Unidimensional Model Research Articles

How well do DSM-5 criteria measure alcohol use disorder in the general population of older Swedish adolescents? An item response theory analysis

BackgroundThis study assesses the psychometric properties of DSM-5 criteria of AUD in older Swedish adolescents using item response theory models, focusing specifically on the precision of the scale at the cut-offs for mild, moderate, and severe AUD. MethodsData from the second wave of Futura01 was used. Futura01 is a nationally representative cohort study of Swedish people born 2001 and data for the second wave was collected when participants were 17/18 years old. This study included only participants who had consumed alcohol during the past 12 months (n = 2648). AUD was measured with 11 binary items. A 2-parameter logistic item response theory model (2PL) estimated the items’ difficulty and discrimination parameters. Results31.8% of the participants met criteria for AUD. Among these, 75.6% had mild AUD, 18.3% had moderate, and 6.1% had severe AUD. A unidimensional AUD model had a good fit and 2PL models showed that the scale measured AUD over all three cut-offs for AUD severity. Although discrimination parameters ranged from moderate (1.24) to very high (2.38), the more commonly endorsed items discriminated less well than the more difficult items, as also reflected in less precision of the estimates at lower levels of AUD severity. The diagnostic uncertainty was pronounced at the cut-off for mild AUD. ConclusionDSM-5 criteria measure AUD with better precision at higher levels of AUD severity than at lower levels. As most older adolescents who fulfil an AUD diagnosis are in the mild category, notable uncertainties are involved when an AUD diagnosis is set in this group.

Assessing nature exposure: a study on the reliability and validity of a Portuguese version of the nature exposure scale.

Engaging with nature has been widely acknowledged for its positive impact on well-being. Traditionally, assessments of nature exposure focus on estimating green space exposure and outdoor time. However, the Nature Exposure Scale (NES) offers a unique approach by evaluating the quality of nature experiences, encompassing both deliberate and spontaneous encounters. This study aimed to assess the validity and reliability of the NES in the context of Portugal. Exploratory Factor Analysis (PCA) and Confirmatory Factor Analysis (CFA) were employed to examine the underlying structure of the scale. Additionally, reliability assessments, along with tests of convergent and divergent validity, were conducted. Data were collected from a sample of 558 adults and 241 adolescents between 2016 and 2018. The findings revealed strong internal consistency of the NES, supported by acceptable correlation values and robust factor loadings within a unidimensional model. The scale effectively predicted variations in nature exposure across diverse professional activities. In summary, the 4-item NES emerged as a reliable tool for assessing nature exposure in various settings, effectively bridging a gap in the Portuguese context. The scale demonstrated potential for cross-cultural research and was particularly adept at predicting nature exposure in different professional contexts. The results suggest that the NES can enhance our understanding of the impact of nature on well-being in diverse cultural settings. The study underscores the reliability of the NES in assessing nature exposure in Portugal, paving the way for further exploration in Portuguese-speaking regions. The scale holds promise for advancing research on the relationship between nature and well-being, contributing valuable insights across different cultural contexts.

Analysis of Mixed-Format Assessments Using Measurement Models and Topic Modeling

ABSTRACT It is common to find mixed-format data results from the use of both multiple-choice (MC) and constructed-response (CR) questions on assessments. Dealing with these mixed response types involves understanding what the assessment is measuring, and the use of suitable measurement models to estimate latent abilities. Past research in educational measurement, however, has often overlooked the written responses in CR items after analyzing the response scores. This study presents a method for bridging this gap by using a topic model called latent Dirichlet allocation to uncover the structure in written answers and then use that information to augment results from traditional measurement models. In this study, a five-step framework is employed for assessing both the examinees’ latent abilities and the internal structure of the written responses obtained from mixed-format assessments. An empirical dataset obtained from Grade 8 examinees in a southeastern state on an English language arts assessment is used for illustration. Based on the dimensionality of the assessment, a unidimensional partial credit model and two multidimensional bi-factor models were fit to the data. A comparison of results from these analyses suggests that a constrained bi-factor model was most useful for detecting the mixed-format score patterns, and that a 5-topic model was determined for the textual responses. The topic distributions were found to be related to the latent abilities estimated from the constrained bi-factor model. This framework combines both the score patterns and textual responses, and highlights the utility of combining the traditional response analysis with a start-of-art language model.

Examining Differential Item Functioning from a Multidimensional IRT Perspective.

