Measurement Model Invariance Research Articles

Measurement Invariance and Cohort Trends for Social and Emotional Learning Measures Across Four Statewide Administrations: Conventional Fit Statistics Versus the RMSEAD

We assessed measurement invariance (MI) and cohort trends for five social and emotional learning (SEL) measures across four administrations (2013, 2016, 2019, 2022) of a statewide student survey including grades 5, 8, 9, and 11 ( n = 626,082). The MI models were compared using conventional fit statistics and the root mean square error of approximation based on the chi-square difference test (RMSEAD) for nested model comparisons with ordinal items. We found that MI held for all measures across years for each grade using both criteria. We then computed standardized mean differences to identify SEL trends for the 2013 and 2016 cohorts for each measure. Overall, there were declines in SEL skills across administrations for both cohorts. Specifically, commitment to learning, family/community support, and teacher/school support SEL measures had notable declines (i.e., large effect size) from 2019–2022 (i.e., before and after the COVID-19 pandemic) for the 2016 cohort. Education implications and directions for future research for evaluating MI and measuring SEL skills and supports were also discussed.

Examination of the psychometric properties of the Ethnic Identity Scale (EIS) and Multicultural Identity Integration Scale (MULTIIS) in a multiracial population in the United States.

Ethnic identity is theorized to be a critical aspect of human development and is shown to be associated with health and well-being. The Ethnic Identity Scale is a widely used measure that assesses key aspects of ethnic identity development (Umaña-Taylor et al., 2004). The Multicultural Identity Integration Scale (MULTIIS) is a measure that has been more recently developed to assess key aspects of identity integration for individuals with multicultural identities (Yampolsky et al., 2016). Despite the ongoing utilization of these instruments, a comprehensive psychometric evaluation within Multiracial populations has yet to be established in extant literature. Addressing this gap, the present study aims to examine the internal consistency, factor structure, and other psychometric characteristics of the Ethnic Identity Scale and MULTIIS within a sample of 1,012 Multiracial adults in the United States. The majority of the sample identified as female (67.5%, n = 683), straight (80.1%, n = 798), having attained less than a college degree (62.3%, n = 627), and having a household income less than $60,000 (57.4%, n = 552). The majority of participants (55%, n = 557) were classified as having White and non-White racial/ethnic backgrounds, 45.0% (n = 455) as non-White. Findings suggest the Ethnic Identity Scale fits the data poorly by all measures, despite supporting the three-factor structure recommended in the original study; the MULTIIS fits the data acceptably by all measures and supports both a three-factor first-order and eight-factor second-order structure recommended in the original study. Analyses of the MULTIIS three-factor first-order model's measurement invariance across race, gender, educational attainment, and household income identified variance for specific latent factors. Overall, the MULTIIS performed acceptably; however, studies relying on the MULTIIS should account for differential measurement. Implications for clinical, scientific, and public health practice are discussed. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Life Project Scale: a new measure to assess the coherence of the intended future

AbstractThis article identifies different types of validity and reliability evidence of the Life Project Scale (LPS), a new psychometric measure to assess the coherence of the intended future. It comprises three studies involving five samples that accounted for 2,808 participants from Brazil and Portugal. The final version of the LPS contains two factors with four items each. The factors assess identification (i.e., awareness of one’s intended future) and involvement (i.e., enactment of plans and actions in favor of one’s intended future). The two factors are indicators of the coherence of life projects’ intentional structures and action. Altogether, the three studies provide validity and reliability evidence based on exploratory and confirmatory factor analyses, the assessment of ceiling and floor effects, four types of reliability coefficients, relations to other measures (authenticity and meaning in life), and measurement invariance models. Therefore, the LPS is ready for further use, development, and testing in other research and practice contexts.

