Multiple-group Confirmatory Factor Analysis Research Articles

An Indian Outlook on Role Clarity, Organizational Citizenship Behavior, and Gender Relationship: Multiple Group Confirmatory Factor Analysis (MGCFA) Approach

This article investigates the empirical association between role clarity and organizational citizenship behavior and how this relation is moderated by gender. The data of 272 executives have been collected from management development programs conducted in various public and private Indian organizations. Chi-square difference test (Bonferroni method), latent mean approach, sub-group analyses, and multiple group confirmatory factor analysis have been used to analyze the responses. Results indicate that role clarity is positively related to all dimensions of organizational citizenship behavior. Out of five, four dimensions show signs of gender’s moderation; women scored higher in role clarity’s relation with altruism, courtesy, and conscientiousness and men showed higher role clarity–civic virtue relationship. Implications and future scope are also discussed.

Measurement invariance of the Eating Disorder Examination in black and white children and adolescents.

The Eating Disorder Examination (EDE) was originally developed and validated in primarily white female samples. Since data indicate that eating pathology impacts black youth, elucidating the psychometric appropriateness of the EDE for black youth is crucial. A convenience sample was assembled from seven pediatric obesity studies. The EDE was administered to all youth. Confirmatory factor analyses (CFA) were conducted to examine the original four-factor model fit and two alternative factor structures for black and white youth. With acceptable fit, multiple-group CFAs were conducted. For measurement invariant structures, the interactive effects of race with sex, BMIz, adiposity, and age were explored (all significance levels p < .05). For both black and white youth (N=820; 41% black; 37% male; 6-18 years; BMIz -3.11 to 3.40), the original four-factor EDE structure and alternative eight-item one-factor structure had mixed fit via CFA. However, a seven-item, three-factor structure reflecting Dietary Restraint, Shape/Weight Overvaluation, and Body Dissatisfaction had good fit and held at the level of strict invariance. Girls reported higher factor scores than boys. BMIz and adiposity were positively associated with each subscale. Age was associated with Dietary Restraint and Body Dissatisfaction. The interactional effects between sex, BMIz, and age with race were not significant; however, the interaction between adiposity and race was significant. At higher adiposity, white youth reported greater pathology than black youth. An abbreviated seven-item, three-factor version of the EDE captures eating pathology equivalently across black and white youth. Full psychometric testing of the modified EDE factor structure in black youth is warranted.

Measurement Invariance Testing with Many Groups: A Comparison of Five Approaches

With the increasing use of international survey data especially in cross-cultural and multinational studies, establishing measurement invariance (MI) across a large number of groups in a study is essential. Testing MI over many groups is methodologically challenging, however. We identified 5 methods for MI testing across many groups (multiple group confirmatory factor analysis, multilevel confirmatory factor analysis, multilevel factor mixture modeling, Bayesian approximate MI testing, and alignment optimization) and explicated the similarities and differences of these approaches in terms of their conceptual models and statistical procedures. A Monte Carlo study was conducted to investigate the efficacy of the 5 methods in detecting measurement noninvariance across many groups using various fit criteria. Generally, the 5 methods showed reasonable performance in identifying the level of invariance if an appropriate fit criterion was used (e.g., Bayesian information criteron with multilevel factor mixture modeling). Finally, general guidelines in selecting an appropriate method are provided.

Students’ Conceptions of Assessment

Demonstrating factor invariance across population characteristics is an important facet of validating instruments. A new instrument to examine students’ conceptions of assessment (SCoA) has been demonstrated in studies to show significant relations to secondary school students’ academic performance in mathematics and reading. The research reported here investigates the stability of the factor structure and structural relationships to achievement in reading comprehension across student sex, year-level, and ethnicity. Multiple group confirmatory factor analysis found, in a sample of 3,506 students, that the factor structure of the SCoA was invariant. Conversely, statistically significant differences for sex, year, and ethnicity were found for how the conceptions of assessment related to academic performance in reading. These differences between the structural models may not indicate a deficiency in the instrument but, rather, sensitivity to real-world differences among subgroups. The structural differences ...

