Testing the Performance of Level-Specific Fit Evaluation in MCFA Models With Different Factor Structures Across Levels

A Monte Carlo study was conducted to compare the performance of a level-specific (LS) fit evaluation with that of a simultaneous (SI) fit evaluation in multilevel confirmatory factor analysis (MCFA) models. We extended previous studies by examining their performance under MCFA models with different factor structures across levels. In addition, various design factors and interaction effects between intraclass correlation (ICC) and misspecification type (MT) on their performance were considered. The simulation results demonstrate that the LS outperformed the SI in detecting model misspecification at the between-group level even in the MCFA model with different factor structures across levels. Especially, the performance of LS fit indices depended on the ICC, group size (GS), or MT. More specifically, the results are as follows. First, the performance of root mean square error of approximation (RMSEA) was more promising in detecting misspecified between-level models as GS or ICC increased. Second, the effect of ICC on the performance of comparative fit index (CFI) or Tucker–Lewis index (TLI) depended on the MT. Third, the performance of standardized root mean squared residual (SRMR) improved as ICC increased and this pattern was more clear in structure misspecification than in measurement misspecification. Finally, the summary and implications of the results are discussed.

Structure of Negative Symptoms in Schizophrenia: An Unresolved Issue

Background: Negative symptoms are core features of schizophrenia and very challenging to be treated. Identification of their structure is crucial to provide a better treatment. Increasing evidence supports the superiority of a five-factor model (alogia, blunted affect, anhedonia, avolition, and asociality as defined by the NMIH-MATRICS Consensus); however, previous data primarily used the Brief Negative Symptoms Scale (BNSS). This study, including a calibration and a cross-validation sample (n = 268 and 257, respectively) of participants with schizophrenia, used the Clinical Assessment Interview for Negative Symptoms (CAINS) to explore the latent structure of negative symptoms and to test theoretical and data-driven (from this study) models of negative symptoms.Methods: Exploratory factor analysis (EFA) was carried out to investigate the structure of negative symptoms based on the CAINS. Confirmatory factor analysis (CFA) tested in a cross-validation sample four competing theoretical (one-factor, two-factor, five-factor, and hierarchical factor) models and two EFA-derived models.Result: None of the theoretical models was confirmed with the CFA. A CAINS-rated model from EFA consisting of five factors (expression, motivation for recreational activities, social activities, vocational, and close/intimate relationships) was an excellent fit to the data (comparative fix index = 0.97, Tucker–Lewis index = 0.96, and root mean square error of approximation = 0.07).Conclusions: This study cannot support recent data on the superiority of the five-factor model defined by the NMIH-MATRICS consensus and suggests that an alternative model might be a better fit. More research to confirm the structure of negative symptoms in schizophrenia, and careful methodological consideration, should be warranted before a definitive model can put forward and shape diagnosis and treatment of schizophrenia.

Transcultural Adaptation and Psychometric Proprieties of the Mental Toughness Inventory for Brazilian Athletes

This study has assessed the psychometric proprieties of the Mental Toughness Inventory (MTI) within the context of Brazilian sports. About 12 professionals participated in the process of adapting and translating the scale to Brazilian Portuguese. Subjects were 575 athletes (23.54 ± 5.79 years old; 58% males) who answered the MTI and the 10-item Connor–Davidson Resilience Scale (CD-RISC-10). Data were analyzed through confirmatory factor analysis (CFA), Cronbach's alpha (α), composite reliability (CR), average variance extracted (AVE), Spearman correlation, and model invariance tests. Results from CFA showed adequate fit for the original 8-item structure of the scale [Chi-square (χ2) = 27.041; p = 0.078; normalized chi-square (χ2/df) = 1.50; comparative fit index (CFI) = 0.988; Tucker–Lewis Index (TLI) = 0.981; root mean square error of approximation (RMSEA) = 0.03 [0.00–0.05]; standardized root mean residual (SRMR) = 0.030] assessing mental toughness (MT) as a single factor and the scale presented satisfactory internal consistency (CR = 0.81; α = 0.82). MT was correlated with resilience (r = 0.607), age (r = 0.276), and time of experience in the sport (r = 0.215). The MTI has also shown partial measurement invariance for sex and complete invariance across sport types. It was concluded that the MTI is a suitable tool for assessing MT in the present sample of Brazilian athletes; this instrument has potential practical application for researchers and sports psychologists who seek to develop the well-being and performance of athletes.