Employability and career beliefs inventory: a brief version for unemployed persons

BackgroundConsidering that beliefs may be assessed and changed, inventories measuring employability and career beliefs may be of utmost importance for career interventions.ObjectiveThis study introduces the psychometric properties of a brief version of the Employability and Career Beliefs Inventory (ECBI) in a sample of unemployed persons.Methods and resultsAltogether, 2023 unemployed persons aged from 18 to 66 years old and living in Southern Portugal participated in an online survey. The ECBI’s original internal structure was tested and did not fit the data. Exploratory and confirmatory factor analyses were implemented, and a three-factor solution was retained. The three factors discriminate three types of beliefs named growth, pessimism, and flexibility. Measurement invariance models identified scalar equivalence across gender and educational degree, and metric invariance across age. All items fit the graded response model’s parameters. The growth and flexibility subscales were less effective in the assessment of low latent trait levels, whereas the opposite was observed with the pessimism subscale. Internal consistency is good yet discrimination between factors is questionable. Correlations to career decision-making self-efficacy evidence validity based on the relations to other constructs.ConclusionDespite the limitations, the brief version of the ECBI proposed in this study is ready for further use and development among unemployed persons.

Factor structure and measurement invariance of the Depression anxiety stress scale (DASS-21) in Chinese left-behind and non-left-behind children: an exploratory structural equation modeling approach

BackgroundComprehensive data has shown that adolescents often suffer from depression, anxiety, and low self-esteem, and are in a particularly fragile stage of psychological, physiological, and social development. Left-behind children in particular tend to have significantly higher, state anxiety and depression compared to non-left-behind children. The Depression, Anxiety, and Stress Scale (DASS-21) is an effective tool for evaluating depression, anxiety, and stress, and is used to measure levels of depression, anxiety, and stress in groups from a variety of backgrounds. The purpose of this study was to determine the effectiveness, reliability, and measurement invariance of the DASS-21 in Chinese left-behind children.MethodThe test and re-test method was used (N = 676), and the exploratory structural equation model (Mplus v.8.3) used to verify basic measurement models. For measurement invariance, the configural, weak, strong, and strict models were tested. The reliability of the DASS-21 was also tested using the collected data.ResultsAnalysis results showed that the DASS-21 had a stable three-factor structure in the sample of left-behind children in China. The measurement invariance test showed that gender and time not only had strong invariance, but also strict invariance. The results of cross left and non-left invariance indicated a lack of strict invariance. Finally, the McDonald’s omega coefficient of the DASS-21 total scale was 0.864, and the internal consistency of each subscale was also good.ConclusionsThe DASS-21 is shown to be an effective and reliable tool for measuring depression, anxiety and stress in Chinese left-behind children.

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.

Psychometric Properties under EFA, CFA, Measurement Invariance, and IRT Models for Older Adults’ First Aids Knowledge Scale among Iranian Grandparents: The Modified Scale

This research aims to create and evaluate an assessment tool termed Older Adults’ First Aid Knowledge Scale, which measures the knowledge and attitude of Iranian grandparents about first aid. In accordance with COSIM guidelines, 485 individuals in southern Iran completed the instrument as part of a psychometric investigation. Rasch partial credit model (PCM), exploratory factor analysis (EFA), confirmatory factor analysis (CFA), and receiver operating characteristic (ROC) analysis were used to analyze the results. The final version of OFAKS consisted of 18 items that were validated through EFA, CFA, and item response theory (IRT) analysis. All items showed measurement invariance and consecutive response groupings in the predictable order, and the instrument had strong internal consistency. Although Rasch’s analysis demonstrated the significance of OFAKS, further investigations and testing in different settings are required to confirm the validity of the tool.