Monitoring Preadolescents’ Well-being: Italian Validation of the Middle Years Development Instrument

The aim of this study was to translate into Italian and validate the Middle Years Development Instrument (MDI). This population-level measure was developed and validated in English by Schonert-Reichl and colleagues Social Indicators Research, 114, 345–369, (2013) to measure well-being in preadolescents who were aged between six and twelve years. Our purpose was to test the MDI in the Swiss State of Canton Ticino, in which Italian is the official language. A total of 1942 6th and 7th grade preadolescents completed the questionnaire (50% girls). To assess the factor structure of the scales, exploratory factor analysis (EFA) and confirmatory factor analyses (CFA) were performed, which identified Social and emotional development, Connectedness and School experiences dimensions of the MDI Italian-version, exploratory factor analysis (EFA) and confirmatory factor analyses (CFAs) were performed. Then, multiple-group CFAs were computed in order to assess gender and native language invariance. We computed Cronbach’s alphas and correlations between scales. EFA and CFAs confirmed that the factor structure of the MDI was adequate. Analyses of invariance supported the factor structure of the MDI and its utility for comparative research. In general, the analyses confirmed that the MDI-Swiss-Italian version is a valid and reliable instrument for measuring preadolescents’ well-being.

To what extent do teachers in European countries differ in their professional community practices?

ABSTRACTWithin comparative school effectiveness research facilitated by large-scale data across countries, this article presents the results of the testing for measurement invariance of the latent concept of Professional Community (PC) across 23 European countries and more than 35,000 teachers in secondary schools. The newly proposed Multiple-Group Factor Analysis Alignment method is applied to obtain the mean differences of the latent PC factor, while also identifying the non-invariant parameters in these different countries. After performing a number of robustness checks, including the Multiple-Group Confirmatory Factor Analysis (MGCFA) exact scalar method, we conclude that teachers in Eastern European countries report the highest average perceived PC practices. The countries with smaller PC mean differences tend to shift places in the ranking when the 2 methods are applied; however, very few changes among the groups with significant latent mean differences are observed.

Validating the shortened Quality of Life in Childhood Epilepsy Questionnaire (QOLCE-55) in a sample of children with drug-resistant epilepsy.

The aim of this study was to validate the newly developed shortened Quality of Life in Childhood Epilepsy Questionnaire (QOLCE-55) in a sample of children with drug-resistant epilepsy. Data came from 136 children enrolled in the Impact of Pediatric Epilepsy Surgery on Health-Related Quality of Life Study (PEPSQOL), a multicenter prospective cohort study. Confirmatory factor analysis was used to assess the higher-order factor structure of the QOLCE-55. Convergent and divergent validity was assessed by correlating subscales of the KIDSCREEN-27 with the QOLCE-55. Measurement equivalence of the QOLCE-55 was evaluated using multiple-group confirmatory factor analysis of children with drug-resistant epilepsy from PEPSQOL versus children with new-onset epilepsy from HERQULES (Health-Related Quality of Life in Children with Epilepsy Study). The higher-order factor structure of the QOLCE-55 demonstrated adequate fit: Comparative Fit Index (CFI) = 0.948; Tucker-Lewis Index (TLI) = 0.946; Root Mean Square of Approximation (RMSEA) = 0.060 (90% confidence interval [CI] 0.054-0.065); Weighted Root Mean Square Residuals (WRMR) = 1.247. Higher-order factor loadings were strong, ranging from λ = 0.74 to 0.81. Internal consistency reliability was excellent (α = 0.97, subscales α > 0.82). QOLCE-55 subscales demonstrated moderate to strong correlations with similar subscales of the KIDSCREEN-27 (ρ = 0.43-0.75) and weak to moderate correlations with dissimilar subscales (ρ = 0.25-0.42). The QOLCE-55 demonstrated partial measurement equivalence at the level of strict invariance - χ2 (2,823) = 3,727.9, CFI = 0.961, TLI = 0.962, RMSEA = 0.049 (0.044, 0.053), WRMR = 1.834. The findings provide support for the factor structure of the QOLCE-55 and contribute to its robust psychometric profile as a reliable and valid measure. Researchers and health practitioners should consider the QOLCE-55 as a viable option for reducing respondent burden when assessing health-related quality of life in children with epilepsy.

Measurement Invariance of a School Readiness Screener for Use in Preschool and Kindergarten

ABSTRACTResearch Findings: Public policy has increasingly focused on expansion of preschool access for underserved students and systematic evaluation of preschool quality and students’ readiness for school. However, such evaluation is limited by a lack of thoroughly validated assessments for use with preschool populations. The present study examined the measurement and structural invariance of the Kindergarten Student Entrance Profile (KSEP) across kindergarten and prekindergarten groups to evaluate its potential use across developmental groups. Participants included 522 kindergarten and 548 prekindergarten students in central California. Invariance was tested by fitting a series of multiple-groups confirmatory factor analysis models with parameter constraints across groups. Results indicated that measurement and structural parameters of the KSEP were invariant across kindergarten and prekindergarten groups. Prekindergarten means on both Social–Emotional Readiness and Cognitive Readiness were significantly lower than kindergarten means. Practice or Policy: These results suggest that the KSEP may potentially be used with prekindergarten students to assess school readiness and inform intervention before kindergarten entry.