Differential item functioning (DIF) is a standard analysis for every testing company. Research has demonstrated that DIF can result when test items measure different ability composites, and the groups being examined for DIF exhibit distinct underlying ability distributions on those composite abilities. In this article, we examine DIF from a two-dimensional multidimensional item response theory (MIRT) perspective. We begin by delving into the compensatory MIRT model, illustrating and how items and the composites they measure can be graphically represented. Additionally, we discuss how estimated item parameters can vary based on the underlying latent ability distributions of the examinees. Analytical research highlighting the consequences of ignoring dimensionally and applying unidimensional IRT models, where the two-dimensional latent space is mapped onto a unidimensional, is reviewed. Next, we investigate three different approaches to understanding DIF from a MIRT standpoint: 1. Analytically Uniform and Nonuniform DIF: When two groups of interest have different two-dimensional ability distributions, a unidimensional model is estimated. 2. Accounting for complete latent ability space: We emphasize the importance of considering the entire latent ability space when using DIF conditional approaches, which leads to the mitigation of DIF effects. 3. Scenario-Based DIF: Even when underlying two-dimensional distributions are identical for two groups, differing problem-solving approaches can still lead to DIF. Modern software programs facilitate routine DIF procedures for comparing response data from two identified groups of interest. The real challenge is to identify why DIF could occur with flagged items. Thus, as a closing challenge, we present four items (Appendix A) from a standardized test and invite readers to identify which group was favored by a DIF analysis.

The appearance of multidimensionality: Exploring the case for unidimensionality of the PTSD Checklist for DSM-5 using a firefighter sample.

Rasmussen et al. (2019) described the proliferation of factors for posttraumatic stress disorder (PTSD) measures and raised concerns about the construct validity of factors that include two or three items. In this brief report, we describe how the pattern of covariation among the responses to items of well-established measures, such as the PTSD Checklist for DSM-5 (PCL-5), can give the appearance of multidimensionality. We evaluated whether the structure of the 20-item PCL-5 is unidimensional, using the methods of multidimensional item response theory (MIRT) and the concept of a testlet. These analyses were done using a sample of trauma-exposed urban firefighters. A unidimensional and a bifactor model, which includes a general factor composed of all items and four specific factors mirroring the DSM-5 conceptualization, were evaluated for both Likert-type multiple-category and binary coding system of the PCL-5 item response data. Seven testlets were created from the 20 PCL-5 items following the seven-factor model (Armour et al., 2015) presented in Table 1 of Rasmussen et al. (2019). Findings using the unidimensional nominal item response theory model for the seven testlets indicated that the PCL-5 may be considered unidimensional with a single score representing individual differences on a continuum that ranges from low to high. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Allostatic Load in Childhood, Adolescence, and Young Adulthood: Are Assumptions of Measurement Invariance Warranted?

Allostatic load represents the cumulative toll of chronic mobilization of the body's stress response systems, as indexed by biomarkers. Higher levels of stress and disadvantage predict higher levels of allostatic load, which, in turn, predict poorer physical and mental health outcomes. To maximize the efficacy of prevention efforts, screening for stress- and disadvantage-associated health conditions must occur before middle age-that is, during childhood, adolescence, and young adulthood. However, this requires that models of allostatic load display properties of measurement invariance across age groups. Because most research on allostatic load has featured older adults, it is unclear if these requirements can be met. To address this question, we fit a series of exploratory and confirmatory analytic models to data on eight biomarkers using a nationally representative sample of N = 4260 children, adolescents, and young adults drawn from the National Health and Nutrition Examination Survey dataset. Exploratory and confirmatory models indicated that, consistent with allostatic load theory, a unidimensional model was a good fit to the data. However, this model did not display properties of measurement invariance; post-hoc analyses suggested that the biomarkers included in the final confirmatory model were most strongly intercorrelated among young adults and most weakly intercorrelated among adolescents. These results underscore the importance of testing assumptions about measurement invariance in allostatic load before drawing substantive conclusions about stress, disadvantage, and health by directly comparing levels of allostatic load across different stages of development, while underscoring the need to expand investigations of measurement invariance to samples of longitudinal data.

