Full Scalar Invariance Research Articles

The Knowledge-about-Older-Patients - Quiz (KOP-Q) for nurses: Cross-cultural validation between the Netherlands and United States of America

BackgroundThe Knowledge about Older Patients-Quiz (KOP-Q) is designed as a unidimensional scale measuring knowledge of hospital nurses about older patients. Furthermore, the KOP-Q measures a second unidimensional construct, certainty of hospital nurses about their knowledge. The KOP-Q is developed and validated in the Netherlands. Whether the KOP-Q can be used in other countries is unknown given the cultural and language differences. ObjectivesInvestigate the level of measurement invariance of the KOP-Q between the Netherlands and United States of America (USA). DesignA multicenter international cross-sectional design. SettingsFour general hospitals in the Netherlands and four general hospitals in the USA. ParticipantsNurses from the Netherlands (n=201) and the USA (n=130) were invited to participate by email from the ward manager, distributing flyers and present messages on the online hospital communication boards. Questions of the KOP-Q were completed online. MethodThe level of measurement invariance (configural, metric or scalar invariance) across countries was tested by running increasingly constrained structural equation models, and testing whether these models fitted the data. ResultsBoth the knowledge and certainty construct of the KOP-Q proved unidimensional in the Netherlands and USA sample. Test results of the measurement invariance across the Netherlands and USA indicated a stable, partial scalar invariance (15 items full scalar invariance) for the knowledge items and full scalar invariance for the certainty items. ConclusionsThe KOP-Q shows to function uniformly across both language groups and can therefore be used to assess nurses' knowledge and their certainty about this knowledge which can be important for educational and/or quality improvement programs in the USA. Furthermore, the KOP-Q is suitable to make comparisons between the Netherlands and the USA using latent variable models. Before the KOP-Q can be used in other countries, cross-cultural tests should again be performed.

Brand Engagement in Self-Concept: Scale Properties and the Global Nature of the Construct

The aim of the two presented studies was to attest whether the brand engagement in self-concept (BESC) has a global nature and to further evaluate the psychometric properties of the BESC Scale. In the first study (N = 997), we tested the psychometric properties of the Polish version of the BESC Scale and confirmed its factorial structure using confirmatory factor analysis. The BESC scale has good internal consistency (α = .95), construct validity, and demonstrate full scalar invariance across gender. The second study (N = 287) provided evidence that BESC has a global nature and that it is significantly related to the assessment of self–brand distance for brands that belong to various product categories and are important to consumers. Overall, current results provide psychometric support for the BESC Scale as an appropriate measure of global propensity to include important brands as part of the self-concept.

Observers' reports of self-regulation: Measurement invariance across sex, low-income status, and race/ethnicity

Few assessments of self-regulation for preschool-aged children have been validated for use in socioeconomically and racially/ethnically diverse populations. We address this gap by exploring the validity of the Preschool Self-Regulation Assessment – Assessor Report (PSRA-AR), an observer report of children's global self-regulation ability, using data from a population-based sample drawn from a large and diverse school district. We found that a three factor structure - representing attention regulation, impulse control, and positive engagement - fit the data well for the entire sample. Then, using multigroup confirmatory factor analyses, we found full scalar invariance across sex and low-income status. Scalar invariance held for all but one item across Black, White, and Hispanic children. These findings suggest that the PSRA-AR is valid for use with children of diverse socioeconomic and racial/ethnic backgrounds. Moreover, the PSRA-AR was moderately correlated with direct assessments of executive function and academic ability, including early math and reading.

Cross‐culture and gender invariance of the Warr (1990) job‐related well‐being measure

The job‐related affective well‐being measure developed by Warr (1990, J. Occup. Psychol., 63, 193) has been a frequently used tool estimating affect in the work context. The main purpose of this study was to test the measurement invariance of the instrument across cultures and genders. The model with four correlated factors, representing anxiety, comfort, depression, and enthusiasm, had a superior fit compared to alternative models in a sample of 807 employees and was used as the basis of measurement invariance tests. The results of multigroup confirmatory factor analysis showed partial metric invariance across samples of employees from the Netherlands (n = 254), Poland (n = 436), and Spain (n = 207), as well as full scalar invariance across genders (female n = 323; male n = 466). This indicates that cross‐cultural comparisons of relationships between job‐related affect as measured with this instrument and other constructs can be meaningful. Additionally, mean scores on the scales of the instrument can be meaningfully compared across genders, but not across countries.Practitioner points Warr's job‐related affective well‐being measure is a valid and reliable operationalization of four correlated affective dimensions: anxiety, comfort, depression, and enthusiasm. Mean scores on these dimensions can be meaningfully compared across genders, but comparisons of mean scores between countries should be interpreted with caution.