Análise da estrutura fatorial do Audit em adolescentes entre 18 e 19 anos

OBJETIVO: Determinar a estrutura fatorial do instrumento Alcohol Use Disorders Identification Test (Audit) em uma amostra representativa de adolescentes de 18 a 19 anos. MÉTODOS: Estudo transversal realizado com adolescentes nascidos em São Luís (MA). A consistência interna do instrumento foi determinada pelo coeficiente alfa de Cronbach e a validade do construto foi avaliada por meio da Análise Fatorial Confirmatória (AFC). O índice de Kaiser-Meyer-Olkin (KMO) foi calculado para analisar a adequação da amostra. A qualidade de ajuste do modelo fatorial foi analisada de acordo com os índices dos testes qui-quadrado de ajustamento, comparative fit index (CFI), Tucker-Lewis index (TLI) e root mean square error of approximation (RMSEA). RESULTADOS: A amostra do estudo foi composta por 1.002 adolescentes entre 18 e 19 anos, sendo 56,8% meninas, 68,5% com 18 anos, 63,3% pardos ou mulatos, 48,6% pertencentes à classe C, 15,4% não trabalhavam e não estudavam e 52,1% tinham pais separados. A amostra foi adequada para a análise fatorial confirmatória (KMO = 0,79) e o coeficiente do alfa de Cronbach foi de 0,70, demostrando consistência interna satisfatória com cargas fatoriais acima de 0,5, com exceção do item 9 “Ficou ferido ou ficou alguém ferido por ter bebido”. A análise fatorial confirmatória revelou a validade do modelo de três fatores para a amostra em estudo com base nos índices de ajustes psicométricos. CONCLUSÃO: A estrutura fatorial do Audit com três fatores foi confirmada para a população de adolescentes entre 18 e 19 anos residentes em São Luís, ratificando os domínios conceituais originais propostos pela Organização Mundial da Saúde. O Audit apresentou-se como um instrumento confiável para a identificação do consumo de álcool.

Incremental Model Fit Assessment in the Case of Categorical Data: Tucker–Lewis Index for Item Response Theory Modeling

The Tucker–Lewis index (TLI; Tucker & Lewis, 1973), also known as the non-normed fit index (NNFI; Bentler & Bonett, 1980), is one of the numerous incremental fit indices widely used in linear mean and covariance structure modeling, particularly in exploratory factor analysis, tools popular in prevention research. It augments information provided by other indices such as the root-mean-square error of approximation (RMSEA). In this paper, we develop and examine an analogous index for categorical item level data modeled with item response theory (IRT). The proposed Tucker–Lewis index for IRT (TLIRT) is based on Maydeu-Olivares and Joe's (2005) $$M_2$$ M 2 family of limited-information overall model fit statistics. The limited-information fit statistics have significantly better Chi-square approximation and power than traditional full-information Pearson or likelihood ratio statistics under realistic situations. Building on the incremental fit assessment principle, the TLIRT compares the fit of model under consideration along a spectrum of worst to best possible model fit scenarios. We examine the performance of the new index using simulated and empirical data. Results from a simulation study suggest that the new index behaves as theoretically expected, and it can offer additional insights about model fit not available from other sources. In addition, a more stringent cutoff value is perhaps needed than Hu and Bentler's (1999) traditional cutoff criterion with continuous variables. In the empirical data analysis, we use a data set from a measurement development project in support of cigarette smoking cessation research to illustrate the usefulness of the TLIRT. We noticed that had we only utilized the RMSEA index, we could have arrived at qualitatively different conclusions about model fit, depending on the choice of test statistics, an issue to which the TLIRT is relatively more immune.

Modelagem de Equações Estruturais e os testes de seleção – Caso do vestibular da Universidade Estadual do Ceará

Resumo Este estudo teve como objetivo fazer uso da Modelagem de Equações Estruturais – MEE, para avaliar a qualidade do modelo empregado pela Universidade Estadual do Ceará (UECE), no processo de seleção de seus vestibulandos, o qual é baseado na Teoria Clássica dos Testes (TCT), assim como propor ajuste a esse modelo pelo uso da Análise Fatorial de Segunda Ordem e da Análise de Regressão, via MEE. Utilizou-se um banco de dados composto dos resultados das provas de 11.060 candidatos ao vestibular de 2018.1, cujo tratamento se deu por meio do software IBM SPSS Amos (2013, v.22), obtendo-se os seguintes indicadores de qualidade: CFI ( Comparative Fit Index ) = 0,925; GFI ( Goodness-of-fit Index ) = 0,965; TLI ( Tucker Lewis Index ) = 0,922, e RMSEA ( Root Mean Square Error of Aproximation ) = 0,019. Juntos, esses indicadores demonstraram que o modelo é robusto e bastante consistente, apresentando um R 2 (Coeficiente de Correlação de Pearson ) = 0,965, indicando que a proporção das covariâncias observadas entre as variáveis manifestas e explicada pelo modelo ajustado é bastante significativa. Todas as variáveis do modelo ajustado apresentaram elevados coeficientes de regressão com valores entre 0,87 e 0,99, permitindo uma boa discriminação entre as notas dos vestibulandos, principalmente aqueles com o mesmo número de questões respondidas corretamente.