Factorial validity and gender invariance of the Patient Health Questionnaire (PHQ-9) in student-athletes

Background/purpose Depressive symptoms are prevalent in student-athletes. Evidence for the factorial validity of measures assessing depressive symptoms in student-athletes is presently absent from the literature. This study examined the best fitting factorial structure and invariance across sexes of the PHQ-9. Methods Data were collected from 1,524 student-athletes from institutions in the North-Eastern United States. Confirmatory factor analyses of 4 factorial structures of the Patient Health Questionnaire-9 (PHQ-9) were conducted, followed by invariance analyses across sexes. Results All tested models showed an excellent fit for all tested models and a unidimensional model of the PHQ-9 was chosen for further analyses. Metric and scalar invariance for the identified model across sexes was supported. The participants in this study reported an unusually low (3.13%) prevalence of elevated depressive symptoms, which may indicate underreporting of symptoms. Conclusion Using a unidimensional, composite score via the PHQ-9 is a valid assessment of depressive symptoms in male and female student-athletes.

Interface fluctuations associated with split Fermi seas

We consider the asymptotic behaviour of a family of unidimensional lattice fermion models, which are in exact correspondence with certain probability laws on partitions and on unitary matrices. These models exhibit limit shapes, and in the case where the bulk of these shapes are described by analytic functions, the fluctuations around their interfaces have been shown to follow a universal Tracy–Widom distribution or its higher-order analogue. Non-differentiable bulk limit shape functions arise when a gap appears in some quantum numbers of the model, in other words when the Fermi sea is split. We show that split Fermi seas give rise to new interface fluctuations, governed by integer powers of universal distributions. This breakdown in universality is analogous to the behaviour of a random Hermitian matrix when the support of its limiting eigenvalue distribution has multiple cuts, with oscillations appearing in the limit of the two-point correlation function. We show that when the Fermi sea is split in the lattice fermion model, there are multiple cuts in the eigenvalue support of the corresponding unitary matrix model.

Exploring and Evaluating the Two-Factor Model of Perfectionism in Sport

Perfectionism is a multidimensional personality trait with two higher-order dimensions; perfectionistic strivings and perfectionistic concerns. The purpose of the present study was to explore and evaluate the two-factor model for the first time using three instruments developed to measure perfectionism in sport. In doing so, we (i) assessed the fit of two-factor models when including and excluding various contentious subscales (other-oriented perfectionism, parental pressure, coach pressure, organisation, and negative reactions to imperfection) and (ii) compared two-factor models to alternative one-factor (or unidimensional) models. Participants were recruited from community and university sports clubs in the UK ( N = 527; M age = 18.07 years, SD = 0.49) and completed the Sport-Multidimensional Perfectionism Scale-2, the Multidimensional Inventory of Perfectionism in Sport, and the Performance Perfectionism Scale-Sport. Support was found for the two-factor model, with superior fit displayed each time the aforementioned subscales were excluded and, in all cases, when compared to a unidimensional model. The findings suggest that the two-factor model is an adequate representation of the underlying structure of instruments designed to measure perfectionism in sport with better fit and conceptual clarity offered by more parsimonious models.

Ant colony optimization for parallel test assembly.

Ant colony optimization (ACO) algorithms have previously been used to compile single short scales of psychological constructs. In the present article, we showcase the versatility of the ACO to construct multiple parallel short scales that adhere to several competing and interacting criteria simultaneously. Based on an initial pool of 120 knowledge items, we assembled three 12-item tests that (a) adequately cover the construct at the domain level, (b) follow a unidimensional measurement model, (c) allow reliable and (d) precise measurement of factual knowledge, and (e) are gender-fair. Moreover, we aligned the test characteristic and test information functions of the three tests to establish the equivalence of the tests. We cross-validated the assembled short scales and investigated their association with the full scale and covariates that were not included in the optimization procedure. Finally, we discuss potential extensions to metaheuristic test assembly and the equivalence of parallel knowledge tests in general.

Psychometric properties of an Arabic translation of the short entrapment scale in a non-clinical sample of young adults

BackgroundEntrapment is the feeling of wanting to leave an unbearable situation but believing that there are no options to do so. An Arabic entrapment Scale will assist healthcare professionals in the region in the prevention of suicide as the tool is tailored to the specific sociocultural context, which would enhance entrapment detection.ObjectiveIn the current study, we aim to evaluate the psychometric properties of a translated Arabic version of the Entrapment Scale Short Form (E-SF).MethodsThree hundred eighty-nine Lebanese citizens were enrolled in this cross-sectional study.ResultsThe mean age of participants was 23.03 years (SD = 2.93), 69.4% being women. To examine the factor structure of the entrapment scale, we used an exploratory-to-confirmatory factor analysis (EFA-to-CFA) strategy. EFA and CFA results indicated that the fit of the unidimensional model of the Arabic Entrapment Scale (A-ES) was generally acceptable. Composite reliability of scores was adequate in the total sample (ω = .87). All indices suggested that configural, metric, and scalar invariance was supported across genders. Entrapment was positively and significantly correlated with suicidal ideation, alcohol use disorder, psychological distress, and orthorexia nervosa, suggesting convergent and divergent validity.ConclusionThe A-ES was found to be a valid and reliable tool to assess the degree of entrapment in Lebanese young adults. The A-ES will assist healthcare professionals in the region in the prevention of suicide as the tool is tailored to the specific sociocultural context, which would enhance entrapment detection.