Measurement invariance of intuitive cancer risk perceptions across diverse populations: The Cognitive Causation and Negative Affect in Risk scales.

Intuitive cancer risk perceptions may inform strategies to motivate cancer prevention behaviors. This study evaluated factor structure and measurement invariance of two new measures of intuitive cancer risk, the Cognitive Causation and Negative Affect in Risk scales. Single- and multiple-group confirmatory factor analysis models were fit to responses from three diverse samples. The confirmatory factor analysis models fit the data well, with all comparative fit indices (CFI) ≥ 0.94. Items flagged by chi-square difference tests as potentially non-invariant were largely invariant between samples according to practical fit indices (e.g. ΔCFI). These novel scales may be particularly relevant in diverse, underserved populations.

Poverty in Ukraine: Development, Validity and Reliability of a New Measure of Financial Strain for Young Adults.

The purpose of this paper was to develop a valid and reliable measure of financial strain for use with college-attending young adults in Ukraine. The newly developed measure represents an alternative approach to currently used objective measures of poverty and economic hardship. Objective measures are not adequate or applicable for use with Ukrainian young adults who are mainly dependent on their families for economic support. Financial strain was conceptualized as the financial adjustments that young adults have to make to meet their basic housing, food and clothing needs. Items were generated to capture the construct of financial strain, and content validity of the scale was assessed. Exploratory factor analysis strongly supported the unidimensional nature of the scale. In addition, findings from simultaneous multiple-group confirmatory factor analyses indicated configural, metric and factor invariance of the measure across the southern and central regions of Ukraine. The new measure of financial strain was positively correlated with emotional distress and violence against romantic partners, indicating good predictive validity. The scale also showed good internal consistency. We suggest that the new measure is appropriate to use with young adults in other Eastern European countries in transition.

Validating the factor structure and testing measurement invariance of modified Short-Form McGill Pain Questionnaire (Ortho-SF-MPQ) for orthodontic pain assessment.

To validate the factor structure of recently modified Short-Form McGill Pain Questionnaire (Ortho-SF-MPQ) to assess orthodontic pain; and to test its Measurement Invariance (MI) across gender. 180 orthodontic patients were enrolled in this study. 0.016 inch Super-elastic NiTi arch wire was used in 0.022″ × 0.028″ slot pre-adjusted edgewise appliance. After initial arch wire placement, pain was assessed at T1 (24 hours), T2 (day 3), and T3 (day 7) by using the Ortho-SF-MPQ which consists of seven sensory (pressure, sore, aching, tight, throbbing, pulling, miserable) and four affective (uncomfortable, strange, frustrating, annoying) descriptors. Confirmatory factor analysis (CFA) models were fitted for analysis. Multiple-groups CFA (MG-CFA) approach was used for MI testing. Data from 172 patients (85 male, 87 female) with mean age 14.2 years (SD 1.4) was analysed. CFA model fit indices value at T1 (RMSEA 0.048; CFI 0.995; TLI 0.995), T2 (RMSEA 0.051; CFI 0.998; TLI 0.997), and T3 (RMSEA 0.040; CFI 0.998; TLI 0.998) confirmed the validity of two-factor structure of Ortho-SF-MPQ in assessing orthodontic pain. MG-CFA model based non-significant scaled chi-square difference test (Satorra-Bentler method) for weak invariance (T1: χ2 = 6.566, df = 9, p = 0.682; T2: χ2 = 14.637, df = 9, p = 0.101; T3 (χ2 = 14.248, df = 9, p = 0.114) and strong invariance (T1: χ2 = 25.874, df = 20, p = 0.170; T2: χ2 = 25.052, df = 20, p = 0.199; T3: χ2 = 18.889, df = 20, p = 0.529) confirmed MI across male and female groups. Two-factor structure (sensory and affective) of Ortho-SF-MPQ is structurally valid and invariant to measure pain in male and female orthodontic patents after initial arch wire placement.

The Metacognitions Questionnaire-30: An Examination of a Bifactor Model and Measurement Invariance Among Men and Women in a Community Sample.

The Metacognitions Questionnaire-30 (MCQ-30) is a self-report measure that assesses metacognitive beliefs (i.e., beliefs about thinking). Prior research has supported a correlated five-factor model, but no known published study has examined the tenability of second-order or bifactor models of the MCQ-30. Results supported a bifactor model of the MCQ-30 in a sample of community adults from the United States ( N = 785), as well as separately among men ( n = 372) and women ( n = 413). Multiple-groups confirmatory factor analysis supported the configural and metric/scalar invariance of the bifactor model among men and women. Results further supported the incremental validity of one of the MCQ-30 domain-specific factors in accounting for unique variance in an index of health anxiety beyond the general metacognition factor. Results provide support for a bifactor conceptualization of the MCQ-30 and the invariance of that model across men and women.

