On assessing model fit for distribution-free longitudinal models under missing data

2013 ◽  
Vol 33 (1) ◽  
pp. 143-157 ◽  
Author(s):  
P. Wu ◽  
X.M. Tu ◽  
J. Kowalski
Keyword(s):  
Missing Data ◽  
Model Fit ◽  
Longitudinal Models ◽  
Distribution Free

scholarly journals Fitting Ordinal Factor Analysis Models With Missing Data: A Comparison Between Pairwise Deletion and Multiple Imputation

2019 ◽  
Vol 80 (1) ◽  
pp. 41-66 ◽  
Author(s):  
Dexin Shi ◽  
Taehun Lee ◽  
Amanda J. Fairchild ◽  
Alberto Maydeu-Olivares
Keyword(s):  
Factor Analysis ◽  
Missing Data ◽  
Multiple Imputation ◽  
Missing At Random ◽  
Complete Data ◽  
Model Fit ◽  
Parameter Estimates ◽  
Fit Indices ◽  
Wide Range ◽  
Analysis Models

This study compares two missing data procedures in the context of ordinal factor analysis models: pairwise deletion (PD; the default setting in Mplus) and multiple imputation (MI). We examine which procedure demonstrates parameter estimates and model fit indices closer to those of complete data. The performance of PD and MI are compared under a wide range of conditions, including number of response categories, sample size, percent of missingness, and degree of model misfit. Results indicate that both PD and MI yield parameter estimates similar to those from analysis of complete data under conditions where the data are missing completely at random (MCAR). When the data are missing at random (MAR), PD parameter estimates are shown to be severely biased across parameter combinations in the study. When the percentage of missingness is less than 50%, MI yields parameter estimates that are similar to results from complete data. However, the fit indices (i.e., χ2, RMSEA, and WRMR) yield estimates that suggested a worse fit than results observed in complete data. We recommend that applied researchers use MI when fitting ordinal factor models with missing data. We further recommend interpreting model fit based on the TLI and CFI incremental fit indices.

scholarly journals On the Treatment of Missing Item Responses in Educational Large-scale Assessment Data: The Case of PISA 2018 Mathematics

2021 ◽  
Author(s):  
Alexander Robitzsch
Keyword(s):  
Missing Data ◽  
Large Scale ◽  
Model Fit ◽  
Large Scale Assessment ◽  
Missing Data Treatments ◽  
Scale Assessment ◽  
Scaling Models ◽  
Item Responses ◽  
Response Propensity ◽  
Missing Item

Missing item responses are prevalent in educational large-scale assessment studies like the programme for international student assessment (PISA). The current operational practice scores missing item responses as wrong, but several psychometricians advocated a model-based treatment based on latent ignorability assumption. In this approach, item responses and response indicators are jointly modeled conditional on a latent ability and a latent response propensity variable. Alternatively, imputation-based approaches can be used. The latent ignorability assumption is weakened in the Mislevy-Wu model that characterizes a nonignorable missingness mechanism and allows the missingness of an item to depend on the item itself. The scoring of missing item responses as wrong and the latent ignorable model are submodels of the Mislevy-Wu model. This article uses the PISA 2018 mathematics dataset to investigate the consequences of different missing data treatments on country means. Obtained country means can substantially differ for the different scaling models. In contrast to previous statements in the literature, the scoring of missing item responses as incorrect provided a better model fit than a latent ignorable model for most countries. Furthermore, the dependence of the missingness of an item from the item itself after conditioning on the latent response propensity was much more pronounced for constructed-response items than for multiple-choice items. As a consequence, scaling models that presuppose latent ignorability should be refused from two perspectives. First, the Mislevy-Wu model is preferred over the latent ignorable model for reasons of model fit. Second, we argue that model fit should only play a minor role in choosing psychometric models in large-scale assessment studies because validity aspects are most relevant. Missing data treatments that countries can simply manipulate (and, hence, their students) result in unfair country comparisons.

Bivariate Structural Regression Analysis: A Tool for the Comparison of Analytic Methods

1987 ◽  
Vol 26 (04) ◽  
pp. 205-214 ◽  
Author(s):  
U. Feldmann ◽  
B. Schneider
Keyword(s):  
Regression Analysis ◽  
Clinical Chemistry ◽  
Statistical Tests ◽  
Model Fit ◽  
Likelihood Method ◽  
Distribution Free ◽  
Accuracy And Precision ◽  
Distribution Free Approach ◽  
Structural Regression ◽  
A New Technique

SummaryThis paper introduces the concept of bivariate structural regression analysis, a new technique which offers some advantages compared to the well-known structural relationship approach. The concept is not restricted to multivariate normal distribution, and without additional constraints the model remains identifiable in the bivariate case. A bivariate calibration line is developed first by the maximum likelihood method and then also distribution-free by applying rank statistics. Both estimation procedures coincide, if the distribution assumption is satisfied. Hence, the distribution-free approach has an efficiency of 100%. Our concept of analysis is applied to the comparison of analytical methods in clinical chemistry. Appropriate statistical tests concerning accuracy and precision as well as the model fit are offered.

scholarly journals Factor structure and convergent validity of the Derriford Appearance Scale-24 using standard scoring versus treating ‘not applicable’ responses as missing data: a Scleroderma Patient-centered Intervention Network (SPIN) cohort study

BMJ Open ◽  
2018 ◽  
Vol 8 (3) ◽  
pp. e018641 ◽  
Author(s):  
Erin L Merz ◽  
Linda Kwakkenbos ◽  
Marie-Eve Carrier ◽  
Shadi Gholizadeh ◽  
Sarah D Mills ◽  
...  
Keyword(s):  
Missing Data ◽  
Factor Structure ◽  
Convergent Validity ◽  
Factor Model ◽  
Negative Evaluation ◽  
Model Fit ◽  
Secondary Outcome ◽  
Scoring Methods ◽  
Two Samples ◽  
The Uk

ObjectiveValid measures of appearance concern are needed in systemic sclerosis (SSc), a rare, disfiguring autoimmune disease. The Derriford Appearance Scale-24 (DAS-24) assesses appearance-related distress related to visible differences. There is uncertainty regarding its factor structure, possibly due to its scoring method.DesignCross-sectional survey.SettingParticipants with SSc were recruited from 27 centres in Canada, the USA and the UK. Participants who self-identified as having visible differences were recruited from community and clinical settings in the UK.ParticipantsTwo samples were analysed (n=950 participants with SSc; n=1265 participants with visible differences).Primary and secondary outcome measuresThe DAS-24 factor structure was evaluated using two scoring methods. Convergent validity was evaluated with measures of social interaction anxiety, depression, fear of negative evaluation, social discomfort and dissatisfaction with appearance.ResultsWhen items marked by respondents as ‘not applicable’ were scored as 0, per standard DAS-24 scoring, a one-factor model fit poorly; when treated as missing data, the one-factor model fit well. Convergent validity analyses revealed strong correlations that were similar across scoring methods.ConclusionsTreating ‘not applicable’ responses as missing improved the measurement model, but did not substantively influence practical inferences that can be drawn from DAS-24 scores. Indications of item redundancy and poorly performing items suggest that the DAS-24 could be improved and potentially shortened.

Sign in / Sign up

Export Citation Format

Share Document