Factor structure of psychosis screening questionnaire in Ugandan adults

BackgroundPsychotic disorders are common and contribute significantly to morbidity and mortality of people with psychiatric diseases. Therefore, early screening and detection may facilitate early intervention and reduce adverse outcomes. Screening tools that lay persons can administer are particularly beneficial in low resource settings. However, there is limited research evaluating the validity of psychosis screening instruments in Uganda. We aimed to assess the construct validity and psychometric properties of the Psychosis Screening Questionnaire (PSQ) in Uganda in a population with no history of a psychotic disorder.MethodsThe sample consisted of 2101 Ugandan adults participating as controls in a larger multi-country case-control study on psychiatric genetics who were recruited between February 2018 and March 2020. Participants were individuals seeking outpatient general medical care, caretakers of individuals seeking care, and staff or students recruited from five medical facilities that were age 18 years or older and able to provide consent. Individuals were excluded who had acute levels of alcohol or substance use, including being under inpatient hospitalization or acute medical care for one of these conditions. We used confirmatory factor analysis (CFA) and item response theory (IRT) to evaluate the factor structure and item properties of the PSQ.ResultsThe overall prevalence screening positive for psychotic symptoms was 13.9% 95% CI (12.4,15.4). “Strange experiences” were the most endorsed symptoms 6.6% 95% CI (5.6,7.8). A unidimensional model seemed to be a good model or well-fitting based on fit indices including the root mean square error of approximation (RMSEA of 0.00), comparative fit index (CFI of 1.000), and Tucker-Lewis Index (TLI of 1.000). The most discriminating items along the latent construct of psychosis were items assessing thought disturbance followed by items assessing paranoia, with a parameter (discrimination) value of 2.53 and 2.40, respectively.ConclusionThe PSQ works well in Uganda as an initial screening tool for moderate to high-level of psychotic symptoms.

Analysis of the dimensional structure of the Patient Health Questionnaire-9 (PHQ-9) in undergraduate students at a public university in Brazil

BackgroundThe prevalence of mental health-related problems has increased in recent years, especially among young adults, such as university students. Several measurement instruments have been developed to identify individuals at risk for depression, such as the Patient Health Questionnaire-9 (PHQ-9). However, different dimensional structures of the PHQ-9 can make it difficult to interpret and compare research results. To analyze the dimensional structure and convergent validity of the PHQ-9 in university students. MethodsThis is a cross-sectional study developed with 3163 students enrolled in different courses at a university in southern Brazil. Exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) were performed to identify the most parsimonious and best-fitting model. Convergent validity was verified through the significant relationship (p < 0.05) between theoretically related constructs (sleep quality, alcohol consumption, and physical activity practice). ResultsThe EFA resulted in two models (unidimensional and two-dimensional), with better indices for the two-dimensional model. In the CFA, both the unidimensional and the two-dimensional models presented satisfactory fit indices that were higher for the unidimensional model. LimitationsThe analysis of convergent validity is limited by the absence of a gold standard for comparison. ConclusionsThis study provided support for the unidimensional structure of the PHQ-9, with adequate convergent validity, among university students.

Psychometric Properties of the Connor-Davidson Resilience Scale (CD-RISC) in Spanish Adolescents.

