2.50
Hdl Handle:
http://hdl.handle.net/10755/163802
Category:
Abstract
Type:
Presentation
Title:
Multiple Imputation for Missing Data
Author(s):
Patrician, Patricia
Author Details:
Patricia Patrician, University of Alabama at Birmingham School of Nursing, Birmingham, Alabama, USA, ppatrici@uab.edu
Abstract:
Purpose: The purpose of this project is to illustrate the process of multiple imputation and demonstrate its utility in overcoming the challenges of missing data. Multiple imputation, a predictive approach to handling missing data in multivariate analyses, creates plausible imputations of missing values, accurately reflects uncertainty and preserves important data relationships. Specific aims: To review the problems associated with missing data particularly in survey and longitudinal research, to discuss conventional methods of reconciling missing data, and to describe and apply multiple imputation techniques. Data from an empirical investigation of AIDS care is used to illustrate the process of multiple imputation. Methods: This exercise employs cumulative logit analysis for an ordered category dependent variable and generalized estimating equations (GEE) to adjust for the non-independent repeated measures (N=approximately 6000 observations from 400 nurses). After obtaining an initial set of results, 25% of values on two variables were randomly removed. Five multiply imputed data sets were generated, and results were combined to yield a single set of parameter estimates and standard errors. Results and Conclusions: The multiply imputed results were very similar to the original results obtained prior to randomly removing information on two variables. Significance levels were identical; parameter estimates and standard errors did not vary considerably. Although listwise deletion and mean imputation are the most common techniques to reconcile missing data, multiple imputation techniques may yield far superior parameter estimates, standard errors and test statistics. Implications: Nurse researchers employing survey methods or conducting secondary analyses of large data sets invariably encounter missing data. Incomplete data are problematic particularly in the presence of substantial absent information and/or systematic non-response patterns. Multiple imputation represents an improvement over more commonly used methods of handling missing data. Disclaimer: The opinions or assertions contained herein are the private views of the author and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defense.
Repository Posting Date:
27-Oct-2011
Date of Publication:
27-Oct-2011
Conference Date:
2001
Conference Name:
ENRS 13th Annual Scientific Sessions
Conference Host:
Eastern Nursing Research Society
Conference Location:
Atlantic City, New Jersey, USA
Sponsors:
Acknowledgments: This work was supported by grants from the National Institute for Nursing Research (RO1 NR02280), Agency for Healthcare Research and Quality (RO1 HS08603) and an educational grant from the Army Nurse Corps.
Note:
This is an abstract-only submission. If the author has submitted a full-text item based on this abstract, you may find it by browsing the Virginia Henderson Global Nursing e-Repository by author. If author contact information is available in this abstract, please feel free to contact him or her with your queries regarding this submission. Alternatively, please contact the conference host, journal, or publisher (according to the circumstance) for further details regarding this item. If a citation is listed in this record, the item has been published and is available via open-access avenues or a journal/database subscription. Contact your library for assistance in obtaining the as-published article.

Full metadata record

DC FieldValue Language
dc.type.categoryAbstracten_US
dc.typePresentationen_GB
dc.titleMultiple Imputation for Missing Dataen_GB
dc.contributor.authorPatrician, Patriciaen_US
dc.author.detailsPatricia Patrician, University of Alabama at Birmingham School of Nursing, Birmingham, Alabama, USA, ppatrici@uab.eduen_US
dc.identifier.urihttp://hdl.handle.net/10755/163802-
dc.description.abstractPurpose: The purpose of this project is to illustrate the process of multiple imputation and demonstrate its utility in overcoming the challenges of missing data. Multiple imputation, a predictive approach to handling missing data in multivariate analyses, creates plausible imputations of missing values, accurately reflects uncertainty and preserves important data relationships. Specific aims: To review the problems associated with missing data particularly in survey and longitudinal research, to discuss conventional methods of reconciling missing data, and to describe and apply multiple imputation techniques. Data from an empirical investigation of AIDS care is used to illustrate the process of multiple imputation. Methods: This exercise employs cumulative logit analysis for an ordered category dependent variable and generalized estimating equations (GEE) to adjust for the non-independent repeated measures (N=approximately 6000 observations from 400 nurses). After obtaining an initial set of results, 25% of values on two variables were randomly removed. Five multiply imputed data sets were generated, and results were combined to yield a single set of parameter estimates and standard errors. Results and Conclusions: The multiply imputed results were very similar to the original results obtained prior to randomly removing information on two variables. Significance levels were identical; parameter estimates and standard errors did not vary considerably. Although listwise deletion and mean imputation are the most common techniques to reconcile missing data, multiple imputation techniques may yield far superior parameter estimates, standard errors and test statistics. Implications: Nurse researchers employing survey methods or conducting secondary analyses of large data sets invariably encounter missing data. Incomplete data are problematic particularly in the presence of substantial absent information and/or systematic non-response patterns. Multiple imputation represents an improvement over more commonly used methods of handling missing data. Disclaimer: The opinions or assertions contained herein are the private views of the author and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defense.en_GB
dc.date.available2011-10-27T11:14:05Z-
dc.date.issued2011-10-27en_GB
dc.date.accessioned2011-10-27T11:14:05Z-
dc.conference.date2001en_US
dc.conference.nameENRS 13th Annual Scientific Sessionsen_US
dc.conference.hostEastern Nursing Research Societyen_US
dc.conference.locationAtlantic City, New Jersey, USAen_US
dc.description.sponsorshipAcknowledgments: This work was supported by grants from the National Institute for Nursing Research (RO1 NR02280), Agency for Healthcare Research and Quality (RO1 HS08603) and an educational grant from the Army Nurse Corps.en_US
dc.description.noteThis is an abstract-only submission. If the author has submitted a full-text item based on this abstract, you may find it by browsing the Virginia Henderson Global Nursing e-Repository by author. If author contact information is available in this abstract, please feel free to contact him or her with your queries regarding this submission. Alternatively, please contact the conference host, journal, or publisher (according to the circumstance) for further details regarding this item. If a citation is listed in this record, the item has been published and is available via open-access avenues or a journal/database subscription. Contact your library for assistance in obtaining the as-published article.-
All Items in this repository are protected by copyright, with all rights reserved, unless otherwise indicated.