Bifactor Structures Research Articles

Psychometric evaluation of the PCL-5: assessing validity, diagnostic utility, and bifactor structures

ABSTRACT Background: The changes DSM-5 brought to the diagnostic criteria for posttraumatic stress disorder (PTSD) resulted in revising the most widely used instrument in assessing PTSD, namely the Posttraumatic Checklist for DSM-5 (PCL-5). Objective: This study examined the psychometric properties of the Romanian version of the PCL-5, tested its diagnostic utility against the Structured Clinical Interview for DSM-5 (SCID-5), and investigated the latent structure of PTSD symptoms through correlated symptom models and bifactor modelling. Method: A total sample of 727 participants was used to test the psychometric properties and underlying structure of the PCL-5 and 101 individuals underwent clinical interviews using SCID-5. Receiver operating characteristic curve (ROC) analyses were performed to test the diagnostic utility of the PCL-5 and identify optimal cut-off scores based on Youden's J index. Confirmatory Factor Analyses (CFAs) and bifactor modelling were performed to investigate the latent structure of PTSD symptoms. Results: Estimates revealed that the PCL-5 is a valuable tool with acceptable diagnostic accuracy compared to SCID-5 diagnoses, indicating a cut-off score of >47. The CFAs provide empirical support for Anhedonia, Hybrid, and bifactor models. The findings are limited by using retrospective, self-report data and the high percentage of female participants. Conclusions: The PCL-5 is a psychometrically sound instrument that can be useful in making provisional diagnoses within community samples and improving trauma-informed practices.

Applying SEM, Exploratory SEM, and Bayesian SEM to Personality Assessments

Despite the importance of demonstrating and evaluating how structural equation modeling (SEM), exploratory structural equation modeling (ESEM), and Bayesian structural equation modeling (BSEM) work simultaneously, research comparing these analytic techniques is limited with few studies conducted to systematically compare them to each other using correlated-factor, hierarchical, and bifactor models of personality. In this study, we evaluate the performance of SEM, ESEM, and BSEM across correlated-factor, hierarchical, and bifactor structures and multiple estimation techniques (maximum likelihood, robust weighted least squares, and Bayesian estimation) to test the internal structure of personality. Results across correlated-factor, hierarchical, and bifactor models highlighted the importance of controlling for scale coarseness and allowing small off-target loadings when using maximum likelihood (ML) and robust weighted least squares estimation (WLSMV) and including informative priors (IP) when using Bayesian estimation. In general, Bayesian-IP and WLSMV ESEM models provided noticeably best model fits. This study is expected to serve as a guide for professionals and applied researchers, identify the most appropriate ways to represent the structure of personality, and provide templates for future research into personality and other multidimensional representations of psychological constructs. We provide Mplus code for conducting the demonstrated analyses in the online supplement.

Revisiting the academic self-concept transcultural measurement model: The case of Spain and China.

Modelling academic self-concept through second-order factors or bifactor structures is an important issue with substantive and practical implications; besides, the bifactor model has not been analysed with a Chinese sample and cross-cultural studies in the academic self-concept are scarce. Likewise, latent structure validity evidence using network psychometrics has not been carried out. The aim of this study is twofold: to analyse (1) the internal structure of ASC through the Self-Description Questionnaire II-Short (SDQII-S) in Chinese and Spanish samples using two approaches, structural equation modelling and network psychometrics conducting an exploratory graph analysis; and (2) the measurement invariance of the best model across countries and investigate the cross-cultural differences in ASC. The sample was composed by 651 adolescents. Seven models of ASC were tested. Results supported the multi-dimensional nature of the data as well as the reliability. The best-fitted model for the two subsamples was the three-factor ESEM model, but only the configural invariance of this model was supported across countries. The graph function shows that the school dimension appears more related to the verbal factor in the Spanish subsample and to the math dimension in the Chinese subsample. Likewise, the relationship between verbal and math factors in Spanish students is non-existent, but this connection is more relevant for Chinese students. These two differences may be behind the difficulty in finding invariance using SEM models. It is a question of the construct's nature, less related to analytical phenomena, and deserves deeper discussion.

