			Home About us Contact

Measurement Invariance (measurement + invariance)

Distribution by Scientific Domains

Psychology	46%
Medical Sciences	38%

Selected Abstracts

Testing Measurement Invariance Using Item Response Theory in Longitudinal Data: An Introduction

CHILD DEVELOPMENT PERSPECTIVES, Issue 1 2010
Roger E. Millsap
Abstract, Item response theory (IRT) consists of a set of mathematical models for the probabilities of various responses to test items as a function of item and person characteristics. In longitudinal data, changes in measured variables can only be interpreted if important psychometric features of the measured variables are assumed invariant across time. Measurement invariance is invariance in the relation of a measure to the latent variable underlying it. Measurement invariance in longitudinal studies concerns invariance over time, and IRT provides a useful approach to investigating longitudinal measurement invariance. Commonly used IRT models are described, along with the representation of measurement invariance in IRT. The use of IRT for investigating invariance is then described, along with practical considerations in using IRT for this purpose. Conceptual issues, rather than technical details, are emphasized throughout. [source]

Comparing Personality Scales Across Time: An Illustrative Study of Validity and Consistency in Life-Span Archival Data

JOURNAL OF PERSONALITY, Issue 1 2000
Leslie R. Martin
The goals of this study were: (a) to examine whether personality scales, meaningful in contemporary terms, could be derived from archival data; and (b) to use these scales to aid our understanding of the relation of personality to mortality. NEO PI-R data and a battery of archival items, taken from Terman's Life Cycle Study, were collected on two new samples (sample 1 mean age = 11.9, n = 167; sample 2 mean age = 22.2, n = 203). Measurement invariance of the archival scales was assessed, and validity was examined using both rational analyses and associations with the Five Factor Model. It was demonstrated that interpretable scales can be derived from 50- to 70-year-old archival data. The archival adult personality data were then used to predict mortality. Conscientiousness remains the strongest personality predictor of longevity. Criteria for establishing the validity of archivally derived scales are suggested. [source]

Testing Measurement Invariance Using Item Response Theory in Longitudinal Data: An Introduction

Factor structure of a conceptual model of oral health tested among 65-year olds in Norway and Sweden

COMMUNITY DENTISTRY AND ORAL EPIDEMIOLOGY, Issue 2 2010
Anne Nordrehaug Åstrøm
Åstrøm AN, Ekbäck G, Ordell S. Factor structure of a conceptual model of oral health tested among 65-year olds in Norway and Sweden. Community Dent Oral Epidemiol 2010. © 2010 John Wiley & Sons A/S Abstract,,, Background:, No studies have tested oral health-related quality of life models in dentate older adults across different populations. Objectives:, To test the factor structure of oral health outcomes within Gilbert's conceptual model among 65-year olds in Sweden and Norway. It was hypothesized that responses to 14 observed indicators could be explained by three correlated factors, symptom status, functional limitations and oral disadvantages, that each observed oral health indicator would associate more strongly with the factor it is supposed to measure than with competing factors and that the proposed 3-factor structure would possess satisfactory cross-national stability with 65-year olds in Norway and Sweden. Methods:, In 2007, 6078 Swedish- and 4062 Norwegian adults borne in 1942 completed mailed questionnaires including oral symptoms, functional limitations and the eight item Oral Impacts on Daily Performances inventory. Results:, Model generation analysis was restricted to the Norwegian study group and the model achieved was tested without modifications in Swedish 65-year olds. A modified 3-factor solution with cross-loadings, improved the fit to the data compared with a 2-factor- and the initially proposed 3-factor model among the Norwegian [comparative fit index (CFI) = 0.97] and Swedish (CFI = 0.98) participants. All factor loadings for the modified 3-factor model were in the expected direction and were statistically significant at CR > 1. Multiple group confirmatory factor analyses, with Norwegian and Swedish data simultaneously revealed acceptable fit for the unconstrained model (CFI = 0.97), whereas unconstrained and constrained models were statistically significant different in nested model comparison. Conclusions:, Within construct validity of Gilbert's model was supported with Norwegian and Swedish 65-year olds, indicating that the 14-item questionnaire reflected three constructs; symptom status, functional limitation and oral disadvantage. Measurement invariance was confirmed at the level of factor structure, suggesting that the 3-factor model is comparable to some extent across 65-year olds in Norway and Sweden. [source]

