Bifactor Measurement Model Research Articles

Measuring Cultural Competencies: The Development and Validation of a Revised Measure of Sociocultural Adaptation

This research extends our work on the conceptualization and measurement of cultural competencies with a revised bifactor measurement model of sociocultural adaptation (the Revised Sociocultural Ada...

Comparing the predictive validity of the four-factor and five-factor (bifactor) measurement structures of the drinking motives questionnaire

IntroductionThe Drinking Motives Questionnaire (DMQ-R) is the most widely administered instrument to assess reasons for consuming alcohol and is conventionally premised on a four-factor structure. Recent research instead reveals that a bifactor measurement model of five motive factors (one general and four specific) represents a superior psychometric embodiment of the scale. The current study evaluated and compared the predictive validity of the four-factor and five-factor models of drinking motives in longitudinally explaining alcohol use and problems. MethodsAdult participants (N=413; age range=18–79 years) completed measures of drinking motives (Time 1) and alcohol use and problems one month later (Time 2). ResultsConfirmatory factor analyses corroborated the four-factor (social, enhancement, conformity, and coping motives) and five-factor (each item double loading on general motives and a specific motives factor) measurement structures, but the latter rendered stronger fit indices. Structural equation models revealed that lower social motives, higher enhancement motives, and higher coping motives prospectively contributed to alcohol use. Furthermore, lower social motives, higher conformity motives, higher coping motives, and greater alcohol use contributed to alcohol problems. DiscussionThe same set of paths emerged as significantly predictive in both models, but general motives additionally explained alcohol use and problems in the five-factor model. The incremental contribution of general motives (beyond the specific motives) on alcohol intake and detrimental consequences supports the predictive validity of the drinking reasons paradigm embodied by the inclusion of a global factor.

Ethnic differences in mothering qualities and relations to academic achievement.

Although qualities of mothering behavior have been consistently linked with children's academic outcomes, mothers from different ethnic groups may emphasize different dimensions with their children. The present investigation aims to evaluate and compare the dimensionality of mothering in low-income African American (n = 151) and Mexican American (n = 182) mothers during early childhood and its predictive utility for children's academic achievement. Video-recorded mother-child interactions with children at 2½ and 3½ years of age were rated using 6 mothering quality items from a widely used global rating system. A bifactor measurement model of these 6 items yielded a general sensitive support factor and a specific intrusive-insensitive factor. The bifactor model fit the data significantly better at both time points than either a single-factor or a 2-factor model. Invariance testing supported the stability of the measurement model across the 2 time points. Invariance testing by ethnicity indicated differences in factor loadings as well as mean levels of the specific factor of intrusive-insensitivity. The specific factor reflecting intrusive-insensitive mothering at age 2½ years was associated with poorer subsequent reading achievement for African American but not Mexican American children, suggesting the specific factor reflected qualitatively different parenting constructs for the 2 ethnic groups. Critical examination of what constitutes more optimal parenting yielded both similar and dissimilar characteristics and their relations across culturally different groups of families. Such knowledge should contribute to the development of more effective interventions for ethnically diverse families. (PsycINFO Database Record

A Bifactor Approach to Model Multifaceted Constructs in Statistical Mediation Analysis.

Statistical mediation analysis allows researchers to identify the most important mediating constructs in the causal process studied. Identifying specific mediators is especially relevant when the hypothesized mediating construct consists of multiple related facets. The general definition of the construct and its facets might relate differently to an outcome. However, current methods do not allow researchers to study the relationships between general and specific aspects of a construct to an outcome simultaneously. This study proposes a bifactor measurement model for the mediating construct as a way to parse variance and represent the general aspect and specific facets of a construct simultaneously. Monte Carlo simulation results are presented to help determine the properties of mediated effect estimation when the mediator has a bifactor structure and a specific facet of a construct is the true mediator. This study also investigates the conditions when researchers can detect the mediated effect when the multidimensionality of the mediator is ignored and treated as unidimensional. Simulation results indicated that the mediation model with a bifactor mediator measurement model had unbiased and adequate power to detect the mediated effect with a sample size greater than 500 and medium a- and b-paths. Also, results indicate that parameter bias and detection of the mediated effect in both the data-generating model and the misspecified model varies as a function of the amount of facet variance represented in the mediation model. This study contributes to the largely unexplored area of measurement issues in statistical mediation analysis.

