Addressing Item-Level Missing Data: A Comparison of Proration and Full Information Maximum Likelihood Estimation

Gina L Mazza,Craig K Enders,Linda S Ruehlman

doi:10.1080/00273171.2015.1068157

Abstract

Often when participants have missing scores on one or more of the items comprising a scale, researchers compute prorated scale scores by averaging the available items. Methodologists have cautioned that proration may make strict assumptions about the mean and covariance structures of the items comprising the scale (Schafer & Graham, 2002; Graham, 2009; Enders, 2010). We investigated proration empirically and found that it resulted in bias even under a missing completely at random (MCAR) mechanism. To encourage researchers to forgo proration, we describe a full information maximum likelihood (FIML) approach to item-level missing data handling that mitigates the loss in power due to missing scale scores and utilizes the available item-level data without altering the substantive analysis. Specifically, we propose treating the scale score as missing whenever one or more of the items are missing and incorporating items as auxiliary variables. Our simulations suggest that item-level missing data handling drastically increases power relative to scale-level missing data handling. These results have important practical implications, especially when recruiting more participants is prohibitively difficult or expensive. Finally, we illustrate the proposed method with data from an online chronic pain management program.

Highlights

Often when participants have missing scores on one or more of the items comprising a scale, researchers compute prorated scale scores by averaging the available items
One of the major goals of this paper is to describe an full information maximum likelihood (FIML) approach to item-level missing data handling that mitigates the loss in power due to missing scale scores and utilizes the available item-level data without altering the substantive analysis
We present two simulation studies that examine its performance, and we demonstrate its application with data from an online chronic pain management program

Summary

Introduction

Often when participants have missing scores on one or more of the items comprising a scale, researchers compute prorated scale scores by averaging the available items. Our simulations suggest that item-level missing data handling drastically increases power relative to scale-level missing data handling These results have important practical implications, especially when recruiting more participants is prohibitively difficult or expensive. When the number of incomplete items exceeded the stated threshold, researchers tended to treat the entire record as missing (deletion) These references suggest that researchers routinely encounter item-level missing data, and they often apply proration to deal with the problem. Recall that proration is equivalent to imputing each participant’s missing scores with the mean of his or her observed scores For these imputations to be valid, the incomplete items must have the same properties as the complete donor items. We present simulation studies that examine this issue later in the paper

Objectives

Results

Conclusion