Is Item Imputation Always Better? An Investigation of Wave-Missing Data in Growth Models

Vera, Juan Diego; Enders, Craig K.

doi:10.1080/10705511.2020.1850289

Cited by 4 publications

(4 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The reason for this finding is straightforward: itemlevel missing data handling leverages strong sources of within-scale covariation among items measuring the same construct. This is a robust finding across different contexts (i.e., cross-sectional data, longitudinal data), with few exceptions (e.g., when entire waves of questionnaire data are missing; Vera & Enders, 2021).…”

mentioning

confidence: 64%

A factored regression model for composite scores with item-level missing data.

Alacam¹,

Enders²,

Du³

et al. 2023

Psychological Methods

View full text Add to dashboard Cite

mentioning

confidence: 64%

A factored regression model for composite scores with item-level missing data.

Alacam¹,

Enders²,

Du³

et al. 2023

Psychological Methods

View full text Add to dashboard Cite

“…For example, the simulated scales often had high internal consistencies, which may have benefited certain procedures; in fact, we also explored perturbations of the scale structure and found that FCS-PI performed more poorly in these conditions (for additional details, see Supplement C in the supplemental online materials). For this reason, future research should evaluate these methods in different contexts, for example, in longitudinal designs that feature large numbers of repeated measurements (see also Vera & Enders, 2021). In addition, we focused on relatively simple types of analyses, and it would be interesting to also consider the impact of the different procedures on the results obtained in more complex types of analyses (e.g., structural equation models; see Gottschall et al, 2012).…”

Section: Discussionmentioning

confidence: 99%

“…Composite score approaches have been used to simplify imputation models by combining some of the variables into composite scores (e.g., sum of mean scores; see Figure 1 for an illustration). For example, in multi-item scales, it has been recommended that items from the same scale should be combined into scale scores (e.g., scale means; Eekhout et al, 2018;Gottschall et al, 2012;Vera & Enders, 2021). Combining items into scale scores simplifies the imputation model by effectively placing constraints on the parameters of the fully-specified model in accordance with the suspected scale structure (see also Alacam et al, 2023).…”

Section: Composite Scoresmentioning

confidence: 99%

Multiple imputation of missing data in large studies with many variables: A fully conditional specification approach using partial least squares

Grund,

Lüdtke,

Robitzsch

2023

Preprint

View full text Add to dashboard Cite

Multiple imputation (MI) is one of the most popular methods for handling missing data in psychological research. However, many imputation approaches are poorly equipped to handle large numbers of variables, which are a common sight in studies that employ questionnaires to assess psychological constructs. In such a case, conventional imputation approaches often become unstable and require that the imputation model be simplified, for example, by removing variables or combining them into composite scores. In this article, we propose an alternative method that extends the fully conditional specification (FCS) approach to MI with dimension reduction techniques such as partial least squares (PLS). To evaluate this approach, we conducted a series of simulation studies, in which we compared it with other approaches that were based on variable selection, composite scores, or dimension reduction through principal components analysis (PCA). Our findings indicate that this novel approach can provide accurate results even in challenging scenarios, where other approaches fail to do so. Finally, we also illustrate the use of this method in real data and discuss the implications of our findings for practice.

show abstract

“…Such problems are commonly encountered in practice when imputing items in large‐scale longitudinal studies because of the need to fit imputation models that contain a large number of highly correlated variables. Another paper noted that item‐level imputation does not produce precision gains in longitudinal studies where entire questionnaire batteries are missing 21 . Furthermore, if the scale score is of interest in the analysis model but is left out of the imputation model, item‐level imputation is not congenial with the analysis model, which could lead to bias 7,13 …”

Section: Introductionmentioning

confidence: 99%

A comparison of multiple imputation strategies for handling missing data in multi‐item scales: Guidance for longitudinal studies

Mainzer

Apajee

Nguyen

et al. 2021

Statistics in Medicine

View full text Add to dashboard Cite

Medical research often involves using multi‐item scales to assess individual characteristics, disease severity, and other health‐related outcomes. It is common to observe missing data in the scale scores, due to missing data in one or more items that make up that score. Multiple imputation (MI) is a popular method for handling missing data. However, it is not clear how best to use MI in the context of scale scores, particularly when they are assessed at multiple waves of data collection resulting in large numbers of items. The aim of this article is to provide practical advice on how to impute missing values in a repeatedly measured multi‐item scale using MI when inference on the scale score is of interest. We evaluated the performance of five MI strategies for imputing missing data at either the item or scale level using simulated data and a case study based on four waves of the Longitudinal Study of Australian Children (LSAC). MI was implemented using both multivariate normal imputation and fully conditional specification, with two rules for calculating the scale score. A complete case analysis was also performed for comparison. Based on our results, we caution against the use of a MI strategy that does not include the scale score in the imputation model(s) when the scale score is required for analysis.

show abstract

Is Item Imputation Always Better? An Investigation of Wave-Missing Data in Growth Models

Cited by 4 publications

References 33 publications

A factored regression model for composite scores with item-level missing data.

A factored regression model for composite scores with item-level missing data.

Multiple imputation of missing data in large studies with many variables: A fully conditional specification approach using partial least squares

A comparison of multiple imputation strategies for handling missing data in multi‐item scales: Guidance for longitudinal studies

Contact Info

Product

Resources

About