Evaluating bifactor models: Calculating and interpreting statistical indices.

Bifactor measurement models are increasingly being applied to personality and psychopathology measures (Reise, 2012). In this work, authors generally have emphasized model fit, and their typical conclusion is that a bifactor model provides a superior fit relative to alternative subordinate models. Often unexplored, however, are important statistical indices that can substantially improve the psychometric analysis of a measure. We provide a review of the particularly valuable statistical indices one can derive from bifactor models. They include omega reliability coefficients, factor determinacy, construct reliability, explained common variance, and percentage of uncontaminated correlations. We describe how these indices can be calculated and used to inform: (a) the quality of unit-weighted total and subscale score composites, as well as factor score estimates, and (b) the specification and quality of a measurement model in structural equation modeling. (PsycINFO Database Record

Do the Autism Spectrum Quotient (AQ) and Autism Spectrum Quotient Short Form (AQ-S) Primarily Reflect General ASD Traits or Specific ASD Traits? A Bi-Factor Analysis.

In the current study, we fit confirmatory bi-factor models to the items of the Autism Spectrum Quotient (AQ) and Autism Spectrum Quotient Short Form (AQ-S) in order to assess the extents to which the items of each reflect general versus specific factors. The models were fit in a combined sample of individuals with and without a clinical diagnosis of autism spectrum disorders. Results indicated that, with the exception of the Attention to Details factor in the AQ and the Numbers/Patterns factors in the AQ-S, items primarily reflected a general factor. This suggests that when attempting to estimate an association between a specific symptom measured by the AQ or AQ-S and some criterion, associations will be confounded by the general factor. To resolve this, we recommend using a bi-factor measurement model or factor scores from a bi-factor measurement whenever hypotheses about specific symptoms are being assessed.

Coming Full Circle With Reactions: Understanding the Structure and Correlates of Trainee Reactions Through the Affect Circumplex

Research suggests that the structure of trainee reactions is captured with as few as one or as many as 11 dimensions. It is commonly understood that reactions contain both affective and cognitive components. To date, however, training research focuses largely on affective reactions that range from pleasant to unpleasant (i.e., valence). Here, we expand and further refine the construct of affective trainee reactions by including reactions that are more and less activating versions of pleasantness (e.g., excitement and calm, respectively) and unpleasantness (e.g., stress and boredom, respectively). We develop and validate a new measure based on this model and argue that the structure of affective reactions has implications for better understanding learning and course reputation outcomes. Results from a short online training indicate that reactions were best explained by four factors: pleasant activation (e.g., excitement), pleasant deactivation (e.g., calm), unpleasant activation (e.g., tension), and unpleasant deactivation (e.g., boredom). In addition to these four factors, results also show a general activation factor underlying all items, which produces a bifactor measurement model. The relationships between these reactions and training outcomes suggest what is most beneficial for course reputation outcomes (i.e., pleasant activating reactions) may not benefit learning; what is most beneficial for learning (i.e., pleasant deactivating reactions) may benefit course reputation outcomes but slightly less so.

The Aphasia Communication Outcome Measure (ACOM): Dimensionality, Item Bank Calibration, and Initial Validation.

