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

Gina L. Mazza, Craig K. Enders, Linda S. Ruehlman

Research output: Contribution to journalArticle

23 Citations (Scopus)

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.

Original languageEnglish (US)
Pages (from-to)504-519
Number of pages16
JournalMultivariate Behavioral Research
Volume50
Issue number5
DOIs
StatePublished - Sep 3 2015

Fingerprint

Missing Data
Maximum Likelihood Estimation
Research Personnel
Data Handling
Pain Management
Chronic Pain
Missing Completely at Random
Auxiliary Variables
Covariance Structure
Pain
Maximum Likelihood
Averaging
Simulation

Keywords

  • auxiliary variables
  • full information maximum likelihood
  • missing data
  • proration
  • questionnaires

ASJC Scopus subject areas

  • Experimental and Cognitive Psychology
  • Statistics and Probability
  • Arts and Humanities (miscellaneous)

Cite this

Addressing Item-Level Missing Data : A Comparison of Proration and Full Information Maximum Likelihood Estimation. / Mazza, Gina L.; Enders, Craig K.; Ruehlman, Linda S.

In: Multivariate Behavioral Research, Vol. 50, No. 5, 03.09.2015, p. 504-519.

Research output: Contribution to journalArticle

Mazza, Gina L. ; Enders, Craig K. ; Ruehlman, Linda S. / Addressing Item-Level Missing Data : A Comparison of Proration and Full Information Maximum Likelihood Estimation. In: Multivariate Behavioral Research. 2015 ; Vol. 50, No. 5. pp. 504-519.
@article{dd28a75b1be04e33960068f19f474535,
title = "Addressing Item-Level Missing Data: A Comparison of Proration and Full Information Maximum Likelihood Estimation",
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.",
keywords = "auxiliary variables, full information maximum likelihood, missing data, proration, questionnaires",
author = "Mazza, {Gina L.} and Enders, {Craig K.} and Ruehlman, {Linda S.}",
year = "2015",
month = "9",
day = "3",
doi = "10.1080/00273171.2015.1068157",
language = "English (US)",
volume = "50",
pages = "504--519",
journal = "Multivariate Behavioral Research",
issn = "0027-3171",
publisher = "Psychology Press Ltd",
number = "5",

}

TY - JOUR

T1 - Addressing Item-Level Missing Data

T2 - A Comparison of Proration and Full Information Maximum Likelihood Estimation

AU - Mazza, Gina L.

AU - Enders, Craig K.

AU - Ruehlman, Linda S.

PY - 2015/9/3

Y1 - 2015/9/3

N2 - 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.

AB - 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.

KW - auxiliary variables

KW - full information maximum likelihood

KW - missing data

KW - proration

KW - questionnaires

UR - http://www.scopus.com/inward/record.url?scp=84944045316&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84944045316&partnerID=8YFLogxK

U2 - 10.1080/00273171.2015.1068157

DO - 10.1080/00273171.2015.1068157

M3 - Article

AN - SCOPUS:84944045316

VL - 50

SP - 504

EP - 519

JO - Multivariate Behavioral Research

JF - Multivariate Behavioral Research

SN - 0027-3171

IS - 5

ER -