A bifactor model supports unidimensionality of the International Knee Documentation Committee Subjective Knee Form in young active patients with anterior cruciate ligament tears: a retrospective analysis of a randomized controlled trial

BackgroundThe International Knee Documentation Committee Subjective Knee Form (IKDC) is the most highly recommended patient reported outcome measure for assessing patients with anterior cruciate ligament (ACL) injuries and those undergoing ACL reconstruction (ACLR) surgery. The IKDC was developed as a unidimensional instrument for a variety of knee conditions. Structural validity, which determines how an instrument is scored, has not been definitively confirmed for the IKDC in respondents with ACL injuries, and in fact an alternative two-factor/subscale structure has been proposed in this population. The purpose of this study was to determine the most appropriate structure and scoring system for the IKDC in young active patients following ACL injury.MethodsIn total, 618 young patients deemed at high risk of graft rupture were randomized into the Stability 1 trial. Of the trial participants, 606 patients (98%) completed a baseline IKDC questionnaire used for this analysis. A cross sectional retrospective secondary data analysis of the Stability 1 baseline IKDC data was completed to assess the structural validity of the IKDC using exploratory and confirmatory factor analyses. Factor analyses were used to test model fit of the intended one-factor structure, a two-factor structure, and alternative four-factor and bifactor structures (i.e., a combination of a unidimensional factor with additional specific factors) of the IKDC, in a dataset of young active ACL patients.ResultsThe simple one-factor and two-factor structures of the IKDC displayed inadequate fit in our dataset of young ACL patients. A bifactor model provided the best fit. This model contains one general factor that is substantially associated with all items, plus four secondary, more specific content factors (symptoms, activity level, activities of daily living, and sport) with generally weaker associations to subsets of items. Although the single-factor model did not provide unambiguous support to unidimensionality of the IKDC based on fit indices, the bifactor model supports unidimensionality of the IKDC when covariance between items with similar linguistic structure, response options, or content are acknowledged.ConclusionsOverall, findings of a bifactor model with evidence of a reliable general factor well defined by all items lends support to continue interpreting and scoring this instrument as unidimensional. This should be confirmed in other samples. Clinically, based on these findings, the IKDC can be represented by a single score for young active patients with ACL tears. A more nuanced interpretation would also consider secondary factors such as sport and activity level.Trial registrationThe Stability 1 trial for which these data were collected was registered on ClinicalTrial.gov (NCT02018354).

Dimensionality assessment in bifactor structures with multiple general factors: A network psychometrics approach.

The accuracy of factor retention methods for structures with one or more general factors, like the ones typically encountered in fields like intelligence, personality, and psychopathology, has often been overlooked in dimensionality research. To address this issue, we compared the performance of several factor retention methods in this context, including a network psychometrics approach developed in this study. For estimating the number of group factors, these methods were the Kaiser criterion, empirical Kaiser criterion, parallel analysis with principal components (PAPCA) or principal axis, and exploratory graph analysis with Louvain clustering (EGALV). We then estimated the number of general factors using the factor scores of the first-order solution suggested by the best two methods, yielding a "second-order" version of PAPCA (PAPCA-FS) and EGALV (EGALV-FS). Additionally, we examined the direct multilevel solution provided by EGALV. All the methods were evaluated in an extensive simulation manipulating nine variables of interest, including population error. The results indicated that EGALV and PAPCA displayed the best overall performance in retrieving the true number of group factors, the former being more sensitive to high cross-loadings, and the latter to weak group factors and small samples. Regarding the estimation of the number of general factors, both PAPCA-FS and EGALV-FS showed a close to perfect accuracy across all the conditions, while EGALV was inaccurate. The methods based on EGA were robust to the conditions most likely to be encountered in practice. Therefore, we highlight the particular usefulness of EGALV (group factors) and EGALV-FS (general factors) for assessing bifactor structures with multiple general factors. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Supplemental Material for Dimensionality Assessment in Bifactor Structures With Multiple General Factors: A Network Psychometrics Approach

Supplemental Material for Dimensionality Assessment in Bifactor Structures With Multiple General Factors: A Network Psychometrics Approach

