scholarly journals Dynamic Fit Index Cutoffs for One-Factor Models

2021 ◽  
Author(s):  
Melissa Gordon Wolf ◽  
Daniel McNeish
Keyword(s):  
Approximate Model ◽  
Factor Models ◽  
Model Fit ◽  
Fit Indices ◽  
Analysis Framework ◽  
Excellent Performance ◽  
Model Fit Indices ◽  
Simulation Results ◽  
Fit Index ◽  
Continuous Measures

Assessing unidimensionality of a scale is a frequent interest in behavioral research. Often, this is done with approximate model fit indices in a factor analysis framework such as RMSEA, CFI, or SRMR. These fit indices are continuous measures, so values indicating acceptable fit are up to interpretation. Cutoffs suggested by Hu and Bentler (1999) are a common guideline used in empirical research. However, these cutoffs were derived with intent to detect omitted cross-loadings or omitted factor covariances in three-factor models. These types of misspecifications cannot exist in one-factor models, so the appropriateness of using these guidelines in one-factor models is uncertain. This paper uses a simulation study to address whether traditional fit index cutoffs are sensitive to the types of misspecifications that can occur in one-factor models. The results showed that traditional cutoffs have very poor sensitivity to misspecification in one-factor models and that the traditional cutoffs generalize poorly to one-factor contexts. As an alternative, we investigate the accuracy and stability of the recently introduced dynamic fit cutoff approach for creating fit index cutoffs for one-factor models. Simulation results indicated excellent performance of dynamic fit index cutoffs to classify correct or misspecified one-factor models and that dynamic fit index cutoffs are a promising approach for more accurate assessment of unidimensionality.

scholarly journals The Eating Motivation Survey: results from the USA, India and Germany

2017 ◽  
Vol 21 (03) ◽  
pp. 515-525 ◽  
Author(s):  
Gudrun Sproesser ◽  
Matthew B Ruby ◽  
Naomi Arbit ◽  
Paul Rozin ◽  
Harald T Schupp ◽  
...  
Keyword(s):  
Factor Structure ◽  
Online Survey ◽  
Total Sample ◽  
Model Fit ◽  
Fit Indices ◽  
Confirmatory Factor Analyses ◽  
Model Fit Indices ◽  
The Usa ◽  
Survey Results ◽  
Fit Index

Abstract Objective Research has shown that there is a large variety of different motives underlying why people eat what they eat, which can be assessed with The Eating Motivation Survey (TEMS). The present study investigates the consistency and measurement invariance of the fifteen basic motives included in TEMS in countries with greatly differing eating environments. Design The fifteen-factor structure of TEMS (brief version: forty-six items) was tested in confirmatory factor analyses. Setting An online survey was conducted. Subjects US-American, Indian and German adults (total N 749) took part. Results Despite the complexity of the model, fit indices indicated a reasonable model fit (for the total sample: χ 2/df=4·03; standardized root-mean-squared residual (SRMR)=0·063; root-mean-square error of approximation (RMSEA)=0·064 (95 % CI 0·062, 0·066)). Only the comparative fit index (CFI) was below the recommended threshold (for the total sample: CFI=0·84). Altogether, 181 out of 184 item loadings were above the recommended threshold of 0·30. Furthermore, the factorial structure of TEMS was invariant across countries with respect to factor configuration and factor loadings (configural v. metric invariance model: ΔCFI=0·009; ΔRMSEA=0·001; ΔSRMR=0·001). Moreover, forty-three out of forty-six items showed invariant intercepts across countries. Conclusions The fifteen-factor structure of TEMS was, in general, confirmed across countries despite marked differences in eating environments. Moreover, latent means of fourteen out of fifteen motive factors can be compared across countries in future studies. This is a first step towards determining generalizability of the fifteen basic eating motives of TEMS across eating environments.

scholarly journals Examining Measurement invariance for Parental Creativity Nurturing Behaviour (PCNB) across middle-East Nations and South-east Asian Nations