Factorial structure and cross-cultural invariance of the Oral Impacts on Daily Performances

EUROPEAN JOURNAL OF ORAL SCIENCES, Issue 3 2009
A. N. Åstrøm
The issue of cross-cultural construct validation and measurement invariance of the Oral Impacts on Daily Performances (OIDP) questionnaire is important. Using confirmatory factor analysis (CFA), this study evaluated a proposed three-factor structure of the OIDP questionnaire in Tanzanian adolescents and adults and assessed whether this model would be replicated in Ugandan adolescents. Between 2004 and 2007, OIDP data were collected from 1,601 Tanzanian adolescents, 1,031 Tanzanian adults, and 1,146 Ugandan adolescents. Model generation analysis was restricted to Tanzanian adolescents, and the model achieved was tested, without modification, in Tanzanian adults and in Ugandan adolescents. A modified three-factor solution with cross-loadings improved the fit of the OIDP model to the data compared with a one-factor model and the original three-factor model within the Tanzanian [comparative fit index (CFI) = 0.99] and Ugandan (CFI = 0.98) samples. Cross-validation in Tanzanian adults provided a reasonable fit (CFI = 0.98). Multiple-group CFA demonstrated acceptable fit [,2 = 140.829, degrees of freedom (d.f.) = 24, CFI = 0.98] for the unconstrained model, whereas unconstrained and constrained models were statistically significantly different. Factorial validity was confirmed for the three-factor OIDP model. The results provide evidence for cross-cultural equivalence of the OIDP, suggesting that this measure is comparable, at least to some extent, across Tanzanian and Ugandan adolescents. [source]

Construct validity and generalizability of the Carver,White behavioural inhibition system/behavioural activation system scales

EUROPEAN JOURNAL OF PERSONALITY, Issue 5 2001
Luigi Leone
The factorial structure and invariance of the BIS/BAS scales of Carver and White were assessed across three samples from the USA, UK, and Italy. Previous validation studies of the BIS/BAS scales relied on individual samples drawn from English-speaking populations only and failed to formally assess generalizability. The current study shows that the four-factor structure proposed by Carver and White,i.e. one BIS and three BAS facets,achieved satisfactory psychometric properties in all three samples and that measurement invariance was obtained across countries. Latent mean differences due to gender and country were also investigated. Theoretical issues concerning the validity of the BIS/BAS scales are addressed. Copyright © 2001 John Wiley & Sons, Ltd. [source]

Ascertaining late-life depressive symptoms in Europe: an evaluation of the survey version of the EURO-D scale in 10 nations.

INTERNATIONAL JOURNAL OF METHODS IN PSYCHIATRIC RESEARCH, Issue 1 2008
The SHARE project
Abstract The reported prevalence of late-life depressive symptoms varies widely between studies, a finding that might be attributed to cultural as well as methodological factors. The EURO-D scale was developed to allow valid comparison of prevalence and risk associations between European countries. This study used Confirmatory Factor Analysis (CFA) and Rasch models to assess whether the goal of measurement invariance had been achieved; using EURO-D scale data collected in 10 European countries as part of the Survey of Health, Ageing and Retirement in Europe (SHARE) (n = 22,777). The results suggested a two-factor solution (Affective Suffering and Motivation) after Principal Component Analysis (PCA) in 9 of the 10 countries. With CFA, in all countries, the two-factor solution had better overall goodness-of-fit than the one-factor solution. However, only the Affective Suffering subscale was equivalent across countries, while the Motivation subscale was not. The Rasch model indicated that the EURO-D was a hierarchical scale. While the calibration pattern was similar across countries, between countries agreement in item calibrations was stronger for the items loading on the affective suffering than the motivation factor. In conclusion, there is evidence to support the EURO-D as either a uni-dimensional or bi-dimensional scale measure of depressive symptoms in late-life across European countries. The Affective Suffering sub-component had more robust cross-cultural validity than the Motivation sub-component. Copyright © 2008 John Wiley & Sons, Ltd. [source]

An Odds Ratio Approach for Assessing Differential Distractor Functioning Effects under the Nominal Response Model