A preliminary examination of the Anxiety Sensitivity Index-3 factor structure in older adults

BackgroundAnxiety sensitivity (AS), as measured by the Anxiety Sensitivity Index-3 (ASI-3), exhibits three-factor and bifactor structures for younger adults. Less is known about the scale's structure within older adult samples. MethodsWe explored the ASI-3's factor structure in a sample of 135 older adults who completed the ASI-3 alongside measures of anxiety, general worry, dementia worry, and depression as part of a larger study. Structural equation modeling was used to determine the factor structure of the ASI-3 in this sample. ResultsA bifactor model was identified and factor loadings supported a general factor and specific physical and cognitive subfactors, but not the social concerns subfactor. Whereas the general factor was associated with anxiety, general worry, dementia worry, and depression, each subfactor had differential associations with these mood/anxiety variables. In particular, the cognitive concerns subfactor was strongly associated with depression and dementia worry. LimitationsGeneralizability of these findings is limited by a homogenous sample. ConclusionsThe observed factor structure of the ASI-3 in our sample aligns with increases in physical and cognitive health concerns during aging. Associations between the cognitive concerns subfactor and dementia worry suggest that this facet of the ASI-3 may be measuring age-related health concerns rather than concerns specific to anxiety symptoms. Future studies should examine whether AS cognitive concerns and dementia worry are overlapping constructs particularly for individuals concerned about their age-related cognitive changes. Age differences in AS and its correlates should be further studied to identify better methods of assessing for AS across the lifespan.

Levels of Personality Functioning Questionnaire 12-18 (LoPF-Q 12-18): Factor Structure, Validity, and Clinical Cut-Offs.

The Levels of Personality Functioning Questionnaire 12-18 (LoPF-Q 12-18) is the only self-report measure informed by the Level of Personality Functioning (Diagnostic and Statistical Manual of Mental Disorders [5th ed.; DSM-5; American Psychiatric Association, 2013]) Alternative Model of Personality Disorders developed for adolescents. The present investigation includes two studies evaluating the English LoPF-Q 12-18. In Study 1, single-factor and bifactor structures (unidimensional severity criterion and four specific factors: identity, self-direction, empathy, intimacy) were evaluated in an ethnically diverse community sample (N = 453; age 10-18; 57% female). Study 2 used a community control (n = 298; age 10-18; 54.4% female) and clinical sample (n = 94; age 11-18; 58.5% female) to examine reliability, validity, and clinical utility. Study 1 results supported the bifactor model, with a robust general factor and little multidimensionality caused by the group factors, suggesting an essentially unidimensional structure. Study 2 revealed good internal consistency and construct validity and provided clinical cut-offs, supporting the use of the LoPF-Q 12-18 total score in research and clinical applications.

Construct Validity and Test–Retest Reliability of the Test of Advanced Movement Skills with a Dual-Outcome Scoring System

ABSTRACT The objective was to verify the construct validity and test–retest reliability of the Test of Advanced Movement Skills (TAMS) with an innovative dual-outcome scoring system. Three statistical approaches – confirmatory factor analysis (CFA), exploratory structural equation modeling (ESEM), and item response theory analysis (IRT) – were applied to the data collected from 88 participants aged 9–12 years. Furthermore, 10 of them were randomly selected to examine the test–retest reliability by calculating intraclass correlation coefficients (ICC). The synthesis of findings from CFA, ESEM, and IRT analyses provides supportive evidence that two- and bi-factor structures demonstrated better model-data fit compared to the one-factor structure. An excellent level of test–retest reliability was confirmed in the TAMS (ICC = .96) and anthropometric instruments (ICC = .986–.998). We conclude that a composite score derived from the TAMS is sufficient and practical to measure one’s general competence in advanced movement skills during late childhood.

Structural Validity of the Hospital Anxiety and Depression Scale Among Family Members of Critically Ill Patients.

