Missing data: Our view of the state of the art.

Schafer, Joseph L.; Graham, John W.

doi:10.1037/1082-989x.7.2.147

Cited by 9,460 publications

(7,877 citation statements)

References 87 publications

Supporting

Mentioning

7,570

Contrasting

Unclassified

112

Order By: Relevance

“…All estimates remained within the limits for unproblematic estimates [27] of ± 0.5 SE from the true value with up to 50% missingness. Only the regression coefficient estimates for stage, pre-operative RT, post-operative RT and indication for CT (Figure 1) could be deemed problematic [27] with 75% missingness. However, the percentage biases were more extreme than the specified accuracy given the number of replications performed for the majority of covariates with 25% or more missingness (Figure 2).…”

Section: Resultsmentioning

confidence: 64%

See 1 more Smart Citation

Comparison of imputation methods for handling missing covariate data when fitting a Cox proportional hazards model: a resampling study

Marshall

Altman

Holder

2010

BMC Med Res Methodol

111

103

View full text Add to dashboard Cite

BackgroundThe appropriate handling of missing covariate data in prognostic modelling studies is yet to be conclusively determined. A resampling study was performed to investigate the effects of different missing data methods on the performance of a prognostic model.MethodsObserved data for 1000 cases were sampled with replacement from a large complete dataset of 7507 patients to obtain 500 replications. Five levels of missingness (ranging from 5% to 75%) were imposed on three covariates using a missing at random (MAR) mechanism. Five missing data methods were applied; a) complete case analysis (CC) b) single imputation using regression switching with predictive mean matching (SI), c) multiple imputation using regression switching imputation, d) multiple imputation using regression switching with predictive mean matching (MICE-PMM) and e) multiple imputation using flexible additive imputation models. A Cox proportional hazards model was fitted to each dataset and estimates for the regression coefficients and model performance measures obtained.ResultsCC produced biased regression coefficient estimates and inflated standard errors (SEs) with 25% or more missingness. The underestimated SE after SI resulted in poor coverage with 25% or more missingness. Of the MI approaches investigated, MI using MICE-PMM produced the least biased estimates and better model performance measures. However, this MI approach still produced biased regression coefficient estimates with 75% missingness.ConclusionsVery few differences were seen between the results from all missing data approaches with 5% missingness. However, performing MI using MICE-PMM may be the preferred missing data approach for handling between 10% and 50% MAR missingness.

show abstract

Section: Resultsmentioning

confidence: 64%

“…The regression coefficient estimates were compared against the "true" values in terms of their bias, coverage and efficiency [27]. The average regression coefficient estimates and associated empirical SE obtained from performing 20000 replications of 1000 cases with complete data were considered as the "true" values.…”

Section: Methodsmentioning

confidence: 99%

Comparison of imputation methods for handling missing covariate data when fitting a Cox proportional hazards model: a resampling study

Marshall

Altman

Holder

2010

BMC Med Res Methodol

111

103

View full text Add to dashboard Cite

show abstract

“…In our sample, 86 (8.1%) were missing one PCL-5 item, 25 (2.4%) were missing two or three PCL-5 items, 201 (18.9%) were missing one SF-36 item, and 90 (8.5%) were missing two to six SF-36 items. Missing data on the PCL-5 and SF-36 were estimated using maximum likelihood (ML) procedures (Schafer & Graham, 2002). Two structural equation models (SEM) were constructed to examine associations between PTSD symptom dimensions and physical/psychosocial HRQoL (see Supplementary Figures 1 and 2).…”

Section: Methodsmentioning

confidence: 99%

DSM-5 posttraumatic stress symptom dimensions and health-related quality of life among Chinese earthquake survivors