An inventory of human light exposure behaviour

Light exposure is an essential driver of health and well-being, and individual behaviours during rest and activity modulate physiologically relevant aspects of light exposure. Further understanding the behaviours that influence individual photic exposure patterns may provide insight into the volitional contributions to the physiological effects of light and guide behavioural points of intervention. Here, we present a novel, self-reported and psychometrically validated inventory to capture light exposure-related behaviour, the Light Exposure Behaviour Assessment (LEBA). An expert panel prepared the initial 48-item pool spanning different light exposure-related behaviours. Responses, consisting of rating the frequency of engaging in the per-item behaviour on a five-point Likert-type scale, were collected in an online survey yielding responses from a geographically unconstrained sample (690 completed responses, 74 countries, 28 time zones). The exploratory factor analysis (EFA) on an initial subsample (n = 428) rendered a five-factor solution with 25 items (wearing blue light filters, spending time outdoors, using a phone and smartwatch in bed, using light before bedtime, using light in the morning and during daytime). In a confirmatory factor analysis (CFA) performed on an independent subset of participants (n = 262), we removed two additional items to attain the best fit for the five-factor solution (CFI = 0.95, TLI = 0.95, RMSEA = 0.06). The internal consistency reliability coefficient for the total instrument yielded McDonald’s Omega = 0.68. Measurement model invariance analysis between native and non-native English speakers showed our model attained the highest level of invariance (residual invariance CFI = 0.95, TLI = 0.95, RMSEA = 0.05). Lastly, a short form of the LEBA (n = 18 items) was developed using Item Response Theory on the complete sample (n = 690). The psychometric properties of the LEBA indicate the usability for measuring light exposure-related behaviours. The instrument may offer a scalable solution to characterise behaviours that influence individual photic exposure patterns in remote samples. The LEBA inventory is available under the open-access CC-BY license. Instrument webpage: https://leba-instrument.org/ GitHub repository containing this manuscript: https://github.com/leba-instrument/leba-manuscript.

Scaling metric measurement invariance models

This paper aims at clarifying the questions regarding the effects of the scaling method on the discrepancy function of the metric measurement invariance model. We provide examples and a formal account showing that neither the choice of the scaling method in general nor the choice of a particular referent indicator affects the value of the discrepancy function. Thus, the test statistic is not affected by the scaling method, either. The results rely on an appropriate application of the scaling restrictions, which can be phrased as a simple rule: "Apply the scaling restriction in one group only!" We develop formulas to calculate the degrees of freedom of χ²-difference tests comparing metric models to the corresponding configural model. Our findings show that it is impossible to test the invariance of the estimated loading of exactly one indicator, because metric MI models aimed at doing so are actually equivalent to the configural model.

Problematic Internet Use in adolescents: New psychometric evidence for the Spanish short form of the Compulsive Internet Use Scale.

Problematic Internet Use (PIU) has become a worldwide problem in recent years. Among screening instruments for PIU, the Compulsive Internet Use Scale (CIUS) is perhaps the most widelyused. Psychometric properties of the full CIUS are not convincing, however, and the short form (CIUS-S) has shown promising results, albeit limited to the English version, with little evidence in Spanish. Therefore, the aim of the present work was to study the psychometric properties of the CIUS-S scores in a large sample of Spanish adolescents. The sample consisted of 1790 participants, 816 male (45.6%), 961 female (53.7%), and 13 other (0.7%) . Mean age was 15.70years (SD=1.26). The five-items one-dimensional model displayed appropriate goodness-of-fit indices. Strong measurement invariance model across age and partial across gender was found. Furthermore, the CIUS-S was positively associated with several indicators of poor well-being and mental health, and negatively associated with prosocial behavior, self-esteem, and feeling of belonging. Overall, the CIUS-S scores appear reliable and valid in its Spanish version for adolescent populations, supporting its aim to detect and prevent a problem that has become a major worldwide issue in the last years.

Confirmatory Factor Analysis and Measurement Invariance of the Functional Assessment of Cancer Therapy Lung Cancer Utility Index (FACT-LUI).

This work suggests that in addition to being originally designed for use as a utility index, the 7 FACT-LUI items together also fit simple CFA and measurement invariance models. This less expected result indicates that these items as a group are also potentially useful in non-preference-based applications.Clinical trials can make for challenging decisions concerning which patient-reported outcome measures to include without being burdensome. However, the literature suggests a need for improved reporting of quality of life in lung cancer in particular as well as cancer in general. Inclusion of more disease-specific items such as the FACT-LUI may allow for information gathering of both preference-based and non-preference-based data with less demand on patients, similar to what has been done with some generic instruments.

