2003
DOI: 10.1016/s0377-2217(02)00578-7
|View full text |Cite
|
Sign up to set email alerts
|

The comparative efficacy of imputation methods for missing data in structural equation modeling

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

1
148
0
1

Year Published

2005
2005
2022
2022

Publication Types

Select...
4
4

Relationship

0
8

Authors

Journals

citations
Cited by 251 publications
(158 citation statements)
references
References 27 publications
1
148
0
1
Order By: Relevance
“…Since the missing data were consistent with the assumption of MAR in this study, statistical analyses were carried out using the FIML, which allows all available information to be used without imputing data (Muthén & Muthén, 1998-2015. Simulation studies show that FIML provides less-biased regression parameter estimates compared to other missing data procedures (Enders, 2001;Olinsky, Chen, & Harlow, 2003). The results of the study will, thus, be reported for the full sample.…”
Section: Aims and Hypothesesmentioning
confidence: 95%
“…Since the missing data were consistent with the assumption of MAR in this study, statistical analyses were carried out using the FIML, which allows all available information to be used without imputing data (Muthén & Muthén, 1998-2015. Simulation studies show that FIML provides less-biased regression parameter estimates compared to other missing data procedures (Enders, 2001;Olinsky, Chen, & Harlow, 2003). The results of the study will, thus, be reported for the full sample.…”
Section: Aims and Hypothesesmentioning
confidence: 95%
“…Another limitation in this study has been the deletion of incomplete records from the analysis (Little and Rubin 2002). It would be worthwhile to investigate the application of missing data imputation techniques in a subsequent paper (Olinsky et al 2003).…”
Section: Structural Equations Modelling Analysismentioning
confidence: 99%
“…While, multiple imputa tion technique is more comprehensive than expectation maximization, it is the most complex of stochastic imputa tion method so far (Schlomer et al 2010). Full Information Maximum Likelihood (FILM), however, represents an al ternative efficient stochastic imputation technique, as it produces unbiased results and estimates accurate standard error and confidence interval by retaining the sample size as well as it produces similar results to each of EM and MI (Enders, Bandalos 2001;Olinsky et al 2003). From another standpoint, while scholars consider that FIML is an effici ent technique and one of the preferred methods for handle missing data (Enders, Bandalos 2001;Schlomer et al 2010), others believe that MI is probably most promising, because of its theoretical and distributional underpinnings, while FIML is a direct model based method for estimating para meters in the presence of missing data and does not actually impute the missing data (Olinsky et al 2003).…”
Section: Missing Data: Better To Manage Not To Damagementioning
confidence: 99%
“…Full Information Maximum Likelihood (FILM), however, represents an al ternative efficient stochastic imputation technique, as it produces unbiased results and estimates accurate standard error and confidence interval by retaining the sample size as well as it produces similar results to each of EM and MI (Enders, Bandalos 2001;Olinsky et al 2003). From another standpoint, while scholars consider that FIML is an effici ent technique and one of the preferred methods for handle missing data (Enders, Bandalos 2001;Schlomer et al 2010), others believe that MI is probably most promising, because of its theoretical and distributional underpinnings, while FIML is a direct model based method for estimating para meters in the presence of missing data and does not actually impute the missing data (Olinsky et al 2003). However, full information maximum likelihood (FIML) is a superior met hod comparing to multiple imputation (MI), as its ability to manage missing data and conduct analyses in one step, as well as it is a superior method comparing to expecta tion maximization (EM), as its ability to estimate accurate standard errors and confidence intervals by retaining the sample size.…”
Section: Missing Data: Better To Manage Not To Damagementioning
confidence: 99%
See 1 more Smart Citation