The Hospital Anxiety and Depression Scale (HADS) is commonly used clinically and scientifically among surrogate decision makers for critically ill patients. The validity of the HADS has been scrutinized, but its use among surrogate decision makers has not been examined. To examine the structural validity of the HADS. This study was a secondary analysis of data obtained from a 3-arm randomized controlled trial of a decision support intervention. Participants were recruited from 6 intensive care units at a tertiary medical center in Northeast Ohio. Participants were adult surrogate decision makers for critically ill, cognitively impaired adults who were not expected to be discharged from the intensive care unit within the subsequent 48 hours. The fit of 2-factor, 3-factor, and bifactor structures of the HADS was tested with confirmatory factor analysis. The bifactor structure, possessing a general psychological distress factor and anxiety and depression group factors, showed a superior fit and met a priori thresholds for acceptable model fit. The general psychological distress factor accounted for more than 75% of the common variance in the HADS items. Confirmatory factor analysis provided evidence supporting a bifactor structure of the HADS. In this sample, the instrument validly measures psychological distress rather than distinct symptoms of anxiety and depression. Replication of these results is encouraged, and use of alternative measures is recommended when measuring distinct symptoms of anxiety and depression among surrogate decision makers for critically ill patients.

Psychometric Properties and Factor Structures of the CERQ, DERS, and RESE Measures: A Bifactor Approach

This study aims to simultaneously compare the psychometric properties and examine the factor structures of 3 emotion regulation (ER) strategy scales using a bifactor approach. Due to good reliability and validity, extensive use, and the same scoring method, the Cognitive Emotion Regulation Questionnaire, Difficulties in Emotion Regulation Scale, and Regulatory Emotional Self-Efficacy Scale were used to assess ER strategies in 1,036 Chinese respondents. A bifactor confirmatory factor analysis was designed to address the multidimensionality of the factor structure, and the corresponding bifactor structures were then applied in a subsequent bifactor multidimensional item response theory (MIRT) analysis. Finally, bifactor MIRT was used to compare the psychometric properties of the 3 measures. The results indicated that bifactor structures were appropriate for the 3 ER strategy measures, which performed well overall. Different measures provide the highest accuracy for specific groups and designs. Some strengths and limitations of this article are discussed.

Psychometric properties of TAS, TAI, FAT test anxiety scales 6 in Chinese university students: a Bifactor IRT study

In this study, the psychometric properties of three commonly used rating scales of test anxiety were examined, including the test anxiety inventory (TAI), the test anxiety scale (TAS) and the Friedman-Bendas Test Anxiety Scale (FAT). Under the framework of item response theory (IRT), the Bifactor multi-dimensional item response model was employed to compare the psychometric properties of the three scales. Results showed that the Bifactor structures were suitable for the three scales, which were then used in the subsequent Bifactor multidimensional item response theory analysis. Although the three commonly used TA scales were likely to measure the same underlying construct—test anxiety, they had very different psychometric properties. The findings of the Bifactor Multi-IRT provided suggestions for determining which scale to use in a given study design: the TAI and the FAT evaluated information at greatly overlapping ranges; however, the TAI, performing a litter better at the same levels of severity of TA, may be a good choice when we recruit those with various levels of TA severity to ensure a high precision. What’s more, FAT may be a good choice for measuring those with moderate TA severity. Meanwhile, the TAS provided more information at the lower level of TA symptomatology, which was to say, TAS was more suitable for epidemiological TA studies and for measuring those with lower TA severity.

Factor structure and gender invariance of the Beck Depression Inventory – second edition (BDI-II) in a community-dwelling sample of adults

ABSTRACT Objective: The main purpose of this study was to investigate the factor structure of the Beck Depression Inventory – Second Edition (BDI-II) in a sample of adults. Specifically, we evaluated the BDI-II based on confirmatory factor analysis of different measurement models, and compared the optimal factor structure of the BDI-II by gender using measurement invariance analysis. Method: A cross-sectional survey with 717 community-dwelling adults was conducted. The Brazilian Portuguese version of the BDI-II was administered. Seven different models (one-, two-, three-factor models and their bifactor structures) were tested through CFA. CFA and multigroup analysis were executed with the software MPLUS (Weighted Least Squares Estimator – WLSMV). Results: Four bifactor models reached acceptable fit indices. A bifactor model with two specific factors (Cognitive–Affective, and Somatic-Affective) provided the best fit to the data. The multigroup analysis of this model demonstrated invariance by gender. Conclusions: Our findings support the use of the total BDI-II score to identify depressive symptoms, including gender comparisons. Since a bifactor structure fit the data better, the scores of the specific factors should not be used as the first choice, or at least should be used with caution. The analysis of the severity of depression, based on a total score, seems to be the most appropriate option.

