to the definition of comments to ignore ( like with the “in another note” tag. this is so that when making notes on chapters of a textbook, i can still export things for manual addition to the obsidian note without having to keep track of what has already been added ). Also moved the version notes to the yaml section • this version added the bookmarks section to auto collect all bookmarks. it also dropped the template auto fill of the initial source field because we never got it working and just overwrote it manually anyway. this version also added logic to handle “in another note” tag. annotations marked with this are completely ignored. used when i have information covered in a bigger note, but still want to highlight that part in text for visual clarity (so as to avoid duplicates of the same information) • running the import more then once will overwrite the file, so do not import again if you have made changes in obsidian that are not in the zotero article.
Dv Authors Catherine M. Bain, Jordan E. Norris, Alex Conley, Patrick D. Manapat, Lauren E. Ethridge,
Dv Initial Source Misophonia MOC
Dv Status #status/Bare-bones-review
Dv Limitations • Nonclinical, university-based convenience sample limits clinical utility; no formal diagnostic criteria applied. • Sample may underrepresent individuals with highly severe symptoms. • Only self-report measures used. • Only one misophonia measure (MQ) used for convergent validity.
Dv Future Directions • Examine the five-factor structure in clinical samples with determined misophonia. • Conduct differential item functioning (DIF) and MIMIC analyses to assess measurement invariance across groups. • Use exploratory structural equation modeling (ESEM) to evaluate cross-loadings in both models. • Incorporate biological/neurophysiological measures beyond self-report. • Validate against additional misophonia measures (e.g., Amsterdam Misophonia Scale, Misophonia Assessment Questionnaire). • Develop items informative at lower misophonia levels to expand scale scope.
Dv Relevance • study did a psychometric analysis of the DMQ, finding evidence for a five-factor model
Notes Audited 2026-06-06T23:35:00
Audit Notes audited as review was done
•
A Psychometric Analysis of the Duke Misophonia Questionnaire
Status:: #status/Bare-bones-review
Metadata
- Item Type: #journalArticle
- Authors:: Catherine M. Bain, Jordan E. Norris, Alex Conley, Patrick D. Manapat, Lauren E. Ethridge,
- Date:
- Date Added: 1755661030000
- URL: https://onlinelibrary.wiley.com/doi/10.1002/jclp.70028
- DOI: 10.1002/jclp.70028
- Cite Key: bainPsychometricAnalysisDuke2025
- Topics: [object Object]
- Tags:: #Bare-bones-review-🦴, #clinical, #DMQ, #measures, #Misophonia,
- PDF Attachments
Abstract
| Background/Purpose | Background- • Misophonia lacks formal diagnostic criteria, reducing standardization across clinical and research studies. • Limited number of psychometrically developed or evaluated misophonia scales exists. • Poor psychometric quality can lead to under-/overestimation of effects and inaccurate construct conclusions. • DMQ developed via refinement of items from prior misophonia measures with clinician, stakeholder, and patient/family input. • DMQ development used within-subscale EFA and IRT but no complete dimensionality assessment. • DMQ scored on nine theoretically derived subscales; overall scale dimensionality never assessed. • Analyzing subscales separately fails to account for inter-subscale correlations, causing information loss and biased factor loadings. • Several DMQ theoretical subscales required complex bi-factor models during development. • Initial constrained CFA approach may have produced over- or underfactoring; overfactoring more likely. • Item analysis during development was within-factor only. Purpose- • Assess dimensionality of the DMQ to determine factor structure via EFA/CFA approach. • Perform exploratory item analysis via MIRT framework given multidimensional structure. • Conduct first psychometric evaluation of the theory-based eight-factor structure via CFA. • Compare theory-defined eight-factor model against most plausible empirically derived model. • Perform content analysis to understand resulting factors and regression analysis for validity evidence. |
| Participants/samples | • Two samples from University of Oklahoma Psychology Research Pool (SONA), recruited online. • Sample 1 (exploratory): N = 318; predominantly female; predominantly White; aged 18–30. • Sample 2 (validation): N = 424; predominantly female; predominantly White; aged 18–26. • Combined full sample: N = 742. • Exclusions: failed attention check; under age 18; prior participation (Sample 2). • No missing data in either sample. • Time-window-based split into exploratory and validation sets. |
| Measures | • DMQ – 86-item self-report; nine theoretical dimensions; trigger items (binary) excluded, leaving 70 nontrigger items; Likert frequency scale (never–always/almost always). • MQ (Misophonia Questionnaire) – 20-item three-part measure; symptoms scale, emotions/behaviors scale, single severity item; used for convergent validity. • ASP (Adolescent and Adult Sensory Profile) – self-report four-factor sensory processing measure (low registration, sensory sensitivity, sensation avoiding, sensation seeking); used for divergent validity. • ASQ (Anxiety Symptoms Questionnaire) – self-report frequency/severity of anxiety symptoms; cognitive and somatic items; used for divergent validity. |
| Procedures | • EFA conducted on Sample 1 using polychoric correlations, OLS estimation, oblique (direct quartimin) rotation; psych package in R. • Factor number determined by scree plot and parallel analysis; models with two through eight factors estimated. • Solution selection based on simple structure, item retention, and construct validity (loading threshold 0.32). • CFA via lavaan with DWLS/WLSMV estimator and polychoric correlations; fit indices χ², TLI, CFI, SRMR, RMSEA. • Plausible models fit as CFA in Sample 1, validated in Sample 2, final estimates from combined full sample. • MGRM fit to full sample using mirt package; MH-RM estimation algorithm. • Internal consistency via coefficient alpha and omega. • EAP scores entered as simultaneous predictors in regression analyses for convergent (MQ) and divergent (ASQ, ASP) validity. |