2021 ◽  
Author(s):  
Ekta Sharma ◽  
Sandeep Sharma ◽  
Muhammad Al-Kudah ◽  
Angham Al-Tamimi
Keyword(s):  
Middle East ◽  
Measurement Invariance ◽  
East Asian ◽  
Factor Loading ◽  
Model Fit ◽  
Fit Indices ◽  
Chi Square ◽  
Model Fit Indices ◽  
The One ◽  
Fit Index

Abstract Background and Objectives Examining the Usefulness of the Psychological instruments / scales /framework which are developed in one nation for other nations is vital step in establishing generalization for use across nations of Parental creativity nurturing behavior , Who are going to play most important role to develop Creative potential of children in 21st century.However The instrument used to measure the parental creativity nurturing behavior and the theoretical constructs of the subject exhibits good cross nation equivalence of PCNB scale.We reviewed various forms of measurement invariance for sample of parents , systematically, scientifically analyzed, and statistically proved the measurement invariance ,across gender, Age , culture ,Language of the PCNB scale by comparing MEN and SEAN . This would benefit the Parents in middle east Nations (MEN) and South-east Asian Nations(SEAN) . It will help parents to monitor their creativity nurturing behavior (CNB) and take appropriate steps to enrich PCNB and play important role to help their child lead a creativity accompanied life in 21st century.It will also Approve the wider use and acceptability of the PCNB scale.Research design and methodBased on data of 931 { 423 & 508 parents respectively from South East Asian Nations(SEAN ,Representing Nation -India) and Middle East Nations (MEN, representing nation Jordon)} ,we used PCNB Scale(Sharma & Sharma 2021) that focused on 4 major factors that are crucial to examine and assess validity and reliability of PCNB scale. Multisampling confirmatory factor analysis was done to identify the factors and then the model fit for the construct of PCNB. The selection of India as representative nation for Southeast Asian Nations and Jordon as representative nation of Middle east nations was done randomly as both are emerging educational nations in respective geographical areas in Asia.ResultsThe scale was found to be both valid and reliable with an excellent model fit with composite reliability (CR-SEAN) (CR factor1 =0.818 , CR factor2 =0.872 , CR factor3 =0.670 , CR factor4 =0.729), factor loading (>0.4 ) , Cronbach Alpha(SEAN)=0.712 and model fit indices SEAN (comparative fit index(CFI)=0.967,Chi-square/df=1.492, GFI=0.942,TLI=0.95,RFI=0.861,RMSEA=0.059,IFI=0.961,SRMR=0.051 ).This suggests four factors model ,supporting the one-dimensional reliable construct of PCNB at baseline.The scale was found to be both valid and reliable with an excellent model fit with composite reliability (CR-MEN) (CR factor1 =0.955 , CR factor2 =0.950 , CR factor3 =0.918 , CR factor4 =0.857), factor loading (>0.4 ) , Cronbach Alpha(MEN)=0.57 and model fit indices MEN (comparative fit index(CFI)=0.967,Chi-square/df=1.492, GFI=0.942,TLI=0.95,RFI=0.861,RMSEA=0.059,IFI=0.961,SRMR=0.038 ).This suggests four factors ,supporting the one-dimensional reliable construct of PCNB at baseline.Discussion and implicationsPCNB is a valid and reliable measure of creativity nurturing behavior of parents and thus enables a comprehensive evaluation of measure of creativity nurturing behavior in Parents. This scale would prove to be an important tool for assessment of parental creativity nurturing behavior and their parenting style. Once put to use on large-scale, would benefit and help the parents to identify their behavior and nurture creativity behavior in children well before they join the school, and would continue to nurture creativity even during the schooling and further. Basically this would help parents identify the need to get counseled (trained) formally or informally to nurture creativity in their child. SignificancePCNB Scale is a valid and reliable scale that would be globally useful for assessment of Parents creativity nurturing behavior (PCNB) across gender, countries, cultures, and age. This scale is available with strong psychometric properties and can prove useful for planning parent counseling programs to help them exhibit creativity nurturing behavior, and thus contribute to society by identifying parents to nurture creativity in children for finding creative solutions to different problems and challenges that world faces in the times to come. This could also help to develop new teaching methods and pedagogies in the direction of creativity nurturing. This research will add to Current sciences of Creativity nurturing behavior of parents.