Determining the Number of Factors When Population Models Can Be Closely Approximated by Parsimonious Models

Despite the existence of many methods for determining the number of factors, none outperforms the others under every condition. This study compares traditional parallel analysis (TPA), revised parallel analysis (RPA), Kaiser’s rule, minimum average partial, sequential χ2, and sequential root mean square error of approximation, comparative fit index, and Tucker–Lewis index under a realistic scenario in behavioral studies, where researchers employ a closing–fitting parsimonious model with K factors to approximate a population model, leading to a trivial model-data misfit. Results show that while traditional and RPA both stand out when zero population-level misfits exist, the accuracy of RPA substantially deteriorates when a K-factor model can closely approximate the population. TPA is the least sensitive to trivial misfits and results in the highest accuracy across most simulation conditions. This study suggests the use of TPA for the investigated models. Results also imply that RPA requires further revision to accommodate a degree of model–data misfit that can be tolerated.

The creation and validation of a Jazz Theory Measure: Implications for improvisation education

The purpose of this study was to create and validate a measure for jazz theory knowledge assessment. The Jazz Theory Measure (JTM) was created by exploring publications on jazz improvisation instruction and research studies that utilized jazz theory as a significant aspect of the study (i.e., dependent variable, independent variable, data collection). Latent jazz theory components (i.e., Key Knowledge, Scale Knowledge, and Chord Knowledge) were identified from an exploration of the literature and each was paired with several subtests. Items representing the JTM ( N = 60) demonstrate significant representation of the jazz theory components through alpha reliability (.98), and confirmatory factor analysis (standardized root mean square residual [SRMR] = .0327, comparative fit index [CFI] = .930, Tucker–Lewis index [TLI] = .907) providing support for the JTM as a valid and reliable measure of jazz theory knowledge. Further analysis suggests that music educators providing instruction in jazz improvisation should focus on building a foundation of jazz theory in the first 5 years of jazz instruction.

Diseño y validación de una escala de estrés financiero en trabajadores dependientes peruanos durante la pandemia del COVID-19 (EFT-Cov19)

La pandemia del COVID-19 ha generado estragos en la vida de los trabajadores de diferentes partes del mundo. La inestabilidad propia de esta etapa de emergencia sanitaria ha repercutido en la salud mental de esta población. El objetivo fue evaluar las propiedades psicométricas de una escala de estrés financiero para trabajadores dependientes peruanos. Estudio observacional, analítico, instrumental y transversal en 749 trabajadores. Se realizó un Análisis Factorial Exploratorio (AFE), por mínimos cuadrados no ponderados, previo análisis de la prueba de Bartlett y el coeficiente Kaiser-Meyer-Olkin (KMO). La bondad de ajuste absoluta e incremental se determinó a través del índice de ajuste comparativo (CFI) y el índice de Tucker-Lewis Index (TLI). Se realizó un AFE previo análisis del índice Kaiser-Meyer-Olkin (KMO = 0,903) y la prueba de esfericidad de Bartlett (1751.9; gl = 36; p < 0,001), los cuales fueron adecuados. Los ítems convergieron en un solo factor. La EFT-Cov19 correlacionó positivamente con la LABOR-PE (r = 0,564, p < 0,01) y con un tamaño del efecto mediano. La fiabilidad de la EFT-Cov19 se calculó con el coeficiente α de Cronbach, obteniendo un valor aceptable (α = 0,896; IC 95% = 0,88 – 0,90). En conclusión, la escala EFT-Cov19 es una escala valida, confiable y adecuada para medir el estrés financiero en trabajadores dependientes durante la pandemia del COVID-19.

Internet and decision-making regarding health among pregnant woman: cross-cultural adaptation of a questionnaire for use in Brazil

The aim of this study was to translate and cross-culturally adapt the Women’s Use of the Internet in Pregnancy Questionnaire (WUIPQ) to Brazilian Portuguese and analyze the psychometric properties of the Preparation for Decision Making Scale (PDMS). This study consisted of the following steps: translation, synthesis, back-translation, evaluation by the author of the original questionnaire, review by the panel of experts, and pretest of the WUIPQ. For such, Brazilian pregnant women and mothers who were members of Facebook groups participated in the study. We measured test-retest reliability as well as internal consistency and performed confirmatory factor analysis (CFA) of the B-PDMS. In the pretest, 88.14% of the participants considered the items of the B-WUIPQ to be clear and pertinent, and 84.09% rated the sequence and organization of the questionnaire as excellent/good. The intraclass correlation coefficient and Cronbach’s alpha coefficient for the B-PDMS were 0.850 (95%CI: 0.791-0.899) and 0.91, respectively. CFA revealed factor loadings higher than 0.70 for most items, with a comparative fit index of 0.989, Tucker-Lewis index of 0.984, and root mean square error of approximation of 0.08 (95%CI: 0.06-0.09). The B-WUIPQ presented cross-cultural adapted, and the B-PDMS demonstrated satisfactory psychometric proprieties to Brazilian pregnant women.