The Factor Structure and Measurement Invariance of Positive and Negative Affect

Abstract. This study investigated the factor structure and measurement invariance of the Mroczek and Kolarz’s scales of positive and negative affect in Iran (N = 2,391) and the USA (N = 2,154), and across gender groups. The two-factor model of affect was supported across the groups. The results of measurement invariance testing confirmed full metric and partial scalar invariance of the scales across cultural groups, and full metric and full scalar invariance across gender groups. The results of latent mean analysis revealed that Iranians scored lower on positive affect and higher on negative affect than Americans. The analyses also showed that American men scored significantly lower than American women on negative affect. The significance and implications of the results are discussed.

Factorial invariance of the Satisfaction with Life Scale in adolescents from Spain and Portugal.

The Satisfaction with Life Scale is one of the most widely used scales to measure the global cognitive judgment of satisfaction with one’s life. This study assesses the equivalence of the SWLS across Spanish and Portuguese adolescents, using multi-sample Confirmatory Factor Analysis. Participants were Spanish (N = 2183) and Portuguese (N = 4082) junior high school. The results provide high support for the internal consistency of both the Spanish and Portuguese versions of this scale. The results also showed that factor structure, factor loadings, could be considered invariant across groups. However, the full scalar invariance between Spanish and Portuguese samples was not found, with the intercept for SWLS item 5 varying across countries. Similar findings have also been found in other cross-national studies with this scale. Implications of the findings are discussed and we conclude that the Spanish and Portuguese versions of the SWLS can be used for cross-national comparisons with Spanish and Portuguese adolescents.

Establishing the Measurement Invariance of the Very Short Form of the Infant Behavior Questionnaire Revised for Mothers Who Vary on Race and Poverty Status

ABSTRACTWe examined the measurement invariance of the Infant Behavior Questionnaire Revised–Very Short Form (IBQR–VSF; Putnam, Helbig, Gartstein, Rothbart, & Leerkes, 2014) in a sample of 470 racially (185 White, 285 African American) and socioeconomically diverse mothers (158 below federal poverty threshold, 296 above federal poverty threshold) of infants. Using multigroup confirmatory factor analysis, we demonstrated configural, full metric, and full scalar invariance demonstrating that the 3-factor structure (negative emotionality, positive affectivity/surgency, orienting/regulatory capacity), pattern of item loadings, and item means were comparable for White and African American mothers, and for poor and not poor mothers. In addition, we demonstrated full error invariance across racial groups and partial error variance invariance across poverty status, demonstrating that item reliability was comparable for White and African American mothers, and both those above and below the poverty line (with the exception of a subset of items). Thus, the IBQR–VSF appears appropriate for use in racially and socioeconomically diverse samples.

The Value of Exploratory Structural Equation Modeling in Identifying Factor Overlap in the Mental Health Continuum-Short Form (MHC-SF): A Study with a New Zealand Sample

This study is the first to investigate the factor structure of the Mental Health Continuum-Short Form (MHC-SF) in New Zealand. Towards this end, traditional Confirmatory Factor Analysis (CFA) and the new method of Exploratory Structural Equation Modeling (ESEM) were used. Both ESEM and CFA supported the tripartite model of mental well-being in comparison to the one- and two-factor models; however, ESEM provided better fit with the data. Moreover, interfactor correlations were considerably lower in ESEM than they were in CFA, indicating greater factor distinctiveness in ESEM. ESEM also revealed a number of important cross-loadings for items in the measurement model of the MHC-SF. The results supported full metric and full scalar invariance of the MHC-SF across gender. The attenuated correlations among well-being factors obtained by ESEM here provide an important insight about the ongoing controversy regarding the failure of empirical research to identify distinct eudaimonic and hedonic factors in well-being measures. An overreliance on CFA methods may have led the field to rely on inflated estimates of shared variance between eudaimonia and hedonia.