JOURNAL OF EDUCATIONAL MEASUREMENT, Issue 3 2008
Randall D. Penfield
Investigations of differential distractor functioning (DDF) can provide valuable information concerning the location and possible causes of measurement invariance within a multiple-choice item. In this article, I propose an odds ratio estimator of the DDF effect as modeled under the nominal response model. In addition, I propose a simultaneous distractor-level (SDL) test of invariance based on the results of the distractor-level tests of DDF. The results of a simulation study indicated that the DDF effect estimator maintained good statistical properties under a variety of conditions, and the SDL test displayed substantially higher power than the traditional Mantel-Haenszel test of no DIF when the DDF effect varied in magnitude and/or size across the distractors. [source]

Perspectives of fathers and mothers of children in early intervention programmes in assessing family quality of life

JOURNAL OF INTELLECTUAL DISABILITY RESEARCH, Issue 12 2006
M. Wang
Abstract Background Family quality of life (FQOL), as a family outcome measure of early intervention and other services, has increasingly drawn attention of researchers, policymakers and service providers. Developing an index of family QOL requires a measure suitable for use with multiple family members. The purpose of this study was to test whether mothers and fathers similarly view the conceptual model of FQOL embodied in one measure. Method This study involved fathers and mothers of 107 families who have a young child (birth to five) with a disability enrolled in an early intervention programme. Data from couples completing the Beach Center FQOL measure were analysed using structural equation modelling (SEM) to determine similarities or differences between fathers and mothers with respect to their assessment of FQOL. Results The analysis of measurement invariance of the FQOL construct across the father and mother groups indicates that the Beach Center FQOL Scale measures equally the underlying FQOL construct across fathers and mothers in this sample. Fathers do not differ from mothers in perceived importance of factors related to FQOL items, nor did they differ in their overall satisfaction with FQOL. Conclusion These results suggest that fathers and mothers respond similarly to the latent constructs within the Beach Center FQOL Scale; therefore, it holds promise for use with both fathers and mothers in assessing FQOL across multiple family members. Further implications for research and practice are discussed. [source]

Factorial validity of the Reynolds Intellectual Assessment Scales for referred students

PSYCHOLOGY IN THE SCHOOLS, Issue 10 2009
A. Alexander Beaujean
The Reynolds Intellectual Assessment Scales (RIAS) is a recently developed, individually administered psychometric instrument designed to measure general cognitive ability, as well as verbal (crystallized) intelligence, nonverbal (fluid) intelligence, and memory. Test reviewers have recommended the use of the RIAS despite the fact that, although the RIAS has been in circulation for more than four years, there is a paucity of independent research published about its psychometric properties. The purpose of the current study was to examine the factor structure of the RIAS across three samples of school-age children: the RIAS norming sample; the data reported by Nelson, Canivez, Lindstrom, and Hatt; and a new, independent sample of students referred for special education services. Using confirmatory factor analytic techniques, this study found that a two-factor model, positing verbal and nonverbal factors, fit all three data sets better than a one-factor model. Furthermore, the two-factor model demonstrated partial measurement invariance across the three samples, although the verbal factor showed much stronger invariance, construct reliability, and overall interpretability than did the nonverbal factor. Implications of this study for practitioners are discussed. © 2009 Wiley Periodicals, Inc. [source]

Psychometric properties of the Swedish PedsQL, Pediatric Quality of Life Inventory 4.0 generic core scales

ACTA PAEDIATRICA, Issue 9 2009
Solveig Petersen
Abstract Aim:, To study the psychometric performance of the Swedish version of the Pediatric Quality of Life Inventory (PedsQL) 4.0 generic core scales in a general child population in Sweden. Methods:, PedsQL forms were distributed to 2403 schoolchildren and 888 parents in two different school settings. Reliability and validity was studied for self-reports and proxy reports, full forms and short forms. Confirmatory factor analysis tested the factor structure and multigroup confirmatory factor analysis tested measurement invariance between boys and girls. Results:, Test-retest reliability was demonstrated for all scales and internal consistency reliability was shown with , value exceeding 0.70 for all scales but one (self-report short form: social functioning). Child-parent agreement was low to moderate. The four-factor structure of the PedsQL and factorial invariance across sex subgroups were confirmed for the self-report forms and for the proxy short form, while model fit indices suggested improvement of several proxy full-form scales. Conclusion:, The Swedish PedsQL 4.0 generic core scales are a reliable and valid tool for health-related quality of life (HRQoL) assessment in Swedish child populations. The proxy full form, however, should be used with caution. The study also support continued use of the PedsQL as a four-factor model, capable of revealing meaningful HRQoL differences between boys and girls. [source]