The purpose of this study is to investigate the structure and measurement properties of the Aphasia Communication Outcome Measure (ACOM), a patient-reported outcome measure of communicative functioning for persons with aphasia. Three hundred twenty-nine participants with aphasia responded to 177 items asking about communicative functioning. The data were analyzed using a categorical item factor analysis approach. Validity of ACOM scores on the basis of their convergence with performance-based, clinician-reported, and surrogate-reported assessments of communication was also assessed. Fifty-nine items that obtained adequate fit to a modified bifactor measurement model and functioned similarly across several demographic and clinical subgroupings were identified. The factor model estimates were transformed to item response theory graded response model parameters, and the resulting score estimates showed good precision and moderately strong convergence with other measures of communicative ability and functioning. A free software application for administration and scoring of the ACOM item bank is available from the first author. The ACOM provides reliable measurement of patient-reported communicative functioning in aphasia. The results supported the validity of ACOM scores insofar as (a) factor analyses provided support for a coherent measurement model, (b) items functioned similarly across demographic and clinical subgroups, and (c) scores showed good convergence with measures of related constructs.

Disentangling Self-Control from Its Elements: A Bifactor Analysis

Disentangle self-control from its elements and provide several new insights into the self-control dimensionality debate including: the proportion explained variance in scale items attributed to self-control and its elements, the viability of using total and individual scores to measure self-control and its elements in observed variable analyses, and the unique effects of general (self-control) and specific (elements) latent factors on crime and victimization. The current study utilizes bifactor measurement and structural equation models to address the research objectives. The sample consists of Florida jail inmates and self-control and its elements are measured with the Grasmick et al. scale. Results indicate the elements exist above and beyond the general factor of self-control, and that these specific factors collectively account for nearly one-third of the total proportion explained variance in the scale items. Findings from omega reliability analyses provide evidence supporting the use of a total score to measure self-control, but discouraging the use of subscales to measure the individual elements, when measurement error is not taken into account. Results from a bifactor structural equation model predicting crime and victimization reveal that the effects of three latent specific factors (temper, risk-seeking, and self-centeredness) are substantially larger than the effects of the general factor (self-control). Bifactor methods placed self-control and the elements on equal conceptual footing and found both to explain variation in Grasmick et al. item responses and both to influence crime and victimization. Future work should examine the origins and stability of self-control vis-a-vis the individual elements.

The Rediscovery of Bifactor Measurement Models

Bifactor latent structures were introduced over 70 years ago, but only recently has bifactor modeling been rediscovered as an effective approach to modeling construct-relevant multidimensionality in a set of ordered categorical item responses. I begin by describing the Schmid-Leiman bifactor procedure (Schmid & Leiman, 1957) and highlight its relations with correlated-factors and second-order exploratory factor models. After describing limitations of the Schmid-Leiman, 2 newer methods of exploratory bifactor modeling are considered, namely, analytic bifactor (Jennrich & Bentler, 2011) and target bifactor rotations (Reise, Moore, & Maydeu-Olivares, 2011). Then I discuss limited- and full-information estimation approaches to confirmatory bifactor models that have emerged from the item response theory and factor analysis traditions, respectively. Comparison of the confirmatory bifactor model to alternative nested confirmatory models and establishing parameter invariance for the general factor also are discussed. Finally, important applications of bifactor models are reviewed. These applications demonstrate that bifactor modeling potentially provides a solid foundation for conceptualizing psychological constructs, constructing measures, and evaluating a measure's psychometric properties. However, some applications of the bifactor model may be limited due to its restrictive assumptions.

Disordered brain development and abnormal connectivity

Self-Concept Focus is a 15-item measure of the disposition to make autobiographical memories central to one's self-concept and thus to rehearse them more frequently. In studies with 400 MTurk workers and 299 undergraduates, Self-Concept Focus had high reliability (α ∼.83), good test-retest stability (r = .66) that did not decline between 7 and 54 days, good psychometric properties in a bifactor measurement model, and results that replicated across studies. A factor analysis of nine measures relevant to self-concept resulted in two factors. Self-Concept Focus, Reflection, Reappraisal and Private Self-Consciousness all loaded on one factor, suggesting an underlying dimension of elaborative rehearsal of memories and emotion regulation. Self-Concept Focus also correlated with PTSD symptoms for a single very negative event and thus with opportunities to modify such memory-based symptoms. Given its association with elaborative rehearsal and emotion regulation, Self-Concept Focus has potential applied relevance in clinical, forensic and consumer contexts.