| Results | • Scree plot suggested two/three factors; parallel analysis suggested eight factors. • Two-, three-, four-, six-, seven-, and eight-factor models failed simple structure and/or lost items with cross-loadings. • Five-factor model contained no cross-loadings and retained all items; labeled beliefs, coping, impairment, aversion, anxiety. • Five-factor model demonstrated adequate fit in validation and full sample. • Content analysis: beliefs factor mapped onto original beliefs subscale; coping collapsed prior coping subscales; impairment absorbed an item about hurting the sound-maker; aversion captured negative emotionality and sound avoidance/stopping; anxiety separated worry, somatic arousal, and powerlessness items. • MIRT slopes varied across items, indicating differential relation to the misophonia construct. • DMQ more precise at moderate-to-high misophonia levels; functions less well at low levels; aversion factor evenly distributed across the range while beliefs provides little information at low levels. • Internal consistency high across all five factors. • Convergent validity (MQ): all factors except anxiety predicted severity; all factors predicted emotions/behaviors; all factors except beliefs predicted symptoms. • Divergent validity (ASQ): all factors except coping predicted anxiety total; substantial unexplained variance in DMQ anxiety. • Divergent validity (ASP): factors differentially predicted sensory subscales; partial divergence with substantial unexplained variance in anxiety and beliefs. • Eight-factor theoretical model demonstrated adequate fit; no content analysis repeated. • Five-factor model more parsimonious and slightly better fitting than eight-factor model. |
| Discussion/Limits | Conclusion- • This represents the first complete dimensionality assessment of the overall DMQ. • A novel, more parsimonious five-factor conceptualization of misophonia retains all 70 substantive items. • Structural evidence supports the theoretically derived eight-factor model from Rosenthal et al. (2021). • Separation of a distinct anxiety factor may help differentiate general anxiety from misophonia-specific impairment. • The aversion factor captures core experiential features of misophonia and may index intervention effects on aversive reactions. • Five-factor model is recommended for screening, outcome tracking, and large-scale research; eight-factor model suits granular symptom profiling and individualized treatment targeting. • Misophonia appears conceptually distinct, though partially overlapping, with anxiety and sensory processing constructs. • All 70 items contributed differential information; DMQ functions best for higher-severity individuals. • Misophonia is likely best understood as a spectrum including nonpathological emotional responses. |
| Limitations:: | • Nonclinical, university-based convenience sample limits clinical utility; no formal diagnostic criteria applied. • Sample may underrepresent individuals with highly severe symptoms. • Only self-report measures used. • Only one misophonia measure (MQ) used for convergent validity. |
| Future Directions:: | • Examine the five-factor structure in clinical samples with determined misophonia. • Conduct differential item functioning (DIF) and MIMIC analyses to assess measurement invariance across groups. • Use exploratory structural equation modeling (ESEM) to evaluate cross-loadings in both models. • Incorporate biological/neurophysiological measures beyond self-report. • Validate against additional misophonia measures (e.g., Amsterdam Misophonia Scale, Misophonia Assessment Questionnaire). • Develop items informative at lower misophonia levels to expand scale scope. |
| Relevance:: | • study did a psychometric analysis of the DMQ, finding evidence for a five-factor model |
Notes
key terms and concepts
-
Misophonia – a condition characterized by a decreased tolerance for specific auditory stimuli (triggers) followed by strong, negative emotional, physiological, and behavioral responses to trigger exposure.
-
DMQ (Duke Misophonia Questionnaire) – an 86-item self-report instrument developed to comprehensively assess misophonia, consisting of nine theoretically derived dimensions.
-
EFA (exploratory factor analysis) – a less constrained version of the common factor model used at early/exploratory phases of scale development to prevent under- and overfactoring.
-
CFA (confirmatory factor analysis) – a more constrained factor analytic technique placing a priori constraints on factor number and structure.
-
Simple structure – a factor solution where each item loads strongly onto only one factor and each factor has multiple items loading strongly onto it.
-
IRT (item response theory) – a collection of latent variable models seeking to uncover the underlying process influencing responses to observed variables, driven by properties of individuals (θ) and items.
-
MIRT (multidimensional item response theory) – multidimensional adaptations of IRT models able to account for correlations between factors.
-
MGRM (multidimensional graded response model) – the extension of Samejima's graded response model incorporating multiple dimensions of latent traits.
-
Information – Fisher information; the metric IRT uses to characterize precision/reliability, provided at item level (IIFs) and test level (TIFs).
-
EAP (expected a posteriori) – scale score estimates derived from the mean of the posterior distribution.
-
θ (theta) – level of the latent variable or construct of interest (misophonia in this context).
my engagement notes
my questions
my comments
Bookmarks
reading notes
contextual statement
In a 2025 study, Bain attempted to conduct a complete psychometric evaluation of the Duke Misophonia Questionnaire by applying an EFA/CFA dimensionality assessment and MIRT item analysis to two online university student samples (combined N = 742), using the MQ, ASP, and ASQ as validation measures.