scholarly journals A study on development of Perceived Quality-Image Scale in Sports Organizations

2018 ◽  
Vol 15 (4) ◽  
pp. 2407
Author(s):  
Yeşim Bayrakdaroglu ◽  
Dursun Katkat
Keyword(s):  
Reliability Analysis ◽  
Model Fit ◽  
Measurement Tool ◽  
Fit Indices ◽  
Confirmatory Factor Analyses ◽  
Sports Organizations ◽  
Confirmatory Factor ◽  
Quality Image ◽  
Model Fit Indices ◽  
Marketing Activities

The purpose of this study is to research how marketing activities of international sports organizations are performed and to develop a scale determining the effects of image management on public. The audiences of interuniversity World Winter Olympic sheld in Erzurum in 2011 participated in the research. Explanatory and Confirmatory Factor Analysis, reliability analysis were performed over the data obtained. All model fit indices of 25-item and four-factor structure of quality-image scale perceived in sports organizations applied were found to be at good level. In line with the findings obtained from the explanatory and confirmatory factor analyses and reliability analysis, it can be uttered that the scale is a valid and reliable measurement tool that can be used in field researches.

scholarly journals M117. ELUCIDATING ARCHITECTURE OF NEGATIVE SYNDROME IN SCHIZOPHRENIA

2020 ◽  
Vol 46 (Supplement_1) ◽  
pp. S179-S179
Author(s):  
Mei San Ang ◽  
Gurpreet Rekhi ◽  
Jimmy Lee
Keyword(s):  
Root Mean Square ◽  
Negative Symptom ◽  
Negative Symptoms ◽  
Factor Model ◽  
Five Factor Model ◽  
Factor Models ◽  
Second Order ◽  
Model Fit ◽  
Mean Square ◽  
Fit Indices

Abstract Background The conceptualization of negative symptoms has been refined in the past decades. Two-factor model comprising Motivation and Pleasure (MAP) and Emotional Expressivity (EE), five-factor model representing five domains of negative symptoms and second-order five-factor model incorporating the two-factor and five-factor models (Anhedonia, Asociality and Avolition regressed on MAP; Blunted Affect and Alogia regressed on EE) have been suggested as latent structure of negative symptoms. In most studies, the item “Lack of Normal Distress” in the Brief Negative Symptom Scale (BNSS) did not fit well in factor models. Nevertheless, the reported correlation and item-total correlation of Distress with other negative symptom domains and BNSS items were not negligible. Emotion deficit was also discussed as a part of negative symptoms conceptualization. As a single item may not be sufficient to represent an underlying construct that is potentially abstract and complex, the Schedule for the Deficit Syndrome (SDS) which comprises “Diminished Emotional Range” that is conceptually relevant to the BNSS Distress was employed. The study aimed to reexamine the conceptualization of negative symptoms by examining the model fit of several models when BNSS Distress and SDS Emotion (EMO) were included in the models using confirmatory factor analyses (CFA). Methods Two-hundred and seventy-four schizophrenia outpatients aged 21–65 were assessed on the BNSS and SDS. In the two-factor models, Restricted Affect, Diminished Emotional Range and Poverty of Speech in SDS and all items in BNSS Blunted Affect and Alogia subscales were regressed on EE, Curbing of Interests, Diminished Sense of Purpose and Diminished Social Drive in SDS and all items in BNSS Anhedonia, Asociality and Avolition subscales were regressed on MAP, without EMO, or with EMO regressed on either EE or MAP. Five-factor models and second-order five-factor models were examined, with or without EMO. Lastly, a six-factor model with EMO manifested by the sixth factor and second-order six-factor models in which EMO was regressed on either EE or MAP were tested. Root mean square error of approximation (RMSEA) <0.08, comparative fit index (CFI) >0.95, the Tucker-Lewis Index (TLI) >0.95, and weighted root-mean-square residual (WRMR) <1.0 indicate good model fit. CFAs were conducted using Mplus version 7.4. Results The two-factor models did not yield adequate fit indices. Five-factor model and second-order five-factor model without EMO had good model fit; five-factor model: RMSEA=0.056 (0.044–0.068), CFI=0.996, TFI=0.995, WRMR=0.718; second-order five-factor model: RMSEA=0.049 (0.036–0.061), CFI=0.997, TFI=0.996, WRMR=0.758. When EMO was included as indicator in one of the factors in the five-factor models, only the model in which EMO was regressed on Alogia yielded adequate fit. Similarly, in the second-order five-factor models, adequate fit indices were observed only when EMO was regressed on Alogia and Blunted Affect. The six-factor model fitted the data well, RMSEA=0.053 (0.042–0.064), CFI=0.996, TFI=0.995, WRMR=0.711. Second-order six-factor model with EMO regressed on EE yielded better model fit than MAP, RMSEA=0.050 (0.039–0.061), CFI=0.996, TFI=0.995, WRMR=0.849. Discussion In line with previous studies, five-factor and second-order five-factor models without EMO fitted the data well. When EMO was included, a six-factor model and a second-order six-factor model in which the sixth factor was regressed on EE showed good model fit. Emotion, motivation and behavior are intertwined. Our results showed that diminished emotion may also be one of the components of negative symptoms, which had higher association with EE than MAP.

