A comparison of inclusive and restrictive strategies in modern missing data procedures.

By Linda M. Collins, Joseph L. Schafer, Chi-Ming Kam. Psychological methods, 2001

2,455 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/d4d8636924b12f0d127383733db3c21d5d3789a4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Craig K. Enders
2. S. van Buuren
3. Mike W.-L. Cheung
4. Victoria Savalei
5. Daniel A. Newman
6. O. Lüdtke
7. Ian R. White
8. C. Y. Peng

In Statistics and Probability

1. Mike W.-L. Cheung
2. S. van Buuren
3. Victoria Savalei
4. Craig K. Enders
5. Ian R. White
6. Paul T. von Hippel
7. O. Lüdtke
8. C. Y. Peng

Citations in the corpus

• Analysis of Longitudinal Studies With Missing Data Using Covariance Structure Modeling With Full-Information Maximum Likelihood (2005, methodology): “…the sensitivity of FIML to violations of these two requirements shows some robustness of the method to mild deviations from these assumptions (e.g., Collins et al., 2001; Schafer & Graham, 2002), which suggests that FIML may be used in a variety of empirical settings with missing data across the…”
• Multiple imputation using chained equations: Issues and guidance for practice (2011, neutral): “2) [54].”
• Effects of Missing Data Methods in Structural Equation Modeling With Nonnormal Longitudinal Data (2009, background): “Although Collins, Schafer, and Kam (2001) argued that departures from MAR (i.e., MNAR) might not be large enough to effectively invalidate the results of MARbased analysis, it is true that these methods still rest on unverifiable assumptions and might be sensitive to departures from the assumed…”
• Using an EM Covariance Matrix to Estimate Structural Equation Models With Missing Data: Choosing an Adjusted Sample Size to Improve the Accuracy of Inferences (2004, methodology): “Collins et al. (2001) suggested that coverage rates below 90% are problematic.”
• Review of inverse probability weighting for dealing with missing data (2013, methodology): “When the imputation model uses auxiliary variables, MI is generally easier to implement than full likelihood.(24) The similarity between the two methods means that comments in this article about properties of MI generally apply also to full-likelihood methods.”
• Multiple imputation: review of theory, implementation and software (2007, methodology): “27–28], [12, 14, 15].”