Few studies have examined the psychometric properties of the Connor-Davidson Resilience Scale (CD-RISC) in a large adolescent community sample, finding a significant disparity. This study explores the psychometric properties of the CD-RISC among Spanish adolescents by means of exploratory factor analysis (EFA), Rasch analysis, and measurement invariance (MI) across sex, as well as internal consistency and criterion validity. The sample was comprised of 463 adolescents (231 girls), aged 12 to 18 years, who completed the CD-RISC and other measures on emotional status and quality of life. The EFA suggested that the CD-RISC structure presented a unidimensional model. Consequently, shorter unidimensional CD-RISC models observed in the literature were explored. Thus, the Campbell-Sills and Stein CD-RISC-10 showed the soundest psychometric properties, providing an adequate item fit and supporting MI and non-differential item functioning across sex. Item difficulty levels were biased toward low levels of resilience. Some items showed malfunctioning in lower response categories. With regard to reliability, categorical omega was. 82. Strong associations with health-related quality of life, major depressive disorder symptoms, and emotional symptoms were observed. A weak association was found between resilience and the male sex. Campbell-Sills and Stein's CD-RISC-10 model emerges as the best to assess resilience among Spanish adolescents, as already reported in adults. Thus, independently of the developmental stage, the core of resilience may reside in the aspects of hardiness and persistence.

Applying Rasch Methodology to Examine and Enhance Precision of the Baby Care Questionnaire

The Baby Care Questionnaire (BCQ) is an established ordinal measure of parenting beliefs about infant care, which includes structure and attunement scales. Rasch analysis is a powerful method to examine and improve psychometric properties of ordinal scales. This study aimed to evaluate the psychometric properties and improve precision of the structure and attunement scales of the BCQ using Rasch methodology. A Partial Credit Rasch model was applied to a sample of 450 mothers from the United Kingdom (n = 225) and New Zealand (n = 225) who completed the BCQ. Initial analyses indicated acceptable reliability of the structure and attunement scales of the BCQ, but some items showed misfit to the Rasch model associated with local dependency issues in both scales. After combining locally dependent items into testlets, both scales of the BCQ met expectations of the unidimensional Rasch model and demonstrated adequate and strong reliability and invariance across countries and person factors such as participants’ age and their baby’s sex. This permitted the generation of conversion algorithms to transform ordinal scores into interval data to enhance the precision of both scales of the BCQ. In conclusion, using Rasch methodology, this study demonstrated robust psychometric properties of the structure and attunement scales of the BCQ after minor modifications. The ordinal-to-interval conversion tables published here can be used to further enhance the precision of the structure and attunement scales of the BCQ without changing their original response format. These findings contribute to the enhancement of precision in measuring parenting beliefs about infant care.

Psychometric Evaluation of the Trauma-Informed Care Provider Assessment Tool.

The increasing recognition of adverse childhood experiences as a significant factor in adult health outcomes underscores the need for trauma-informed care (TIC) in healthcare settings. The purpose of this study was to assess the psychometric properties of the TIC Provider Assessment Tool (TIC-PAT) designed for primary care providers. The TIC-PAT aligns with the TIC Pyramid framework and assesses both universal trauma precautions and trauma-specific care. A total of 176 primary care providers in the United States completed the TIC-PAT through an anonymous online survey. Findings through exploratory and confirmatory factor analyses revealed a unidimensional (one-factor) model, consolidating questions into a concise 10-item measure. This study contributes an efficient assessment tool for the provision of TIC by primary care providers in healthcare settings, promoting better patient-provider interactions and enhancing provider awareness of trauma's impact on health.

Psychometric properties of the Sinhala perceived stress questionnaire (PSQ8-11) in Sri Lankan primary school children.

Stress influences examination performance among Sri Lankan students. Validated tests are required to evaluate stress levels among elementary students in Sri Lanka. Therefore, the Perceived Stress Questionnaire 8-11 (PSQ8-11) was translated into a Sinhala version. The aim of this study was to examine the psychometric properties of the translated and adapted scale among elementary level school children in Sri Lanka and examine invariance across male and female children. The participants were 1021 students from seven schools. After removing missing values, responses from 693 students (mean age = 9.65 ± 0.478 years, 51.8% male) were analysed for participant characteristics. Cronbach's alpha, Spearman's correlation, and confirmatory factor analysis with measurement invariance models were conducted after adding one item to the original PSQ8-11 version. The Cronbach's alpha value for the 20-item modified PSQ8-11 Sinhala version was.788. The two subscales, psychological stress (Cronbach's alpha = 0.615) and physiological stress (Cronbach's alpha = 0.711), indicated a satisfactory level of internal consistency. Furthermore, a statistically significant correlation (p < 0.01; 2-tailed) was reported among each of the subscales. Confirmatory factor analyses demonstrated a satisfactory goodness-of-fit across the two models by confirming the theoretical constructs of the PSQ8-11 translated version with its two subscales. The two-factor model has better model fit indices compared to the unidimensional model (χ2/df = 1.447, CFI = 0.947, TLI = 0.938, WRMR = 0.028, RMSEA = 0.026, SRMSR = 0.0341, and PCLOSE = 1 of the two-factor model). Measurement variance across gender was supported by the establishment of configural and metric invariances. Acceptable psychometric properties for the PSQ8-11 Sinhala version were observed in elementary schoolers in Sri Lanka.