Wang

Cao

et al. 2018

European Journal of Psychotraumatology

View full text Add to dashboard Cite

It has been well-documented that posttraumatic stress symptoms cause impairments in health-related quality of life (HRQoL). Until now we have little data on how DSM-5 PTSD symptom dimensions relate to different aspects of HRQoL. Clarifying this question would be informative to improve the quality of life of PTSD patients. This study aimed to investigate the effects of dimensions of a well-supported seven-factor model of DSM-5 PTSD symptoms on physical and psychosocial HRQoL. A total of 1063 adult survivors of the 2008 Wenchuan earthquake took part in this study nine years after the disaster. PTSD symptoms were measured by the PTSD Checklist for DSM-5 (PCL-5). HRQoL was measured by the Medical Outcomes Survey Short Form-36 (SF-36). The associations between PTSD symptom dimensions and HRQoL were examined using structural equation models. Dysphoric arousal symptoms were found to significantly relate to physical HRQoL. Other symptom dimensions were not associated with HRQoL. Our findings contribute to the relationship between DSM-5 PTSD and HRQoL, and carry implications for further clinical practice and research on trauma-exposed individuals.

show abstract

“…In addition, listwise deletion produces unbiased estimates only under the assumption that the missing data occur completely at random, which is not realistic in many cases (Enders, 2006). Accordingly, as noted in the main text, we applied the full information maximum likelihood method (FIML; see Schafer & Graham, 2002). This approach uses the likelihood function, which considers individual patterns of missing data, and which has the important advantage of rendering results that make use of the full information from the entire sample.…”

Section: Missing Datamentioning

confidence: 99%

Breaking the double-edged sword of effort/trying hard: Developmental equilibrium and longitudinal relations among effort, achievement, and academic self-concept.

Marsh

Pekrun²,

Lichtenfeld³

et al. 2016

Developmental Psychology

View full text Add to dashboard Cite

2016)'Breaking the double-edged sword of e ort/trying hard : developmental equilibrium and longitudinal relations among e ort, achievement, and academic self-concept.', Developmental psychology., 52 (8). pp. 1273-1290. Further information on publisher's website:https://doi.org/10.1037/dev0000146Publisher's copyright statement:c 2016 APA, all rights reserved. This article may not exactly replicate the nal version published in the APA journal. It is not the copy of record.Additional information: Use policyThe full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-pro t purposes provided that:• a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders.Please consult the full DRO policy for further details. controlling for the prior effects of all others. There was an asymmetrical pattern of effects for effort that is consistent with the double-edged sword premise: prior school grades had positive effects on subsequent effort, but prior effort had non-significant or negative effects on subsequent grades and ASC. However, on the basis of a synergistic application of new theory and methodology, we predicted and found a significant ASC-by-effort interaction, such that prior effort had more positive effects on subsequent ASC and school grades when prior ASC was high-thus providing a key to breaking the double-edged sword.Keywords: academic effort, academic self-concept, double-edged sword, reciprocal effects models, developmental equilibrium; multigroup longitudinal invariance BREAKING THE DOUBLE-EDGED SWORD 3 Breaking the Double-Edged Sword of Effort/Trying Hard: Developmental Equilibrium andLongitudinal Relations Among Effort, Achievement, and Academic Self-ConceptOur focus is on a longitudinal reciprocal effects model (REM) of the temporal ordering of academic effort, achievement, and academic self-concept (ASC) over the potentially volatile early-to-middle adolescent period. In pursuit of this aim, we introduce a developmental equilibrium hypothesis that posits the consistency of related effects over early-to-middle adolescence, and test it on the basis of longitudinal data.More specifically, in relation to a priori predictions, we formally test developmental equilibrium as the invariance of effects across four waves of data, providing a formal test of the hypothesis that the self-system attains a developmental balance. For the present purposes we focus specifically on the school subject domain of mathematics, which was the basis of the secondary database that we used, noting that the math domain is relevant as an important school subject, and also because of the related psychological processes we are studying, in that effort is likely to be needed to master the skills to be learned.Sel...

show abstract

Missing data: Our view of the state of the art.

Cited by 9,460 publications

References 87 publications

Comparison of imputation methods for handling missing covariate data when fitting a Cox proportional hazards model: a resampling study

Comparison of imputation methods for handling missing covariate data when fitting a Cox proportional hazards model: a resampling study

DSM-5 posttraumatic stress symptom dimensions and health-related quality of life among Chinese earthquake survivors

Breaking the double-edged sword of effort/trying hard: Developmental equilibrium and longitudinal relations among effort, achievement, and academic self-concept.

Contact Info

Product

Resources

About