Factor structure of the ARIC-NCS Neuropsychological Battery: An evaluation of invariance across vascular factors and demographic characteristics.

Neuropsychological test batteries are designed to assess cognition in detail by measuring cognitive performance in multiple domains. This study examines the factor structure of tests from the ARIC-NCS battery overall and across informative subgroups defined by demographic and vascular risk factors in a population of older adults. We analyzed neuropsychological test scores from 6,413 participants in the Atherosclerosis Risk in Communities Neurocognitive Study (ARIC-NCS) examined in 2011-2013. Confirmatory factor analysis (CFA) was used to assess the fit of an a priori hypothesized 3-domain model, and fit statistics were calculated and compared to 1- and 2-domain models. Additionally, we tested for stability (invariance) of factor structures among different subgroups defined by diabetes, hypertension, age, sex, race, and education. Mean age of participants was 76 years, 76% were White, and 60% were female. CFA on the a priori hypothesized 3-domain structure, including memory, sustained attention and processing speed, and language, fit the data better (comparative fit index [CFI] = 0.973, root mean square error of approximation [RMSEA] = 0.059) than the 2-domain (CFI = 0.960, RMSEA = 0.070) and 1-domain (CFI = 0.947, RMSEA = 0.080) models. Bayesian information criterion value was lowest, and quantile-quantile plots indicated better fit, for the 3-domain model. Additionally, multiple-group CFA supported a common structure across the tested demographic subgroups, and indicated strict invariance by diabetes and hypertension status. In this community-based population of older adults with varying levels of cognitive performance, the a priori hypothesized 3-domain structure fit the data well. The identified factors were configurally invariant by age, sex, race, and education, and strictly invariant by diabetes and hypertension status. (PsycINFO Database Record

Two-Step Approach to Partial Factorial Invariance: Selecting a Reference Variable and Identifying the Source of Noninvariance

To date, no effective empirical method has been available to identify a truly invariant reference variable (RV) in testing measurement invariance under a multiple-group confirmatory factor analysis. This study proposes a method that, in selecting an RV, uses the smallest modification index (min-mod). The method’s performance is evaluated using 2 models: (a) a full invariance model, and (b) a partial invariance model. Results indicate that for both models the min-mod successfully identifies a truly invariant RV (Study 1). In Study 2, we use the RV found in Study 1 to further evaluate the performance of item-by-item Wald tests at locating a noninvariant variable. The results indicate that Wald tests overall performed better with an RV selected in a partial invariance model than an RV selected in a full invariance model, although in certain conditions their performances were rather similar. Implications and limitations of the study are also discussed.

The Value Of Item Response Theory In Invariance Testing

The goal of the current study was to assess the Employee Engagement Instrument (EEI) from an item response theory (IRT) perspective, with a specific focus on measurement invariance for annual turnover. The sample comprised 4 099 respondents from all business sectors in South Africa. This article describes the logic and procedures used to test for factorial invariance across groups in the context of construct validation. The procedures included testing for configural and metric invariance in the framework of multiple-group confirmatory factor analysis (CFA).The results confirmed the factor analytic structure of the model fit for some of the individual scales of the EEI. The measurement invariance of the EEI as a function of annual turnover was confirmed. However, the results indicated that the EEI needs to be refined for future research.

EXTENDING THE VALIDITY OF THE FAMILY INVOLVEMENT QUESTIONNAIRE–SHORT FORM FOR CULTURALLY AND LINGUISTICALLY DIVERSE FAMILIES FROM LOW‐INCOME BACKGROUNDS

The construct validity of the Family Involvement Questionnaire–Short Form (FIQ‐SF) was examined in an independent sample of ethnically and linguistically diverse low‐income families (N = 498) enrolled in an urban Head Start program in the Southeast. A series of exploratory and confirmatory factor analyses replicated the three‐factor structure identified in initial validation studies with Northeast samples: home‐school conferencing, home‐based involvement, and school‐based involvement. Findings from multiple group confirmatory factor analyses provided evidence that the three‐factor structure was invariant across family ethnicity. multivariate analyses of variance also confirmed hypothesized mean differences on FIQ‐SF dimension scores across family demographic variables such as caregiver ethnicity, primary home language, caregiver education, employment, and marital status. Findings replicate and extend prior construct validity evidence to support the use of the FIQ‐SF by early childhood education programs such as Head Start serving diverse families from low‐income backgrounds. Implications for future research, practice, and policy applications in early childhood are discussed.