Validation of the French version of the Body Awareness Questionnaire: toward a way to assess alexisomia.

Awareness of one's own states is a particularly important part of cognition and emotion regulation. Recently, the concept of alexisomia has been used to refer to lack of awareness and expression of somatic sensations. Developing self-reported questionnaires to evaluate alexisomia represents a challenge for clinical psychology and medicine. In this context, we suggested to adapt the Body Awareness Questionnaire in French to measure alexisomia and its relation to alexithymia. In fact, we carried out a backtranslation and studied the validity of the construct in relation to proximal constructs around emotional awareness. For this study, 610 university students completed questionnaires measuring a three dimensions alexithymia concept [with The Toronto Alexithymia Scale (TAS-20)] or the five dimensions alexithymia concept [with The Bermond-Vorts Alexithymia Questionnaire (BVAQ-B)] and alexisomia (with the BAQ). Confirmatory factor analyses showed that the BAQ can be envisaged through 4 factors as well as a unidimensional model to refer to alexisomia. We also found that body awareness was negatively related to scores of alexithymia. Results are discussed in light of the construct of alexisomia and its clinical implications in somatic as well as mental disorders. We suggest that the BAQ, which assesses interoception, can contribute, in part, to the assessment of alexisomia. Like alexithymia, this is a key concept to take into consideration when designing treatment and prevention programs.

Адаптация русскоязычной версии шкалы генерализованного тревожного расстройства (Generalized Anxiety Disorder-7)

&lt;p&gt;&lt;strong&gt;Aim.&lt;/strong&gt; The Generalized Anxiety Disorder&amp;ndash;7 is the &amp;laquo;gold standard&amp;raquo; for assessing anxiety in clinical and population studies. The aim of this study was to adapt the Russian version of the GAD&amp;ndash;7. &lt;strong&gt;Method.&lt;/strong&gt; In this cross&amp;ndash;sectional study, 1,130 respondents participated. They were recruited through accidental sampling. The participants completed the Russian versions of the GAD&amp;ndash;7 and the Patient Health Questionnaire&amp;ndash;9 (PHQ&amp;ndash;9) measuring the severity of depressive symptomatology. &lt;strong&gt;Results.&lt;/strong&gt; The factor structure of the Russian version of the GAD&amp;ndash;7 corresponds to both a unidimensional model that measures the general anxiety score and a two&amp;ndash;dimensional model that highlights the cognitive&amp;ndash;emotional and somatic components of anxiety. The adapted scale has high internal reliability (&amp;alpha; = 0,93) and convergent validity, confirmed through statistically significant positive correlations between the GAD&amp;ndash;7 and the PHQ&amp;ndash;9 scores. &lt;strong&gt;Conclusion.&lt;/strong&gt; The Russian version of the GAD&amp;ndash;7 is valid and reliable and can be recommended for assessing anxiety in the general population. Avenues for further psychometric examining lie in expanding validation procedures and involving clinical samples to prove the specificity and sensitivity of the GAD&amp;ndash;7, and to determine its diagnostic capabilities in clinical research and practice.&lt;/p&gt;

Detecting non-content-based response styles in survey data: An application of mixture factor analysis

It is common for some participants in self-report surveys to be careless, inattentive, or lacking in effort. Data quality can be severely compromised by responses that are not based on item content (non-content-based [nCB] responses), leading to strong biases in the results of data analysis and misinterpretation of individual scores. In this study, we propose a specification of factor mixture analysis (FMA) to detect nCB responses. We investigated the usefulness and effectiveness of the FMA model in detecting nCB responses using both simulated data (Study 1) and real data (Study 2). In the first study, FMA showed reasonably robust sensitivity (.60 to .86) and excellent specificity (.96 to .99) on mixed-worded scales, suggesting that FMA had superior properties as a screening tool under different sample conditions. However, FMA performance was poor on scales composed of only positive items because of the difficulty in distinguishing acquiescent patterns from valid responses representing high levels of the trait. In Study 2 (real data), FMA detected a minority of cases (6.5%) with highly anomalous response patterns. Removing these cases resulted in a large increase in the fit of the unidimensional model and a substantial reduction in spurious multidimensionality.