scholarly journals Best Practices in the Use of Bifactor Models: Conceptual Grounds, Fit Indices and Complementary Indicators

2018 ◽  
Vol 18 (3) ◽  
Author(s):  
Pablo Ezequiel Flores-Kanter ◽  
Sergio Dominguez-Lara ◽  
Mario Alberto Trógolo ◽  
Leonardo Adrián Medrano
Keyword(s):  
Best Practices ◽  
Empirical Studies ◽  
Model Fit ◽  
Fit Indices ◽  
Research Article ◽  
Bifactor Models ◽  
Model Fit Indices ◽  
Personality Psychopathology ◽  
Using Data ◽  
Published Research

<p>Bifactor models have gained increasing popularity in the literature concerned with personality, psychopathology and assessment. Empirical studies using bifactor analysis generally judge the estimated model using SEM model fit indices, which may lead to erroneous interpretations and conclusions. To address this problem, several researchers have proposed multiple criteria to assess bifactor models, such as a) conceptual grounds, b) overall model fit indices, and c) specific bifactor model indicators. In this article, we provide a brief summary of these criteria. An example using data gathered from a recently published research article is also provided to show how taking into account all criteria, rather than solely SEM model fit indices, may prevent researchers from drawing wrong conclusions.</p>

Is It Worthy to Take Account of the “Guessing” in the Performance of the Raven Test? Calling for the Principle of Parsimony for Test Validation

2020 ◽  
pp. 073428292093092 ◽  
Author(s):  
Patrícia Silva Lúcio ◽  
Joachim Vandekerckhove ◽  
Guilherme V. Polanczyk ◽  
Hugo Cogo-Moreira
Keyword(s):  
Preschool Children ◽  
Item Response Theory ◽  
Item Response ◽  
Real Data ◽  
Model Fit ◽  
Test Validation ◽  
Fit Indices ◽  
Test Information ◽  
Model Fit Indices ◽  
Principle Of Parsimony