Reliability for Tests With Items Having Different Numbers of Ordered Categories.

This study describes a structural equation modeling (SEM) approach to reliability for tests with items having different numbers of ordered categories. A simulation study is provided to compare the performance of this reliability coefficient, coefficient alpha and population reliability for tests having items with different numbers of ordered categories, a one-factor and a bifactor structures, and different skewness distributions of test scores. Results indicated that the proposed reliability coefficient was close to the population reliability in most conditions. An empirical example was used to illustrate the performance of the different coefficients for a test of items with two or three ordered categories.

Psychometric Properties and Structures of the IAT, GPIUS and GAS Scales: A Bifactor Approach

This study applied a bifactor approach to investigate the structures and simultaneously compare the psychometric properties of three popular self-report internet addiction (IA) instruments. A bifactor confirmatory factor analysis was used to address the structures of the three scales, while the bifactor multidimensional item response model was employed to compare the psychometric properties of the three scales. Results of bifactor confirmatory factor analysis (CFA) showed that the bifactor structures were suitable for the three scales. These corresponding bifactor structures were used in the subsequent bifactor multidimensional item response theory (MIRT) analysis. Results of the bifactor MIRT showed that: three instruments of IA performed well as a whole; the Generalised Problematic Internet Use Scale (GPIUS) and Internet Addiction Test (IAT) provided more test information and had less standard error of measurement, which ranged from −3 to −1 standard deviations of theta or IA severity; the Game Addiction Scale (GAS) performed better than the other two scales in that it can provide more test information in the large area of IA severity (from −1 to +3 SDs). These suggest that the GPIUS and IAT may be the best choice for epidemiological IA studies and for measuring those with lower IA severity. Meanwhile, the GAS may be a good choice when we recruit those with various levels of IA severity.

Giving G a Meaning: An Application of the Bifactor-(S-1) Approach to Realize a More Symptom-Oriented Modeling of the Beck Depression Inventory-II.

The Beck Depression Inventory-II is one of the most frequently used scales to assess depressive burden. Despite many psychometric evaluations, its factor structure is still a topic of debate. An increasing number of articles using fully symmetrical bifactor models have been published recently. However, they all produce anomalous results, which lead to psychometric and interpretational difficulties. To avoid anomalous results, the bifactor-(S-1) approach has recently been proposed as alternative for fitting bifactor structures. The current article compares the applicability of fully symmetrical bifactor models and symptom-oriented bifactor-(S-1) and first-order confirmatory factor analysis models in a large clinical sample (N = 3,279) of adults. The results suggest that bifactor-(S-1) models are preferable when bifactor structures are of interest, since they reduce problematic results observed in fully symmetrical bifactor models and give the G factor an unambiguous meaning. Otherwise, symptom-oriented first-order confirmatory factor analysis models present a reasonable alternative.

Self-reported depression symptoms in haemodialysis patients: Bi-factor structures of two common measures and their association with clinical factors.

To validate the factor structure of two common self-report depression tools in a large sample of haemodialysis (HD) patients and to examine their demographic and clinical correlates, including urine output, history of depression and transplantation. Factor structures of the Beck Depression Inventory (BDI-II) and Patient Health Questionnaire (PHQ-9) were evaluated using confirmatory factor analysis (CFA). Data was utilised from the screening phase (n = 709) of a placebo-controlled feasibility randomised control trial (RCT) of sertraline in HD patients with mild to moderate Major Depressive Disorder. Alternative factor models including bi-factor models for the BDI-II and PHQ-9 were evaluated. Coefficient omega and omega-hierarchical were calculated. For both measures, bi-factor measurement models had the overall best fit to the data, with dominant general depression factors. Omega-hierarchical for the general BDI-II and PHQ-9 factors was 0.94 and 0.88 respectively. Both general factors had high reliability (coefficient omega = 0.97 and 0.94 respectively) and explained over 85% of the explained common variance within their respective models. BDI-II and PHQ-9 general depression factors were negatively associated with age and urine output and positively with a history of depression, antidepressant use within the last 3 months and a history of failed transplantation. In adjusted regression models, age, urine output and a history of depression remained significant. These data suggest that both the BDI-II and PHQ-9 are sufficiently unidimensional to warrant the use of a total score. Younger age, lower urine output and a history of depression appear consistent correlates of depression severity among HD patients.