Measuring college belongingness: Structure and measurement of the Sense of Social Fit Scale.

Sense of belonging is theorized to be a fundamental human need and has been shown to have important implications in many domains of life, including academic achievement. The Sense of Social Fit scale (SSF; Walton & Cohen, 2007) is widely used to assess college belongingness, particularly to study differences in academic experiences along lines of gender and race. Despite its wide use, the instrument's latent factor structure and measurement invariance properties have not been reported in the published literature to date. Consequently, researchers regularly use subsets of the SSF's items without psychometric justification. Here, we explore and validate the SSF's factor structure and other psychometric properties, and we provide recommendations about how to score the measure. A one-factor model in Study 1 showed poor fit, and exploratory factor analyses extracted a four-factor solution. Study 2's confirmatory factor analyses demonstrated superior fit of a bifactor model with four specific factors (from Study 1) and one general factor. Ancillary analyses supported a total scale scoring method for the SSF and did not support computing raw subscale scores. We also tested the bifactor model's measurement invariance across gender and race, compared latent mean scores between groups, and established the model's criterion and concurrent validity. We discuss implications and suggestions for future research. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Understanding Public’s Adoption of Preventive Behavior During COVID-19 Pandemic Using Health Belief Model: Role of Psychological Capital and Health Appraisals

Recognizing the importance of prevention during a highly transmissible disease like COVID-19, this study analyzed the effect of health psychological capital and health appraisals (benefits, barriers, contextual threat, seriousness and susceptibility) on voluntary adoption of preventive behavior. The theoretical framework was based on health belief model. The study also proposed the mediating role of health appraisals between health psychological capital and preventive behavior. Participants from Pakistan ( N = 321) and Malaysia ( N = 343) completed an online survey. Samples from both countries were compared for measurement model invariance and variance in path coefficient. Partial least squares approach to SEM using SMART-PLS3 software was adopted to analyze the measurement model, structural model, importance-performance analysis, and mediation testing. Our finding showed that the model explained a higher variance in preventive behavior for Pakistani samples compared to Malaysian samples. Countries were found to moderate some of the proposed relationships. Benefits and threats are the two most important predictors of preventive behaviors in both countries. Psychological capital is more predictive of coping and threat appraisals and is, directly and indirectly, related to preventive behavior via benefits, barriers, and threats in Pakistani sample, and only indirectly related to preventive behavior via benefits and threats in the Malaysian sample. Perceived threat also mediates the relationship of susceptibility and seriousness with preventive behavior. Public health messages should focus on enhancing health psychological capital, highlight the potential benefit and emphasize the perils of not adopting the precautionary approach.

MIMIC model of teachers and students attitudes towards online learning during Covid-19: A gender perspective

Academic institutions around the globe have shifted to online learning because of the unpredictable spread of COVID-19. The present study aimed to compare teachers’ and students’ attitudes towards online learning during the pandemic and to examine the effects of gender differences on their attitudes. In study 1, we adapted the Test of eLearning Related Attitudes for Pakistani students in three steps: expert review, piloting, and validation. The individual and collective expert review was performed to adapt the teacher version into the student version using the Technique for Research of Information by the Animation of a Group of Experts (TRIAGE). We tested three sets of measurement invariance models for participants’ status and gender in study 2. Data were collected from 289 university teachers (men = 158, women = 131) and 444 undergraduate students (boys = 156, girls = 287). The results demonstrated that both groups had highly positive yet different attitudes towards online learning. Teachers were more satisfied than students. Model fit was poor, and the overall factor structure, factor loadings, and intercepts varied across groups. Intergroup gender invariance illustrated heterogeneity in attitudes towards online learning favoring men teachers and boy students. Study strengths and implications for the promotion of a positive experience of online learning are discussed.