Psychometric Properties of the Generalized Anxiety Disorder Scale-7 (GAD-7) in Outpatients with Anxiety and Mood Disorders.

This study examined the psychometric properties of the Generalized Anxiety Disorder Scale-7 (GAD-7) in a sample of 536 outpatients presenting at a specialty clinic for anxiety and mood disorders. A confirmatory factor analysis (CFA) was used to test the unidimensionality of the GAD-7. This model did not fit the data well. The CFA solution was respecified correlating residuals among items assessing somatic symptoms. This respecified model fit the data well. A series of multiple-groups CFAs determined that the measurement properties of the GAD-7 were invariant between sexes. Scale reliability estimates of the GAD-7 were favorable for the full sample, and for males and females. Sensitivity and specificity could not be balanced at any cut-point. Findings attest to the value of this instrument as a dimensional indicator of GAD severity rather than a screening tool for the presence or absence of the disorder in outpatients with anxiety and mood disorders.

Analyzing the Measurement Equivalence of a Translated Test in a Statewide Assessment Program

When tests are translated into one or more languages, the question of the equivalence of items across language forms arises. This equivalence can be assessed at the scale level by means of a multiple group confirmatory factor analysis (CFA) in the context of structural equation modeling. This study examined the measurement equivalence of a Spanish translated version of a statewide Mathematics test originally constructed in English by using a multi-group CFA approach. The study used samples of native speakers of the target language of the translation taking the test in both the source and target language, specifically Hispanics taking the test in English and Spanish. Test items were grouped in twelve facet-representative parcels. The parceling was accomplished by grouping items that corresponded to similar content and computing an average for each parcel. Four models were fitted to examine the equivalence of the test across groups. The multi-group CFA fixed factor loadings across groups and results supported the equivalence of the two language versions (English and Spanish) of the test. The statistical techniques implemented in this study can also be used to address the performance on a test based on dichotomous or dichotomized variables such as gender, socioeconomic status, geographic location and other variables of interest.

Quality of urban life among older adults in the world major metropolises: a cross-cultural comparative study

ABSTRACTThe concept of quality of urban life (QoUL) can be interpreted quite differently across different cultures. Little evidence has shown that the measure of QoUL, which is based on Western culture, can be applied to populations cross-culturally. In the current study, we use data from the 2006 Assessing Happiness and Competitiveness of World Major Metropolises study to identify underlying factors associated with QoUL as well as assess the consistency of the QoUL measurement among adults, aged 60 and older, in ten world major metropolises (i.e. New York City, Toronto, London, Paris, Milan, Berlin, Stockholm, Beijing, Tokyo and Seoul). Exploratory factor analysis and multiple-group confirmatory factor analysis (CFA) are used to analyse the data. Findings of the study suggest that the measure of QoUL is sensitive to socio-cultural differences. Community factor and intrapersonal factor are two underlying structures that are related to QoUL among older adults in ten metropolises cross-culturally. Results from the CFA indicate that Toronto is comparable with Beijing, New York City, Paris, Milan and Stockholm in QoUL, while other cities are not. The results provide insights into the development of current urban policy and promotion of quality of life among older residents in major metropolitan areas. Future researchers should continue to explore the relationship between QoUL and socio-cultural differences within international urban settings, while remaining cautious when making cross-cultural comparisons.

Locals’ Bidimensional Acculturation Model

Across two studies, we tested whether members of host communities (i.e., locals) can themselves simultaneously maintain their national culture maintenance and adapt toward cultural diversity (i.e., multiculturalism) in their own home country, supporting a bidimensional model of acculturation, or whether these strategies are incompatible, supporting a unidimensional model of acculturation. We modified the Vancouver Index of Acculturation (Multi-VIA) to assess locals’ national culture maintenance and multicultural adaptation within their own home country. Study 1 supported the bidimensionality of the Multi-VIA in an American sample ( n = 218). Moreover, we found an oblique association between locals’ national culture maintenance and multicultural adaptation. In Study 2, we tested the Multi-VIA’s psychometric properties across three continent groups (North America, Europe, and Asia; N = 619). Multiple-group confirmatory factor analysis demonstrated good model fit for the entire sample. Nevertheless, the association between national culture maintenance and multicultural adaptation was orthogonal for Asians and oblique for Americans and Europeans. In addition, national culture maintenance predicted higher levels of locals’ life satisfaction, whereas multicultural adaptation was associated with less acculturative stress and greater intercultural sensitivity.