Analyzing Incomplete Item Scores in Longitudinal Data by Including Item Score Information as Auxiliary Variables
- 27 January 2015
- journal article
- research article
- Published by Taylor & Francis in Structural Equation Modeling: A Multidisciplinary Journal
- Vol. 22 (4) , 588-602
Abstract
The aim of this study is to investigate a novel method for dealing with incomplete scale scores in longitudinal data that result from missing item responses. This method includes item information as auxiliary variables, which is advantageous because it incorporates the observed item-level data while maintaining the scale scores as the focus of the analysis. These auxiliary variables do not change the analysis model, but improve missing data handling. The investigated novel method uses the item scores or some summary of a parcel of item scores as auxiliary variables, while treating the scale scores missing in a latent growth model. The performance of these methods was examined in several simulated longitudinal data conditions and analyzed through bias, mean square error, and coverage. Results show that including the item information as auxiliary variables results in rather dramatic power gains compared with analyses without auxiliary variables under varying conditions.Keywords
This publication has 18 references indexed in Scilit:
- Examining Measure Correlations With Incomplete Data SetsStructural Equation Modeling: A Multidisciplinary Journal, 2014
- Missing data in a multi-item instrument were best handled by multiple imputation at the item score levelJournal of Clinical Epidemiology, 2014
- Missing DataEpidemiology, 2012
- A Comparison of Item-Level and Scale-Level Multiple Imputation for Questionnaire BatteriesMultivariate Behavioral Research, 2012
- Analyzing longitudinal data with missing values.Rehabilitation Psychology, 2011
- A Note on the Use of Missing Auxiliary Variables in Full Information Maximum Likelihood-Based Structural Equation ModelsStructural Equation Modeling: A Multidisciplinary Journal, 2008
- Adding Missing-Data-Relevant Variables to FIML-Based Structural Equation ModelsStructural Equation Modeling: A Multidisciplinary Journal, 2003
- Manual Therapy, Physical Therapy, or Continued Care by a General Practitioner for Patients with Neck PainAnnals of Internal Medicine, 2002
- Influence of Imputation and EM Methods on Factor Analysis when Item Nonresponse in Questionnaire Data is NonignorableMultivariate Behavioral Research, 2000
- Inference and missing dataBiometrika, 1976