Exploring the relationship between process data and contextual variables among Scandinavian students on PISA 2012 mathematics tasks

Given the ongoing development of computer-based tasks, there has been increasing interest in modelling students’ behaviour indicators from log file data with contextual variables collected via questionnaires. In this work, we apply a latent regression model to analyse the relationship between latent constructs (i.e., performance, speed, and exploration behaviour) and contextual variables among Scandinavian students (i.e., Norway, Sweden, and Denmark) during their completion of three interactive mathematics tasks in the 2012 Program for International Student Assessment (PISA). The purpose of this study is threefold: (1) to investigate whether a factor model is adequate for explaining the relationship between extracted process indicators from log files and students’ performance on the three items; (2) to evaluate whether measurement invariance holds for the three analysed countries; and (3) to determine how well contextual variables [i.e., a student’s background, availability, use of information and communication technology (ICT) resources, and learning conditions for mathematical literacy] correlate with the model’s latent constructs. Our findings reveal that a three-factor CFA model is well-suited to the analysed data and that a weak measurement invariance model performs best. By including contextual variables in the modelling, we also highlight the differences in Scandinavian students’ profiles. Specifically, higher economic social cultural status (ESCS) consistently led to higher math proficiency across all three countries. Norway did not show any gender differences in math proficiency, response time, or exploration behaviours. Experience with math tasks at school led to more exploration behaviours across all three countries. Swedish and Norwegian students who had more experience with pure math tasks at school were likely to obtain higher mathematics achievement scores and tended to interact more frequently with the testing platform when answering the three math items. When Danish students had higher ICT scores (i.e., more ICT resources available at school and home), they had lower response accuracy. There was, however, no effect of ICT on the three factors among Swedish and Norwegian students. Finally, we also discuss the implications and limitations of this study.

Validação de Construto da Versão Brasileira da Dyadic Adjustment Scale

Abstract This study sought evidence of construct validity for the Brazilian versions of the Dyadic Adjustment Scale and Revised Dyadic Adjustment Scale. A total of 448 individuals participated in the research, 253 women and 195 men from several regions of Brazil who had been cohabiting with their marital partners for an average of 14.7 years. Several proposed models for the measure were tested in Structural Equation Modeling. In the Confirmatory Factor Analysis, the four-factor and hierarchical models of the Revised Dyadic Adjustment Scale showed good overall adjustments. Evidence of factor, convergent, and discriminant validity were also found. Composite reliability revealed adequate levels of internal consistency. The Multigroup Confirmatory Factor Analysis demonstrated a strong measurement invariance model for men and women.

Measurement equivalence of the NACC UDS neuropsychological battery across race/ethnic groups

AbstractBackgroundThe National Alzheimer’s Coordinating Center (NACC) maintains a national database of clinical and cognitive data collected under a standard protocol set by the National Institute on Aging (NIA) called the Uniform Data Set (UDS). Development of factor scores unbiased by race/ethnicity will allow for comparison of other lifestyle and psychosocial factors within the NIA health disparities framework. The present study examined the dimensional structure of the comprehensive neuropsychological test battery and factorial equivalence across racial/ethnic groups.Method43,999 participants (Table 1; Age: Mean = 71.6, Education: Mean = 15.8; Latinx: N = 3549; Asian: N = 1141; Black: N = 5989; White: N = 32832) were assessed at baseline and included in the present study. Clinical status varied from cognitively normal to dementia. The sample was divided into learning and cross‐validation samples; the best fitting model was developed in the learning sample and tested in the cross‐validation sample. Multiple group confirmatory factor analysis (CFA) was used to evaluate measurement equivalence across race/ethnic groups. Measurement invariance models developed in the learning sample were tested in the cross‐validation sample.ResultResults from the full sample identified a best fitting four‐factor model (Figure 1; episodic memory, attention, executive function, language). Multiple group analyses confirmed the presence of metric invariance (Table 2; invariant factor loadings). Results supported partial scalar invariance—all domains had invariant intercepts but also non‐invariant intercepts. Most memory factor indicators had invariant intercepts (5/7). Attention and language had more non‐invariant intercepts. Category fluency, which loaded onto language and executive function factors, was particularly non‐invariant.ConclusionResults showed that the UDS neuropsychological battery measures the same four factors across race/ethnic groups. Crucially, there are important differences in measurement parameters across race/ethnic groups that must be accounted for in studies using these neuropsychological tests as outcomes. The four‐factor model effectively summarizes the comprehensive neuropsychological test battery and could provide a useful tool for studies of cognitive aging and Alzheimer’s disease and related dementias. Importantly, these cognitive factor scores can be used to clarify the influence of lifestyle and psychosocial factors on cognition across these groups.