The present study compares the fit of two- and three-parameter logistic (2PL and 3PL) models of item response theory in the performance of preschool children on the Raven’s Colored Progressive Matrices. The test of Raven is widely used for evaluating nonverbal intelligence of factor g. Studies comparing models with real data are scarce on the literature and this is the first to compare models of two and three parameters for the test of Raven, evaluating the informational gain of considering guessing probability. Participants were 582 Brazilian’s preschool children ( Mage = 57 months; SD = 7 months; 46% female) who responded individually to the instrument. The model fit indices suggested that the 2PL fit better to the data. The difficulty and ability parameters were similar between the models, with almost perfect correlations. Differences were observed in terms of discrimination and test information. The principle of parsimony must be called for comparing models.

scholarly journals The Poor Fit of Model Fit for Selecting Number of Factors in Exploratory Factor Analysis for Scale Evaluation

2020 ◽  
pp. 001316442094289
Author(s):  
Amanda K. Montoya ◽  
Michael C. Edwards
Keyword(s):  
Factor Analysis ◽  
Root Mean Square ◽  
Exploratory Factor Analysis ◽  
Model Fit ◽  
Mean Square ◽  
Fit Indices ◽  
Scale Evaluation ◽  
Number Of Factors ◽  
Model Fit Indices ◽  
The Impact

Model fit indices are being increasingly recommended and used to select the number of factors in an exploratory factor analysis. Growing evidence suggests that the recommended cutoff values for common model fit indices are not appropriate for use in an exploratory factor analysis context. A particularly prominent problem in scale evaluation is the ubiquity of correlated residuals and imperfect model specification. Our research focuses on a scale evaluation context and the performance of four standard model fit indices: root mean square error of approximate (RMSEA), standardized root mean square residual (SRMR), comparative fit index (CFI), and Tucker–Lewis index (TLI), and two equivalence test-based model fit indices: RMSEAt and CFIt. We use Monte Carlo simulation to generate and analyze data based on a substantive example using the positive and negative affective schedule ( N = 1,000). We systematically vary the number and magnitude of correlated residuals as well as nonspecific misspecification, to evaluate the impact on model fit indices in fitting a two-factor exploratory factor analysis. Our results show that all fit indices, except SRMR, are overly sensitive to correlated residuals and nonspecific error, resulting in solutions that are overfactored. SRMR performed well, consistently selecting the correct number of factors; however, previous research suggests it does not perform well with categorical data. In general, we do not recommend using model fit indices to select number of factors in a scale evaluation framework.

The poor fit of model fit for selecting number of factors in exploratory factor analysis for scale evaluation

2020 ◽  
Author(s):  
Amanda Kay Montoya ◽  
Michael C. Edwards
Keyword(s):  
Factor Analysis ◽  
Exploratory Factor Analysis ◽  
Evaluation Framework ◽  
Model Fit ◽  
Model Specification ◽  
Fit Indices ◽  
Scale Evaluation ◽  
Number Of Factors ◽  
Model Fit Indices ◽  
The Impact

Model fit indices are being increasingly recommended and used to select the number of factors in an exploratory factor analysis. Growing evidence suggests that the recommended cutoff values for common model fit indices are not appropriate for use in an exploratory factor analysis context. A particularly prominent problem in scale evaluation is the ubiquity of correlated residuals and imperfect model specification. Our research focuses on a scale evaluation context and the performance of four standard model fit indices: root mean squared error of approximate (RMSEA), standardized root mean squared residual (SRMR), comparative fit index (CFI), and Tucker-Lewis index (TLI), and two equivalence test-based model fit indices: RMSEAt and CFIt. We use Monte Carlo simulation to generate and analyze data based on a substantive example using the positive and negative affective schedule (N = 1000). We systematically vary the number and magnitude of correlated residuals as well as nonspecific misspecification, to evaluate the impact on model fit indices in fitting a two-factor EFA. Our results show that all fit indices, except SRMR, are overly sensitive to correlated residuals and nonspecific error, resulting in solutions which are over-factored. SRMR performed well, consistently selecting the correct number of factors; however, previous research suggests it does not perform well with categorical data. In general, we do not recommend using model fit indices to select number of factors in a scale evaluation framework.

Sign in / Sign up

Export Citation Format

Share Document