Comparing future teachers’ beliefs across countries: approximate measurement invariance with Bayesian elastic constraints for local item dependence and differential item functioning

Using data from the international Teacher Education and Development Study: Learning to Teach Mathematics (TEDS-M), the measurement equivalence of teachers’ beliefs across countries is investigated for the case of ‘mathematics-as-a fixed-ability’. Measurement equivalence is a crucial topic in all international large-scale assessments and widely undervalued for non-achievement measures. Although it is well known that cultural differences may affect response styles or meaning of constructs, and although comparability is often not specifically examined, international large-scale assessment results are typically compared across countries in terms of scale means. Full scalar invariance is typically unattainable in such international assessments due to their sheer size and local item dependence issues. To deal with both challenges, a Bayesian approach is introduced that uses informative priors to define elastic equivalence constraints to investigate the comparability of TEDS-M future teachers’ beliefs under a working assumption of approximate measurement invariance. Substantively, a clear pattern of three groups of countries emerged that corroborates preliminary findings in other studies for other teacher beliefs. Future mathematics teachers in Thailand, Malaysia and the Philippines believed more strongly that mathematics is a fixed ability, whereas the USA, Chile, Norway, Germany and Switzerland believed less strongly in this. Taiwan, Singapore, Poland and Russia took the middle ground.

Psychometric Properties of the Mental Health Continuum-Short Form: A Study of Portuguese Speaking Children/Youths

The development of measures of children and adolescents’ subjective well-being is crucial to the conceptualization and evaluation of positive mental health. The aim of this study was to investigate the factorial structure and invariance of the Mental Health Continuum-Short Form (MHC-SF—adolescents) across children and youths. Participants were from two different samples: 208 elementary school children (sample 1) and 216 middle school youths (sample 2). Results confirmed the three-dimensional structure of subjective well-being in both samples. The three sub-scales of the MHCSF yielded high internal consistency and results from the HTMT85 indicated discriminant validity. Measurement invariance testing across three different age groups (7–8, 9–10 and 11–14 years) confirmed the full metric and approximate scalar invariance of the MHC-SF. Full scalar invariance was achieved across gender. The study also compared the latent means for mental well-being in the three age groups, and found that the younger groups showed significantly higher levels of wellbeing. The present research study strongly suggests that the MHC-SF (adolescents) is an appropriate instrument to measure the positive mental health and well-being of children and pre-adolescents as multidimensional construct.

Messäquivalenz der Allgemeinen Depressionsskala (ADS 20) bei Menschen mit und ohne Migrationshintergrund unter geschlechtsspezifischer Perspektive

The measurement invariance of the German version of the Center of Epidemiological Studies Depression scale (CES-D 20) among men and women with and without migration background has not been studied yet. A total of 2786 employed persons without (female n = 1772, male n = 598) and with an immigrant background (female n = 301, male n = 115) were included in the study. The multi-group confirmatory factor analysis was used to test for the strong factorial invariance and to estimate latent mean differences. The factorial structure is invariat 1) across the female sub-sample, full scalar invariance is achieved, 2) compared to male non-migrants and female sub-sample, partial scalar invariance is achieved. 3) The male migrants show a different, three-factor structure with a mix-factor of depressive affect/somatic complaints. 4) Female migrants have significantly higher latent mean values in subscales. The measurement invariance of the German CES-D 20 is given to that extent that meaningful comparisons in correlative coefficient and latent mean values can be made.

Analyses of measurement equivalence across gender in the Mentoring Functions Questionnaire (MFQ-9)

The purpose of this study is to examine the measurement equivalence/invariance (ME/I) of the 9-item Mentoring Functions Questionnaire (MFQ-9) across gender. Although ME/I is a prerequisite for examining cross-group differences, this assumption is rarely examined in mentoring research, particularly for comparing gender differences in mentoring functions protégés receive. Following Vandenberg and Lance’s (2000) suggestion, a series of multi-group confirmatory factor analyses (MGCFAs) were conducted to investigate ME/I of the MFQ-9 across male ( N = 201) and female ( N = 312) protégés. The results supported full configural invariance, full metric invariance, full scalar invariance, partial uniqueness invariance and partial factor variance–covariance matrix invariance across gender. Implications and application of the study findings are discussed.