Leading indicators of readiness among the general Army and Special Operations Forces: Predictive and psychometric analysis of the Global Assessment Tool

ABSTRACT Self-assessments are commonly used to track Army readiness in specialized communities, but they are rarely analyzed for reliability and predictive validity. Before introducing new assessments, existing ones should be reevaluated. We examined the Global Assessment Tool (GAT), an annual Army-required self-assessment with multiple psychosocial and health behavior short scales. Psychometric analyses on nine scales included item response theory (IRT) and measurement invariance models across total Army (n = 743,057) and special operations forces (SOF; n = 3,478) cohorts. Predictive analyses examined demographic-adjusted associations between GAT scales and one-year incident medical non-readiness (MNR). Most scales had adequate reliability, although some exhibited highly skewed distributions, which likely increased measurement error. Most scales exhibited metric and scalar measurement equivalence across total Army and SOF groups. Scores from scales measuring positive characteristics were associated with lower odds of MNR (good coping, flexibility, optimism, positive affect, work engagement, friendship, organization trust; adjusted odds ratios ≤ 0.75); scores from scales measuring negative characteristics were associated with increased odds of MNR (poor sleep, depression, negative affect, loneliness; adjusted odds ratios ≥ 1.4). Associations were similar across Army and SOF cohorts. In conclusion, self-report data can potentially contribute to command surveillance, but iterative quality-checks are necessary after deployment.

Meta-analytic tests of measurement invariance of internalizing and externalizing psychopathology across common methodological characteristics.

Factor analytic models of common mental disorders have been hypothesized to be affected by various methodological features, which could undermine the assumption that Internalizing and Externalizing reflect part of the natural structure of psychopathology. In this study, we addressed this issue by testing whether and how methodological features affect the empirical structure of psychopathology using meta-analytic measurement invariance models of Internalizing and Externalizing across multiple sample characteristics. Published studies estimating factor analytic models from Diagnostic and Statistical Manual of Mental Disorders (DSM) diagnoses were screened. A total of 47 samples (N = 118,966) were included in the meta-analysis. Data were pooled into meta-analytic correlation matrices using random effects models that accounted for sampling variability. Multigroup confirmatory factor analytic models of Internalizing and Externalizing were fit to the pooled matrix of DSM diagnoses to test invariance of the structure, factor loadings, and factor covariance. Results supported partial or full invariance across (a) samples selected versus not selected for psychopathology, (b) diagnoses defined by binary DSM thresholds versus dimensional symptoms counts, and (c) diagnoses based on lifetime versus current symptoms. Tentative analyses indicated noninvariance across samples that made diagnoses using hierarchical exclusion rules versus those that did not. Our study suggests that the Internalizing and Externalizing structure is largely robust to common methodological characteristics thought to impact factor analytic models. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Examining the Invariance of a Measurement Model by Using the Covariance Structure Approach

The primary aim of the present study is to examine the measurement invariance of the structural equating model constructed on the numerical and verbal abilities test for sixth grade students across gender, amount of weekly pocket money and students’ perceptions of the sufficiency of their pocket money. The secondary aim is to illustrate the use of the IBM AMOS-24 software package step by step with examples to address invariance using the covariance structural analysis approach. The research data were collected from 2304 sixth grade students enrolled in public schools within the Keçiören and Pursaklar suburbs in Ankara. The covariance structures analysis approach was employed during the examination of the measurement model invariance. The study revealed that invariance was achieved in terms of configural, measurement (in both measurement weights and measurement residuals) and structural invariance with respect to all subgroups.