Internal structure of academic self-concept through the Self-Description Questionnaire II-Short (SDQII-S)

Modelling academic self-concept through second-order factors or bifactor structures is an important issue with substantive and practical implications. To contribute to the discussion, this research analyzes the structure of academic self-concept through the Self-Description Questionnaire-II short-form (SDQII-S) and assesses its invariance across gender and age in a sample of secondary students (N = 517). Results show a better fit for the incomplete bifactor model and the estimation of several statistical indices support the multidimensional nature of the data as well as the reliability of the domain scores. In terms of invariance regarding age and gender, partial invariance is concluded. Girls showed higher latent means in verbal self-concept, while boys had higher values in mathematics self-concept; as the student’s educational level increased, domain-specific and general factor means decreased. Practical implications associated with the use of domain scores are discussed.

Direct Schmid–Leiman Transformations and Rank-Deficient Loadings Matrices

The Schmid-Leiman (S-L; Psychometrika 22: 53-61, 1957) transformation is a popular method for conducting exploratory bifactor analysis that has been used in hundreds of studies of individual differences variables. To perform a two-level S-L transformation, it is generally believed that two separate factor analyses are required: a first-level analysis in which k obliquely rotated factors are extracted from an observed-variable correlation matrix, and a second-level analysis in which a general factor is extracted from the correlations of the first-level factors. In this article, I demonstrate that the S-L loadings matrix is necessarily rank deficient. I then show how this feature of the S-L transformation can be used to obtain a direct S-L solution from an unrotated first-level factor structure. Next, I reanalyze two examples from Mansolf and Reise (Multivar Behav Res 51: 698-717, 2016) to illustrate the utility of 'best-fitting' S-L rotations when gauging the ability of hierarchical factor models to recover known bifactor structures. Finally, I show how to compute direct bifactor solutions for non-hierarchical bifactor structures. An online supplement includes R code to reproduce all of the analyses that are reported in the article.

Characterising the latent structure and organisation of self-reported thoughts, feelings and behaviours in adolescents and young adults.

Little is known about the underlying relationships between self-reported mental health items measuring both positive and negative emotional and behavioural symptoms at the population level in young people. Improved measurement of the full range of mental well-being and mental illness may aid in understanding the aetiological substrates underlying the development of both mental wellness as well as specific psychiatric diagnoses. A general population sample aged 14 to 24 years completed self-report questionnaires on anxiety, depression, psychotic-like symptoms, obsessionality and well-being. Exploratory and confirmatory factor models for categorical data and latent profile analyses were used to evaluate the structure of both mental wellness and illness items. First order, second order and bifactor structures were evaluated on 118 self-reported items obtained from 2228 participants. A bifactor solution was the best fitting latent variable model with one general latent factor termed ‘distress’ and five ‘distress independent’ specific factors defined as self-confidence, antisocial behaviour, worry, aberrant thinking, and mood. Next, six distinct subgroups were derived from a person-centred latent profile analysis of the factor scores. Finally, concurrent validity was assessed using information on hazardous behaviours (alcohol use, substance misuse, self-harm) and treatment for mental ill health: both discriminated between the latent traits and latent profile subgroups. The findings suggest a complex, multidimensional mental health structure in the youth population rather than the previously assumed first or second order factor structure. Additionally, the analysis revealed a low hazardous behaviour/low mental illness risk subgroup not previously described. Population sub-groups show greater validity over single variable factors in revealing mental illness risks. In conclusion, our findings indicate that the structure of self reported mental health is multidimensional in nature and uniquely finds improved prediction to mental illness risk within person-centred subgroups derived from the multidimensional latent traits.