1987
DOI: 10.1177/0013164487471002
|View full text |Cite
|
Sign up to set email alerts
|

A Comparison of Methods for Treating Incomplete Data in Selection Research

Abstract: Data were simulated to conform to covariance patterns taken from the personnel selection literature. Two, six, and ten percent of the values were deleted from one of three predictor variables in sample sizes of 50, 100 and 200. Incomplete data matrices were treated by four methods: (a) elimination of cases with incomplete data records; (b) substitution of missing values with the variable mean; (c) replacement of missing values with an estimate obtained from simple regression; and (d) replacement of missing val… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

7
134
0
1

Year Published

1996
1996
2022
2022

Publication Types

Select...
6
2
2

Relationship

0
10

Authors

Journals

citations
Cited by 121 publications
(142 citation statements)
references
References 13 publications
7
134
0
1
Order By: Relevance
“…Cohen and Cohen (1983) suggest that missing data up to 10% on a specific variable is considered small and the variable should be retained in the analysis. Furthermore, Raymond and Roberts (1987) estimated that a variable should be retained with 40% or less missing data. Removing cases with missing data from data analysis procedures can result in reduced sample size, compromised statistical power, and inaccurate parameter estimates (Barnard & Meng, 1999;Patrician, 2002;Tabachnick & Fidell, 2001).…”
Section: Discussionmentioning
confidence: 99%
“…Cohen and Cohen (1983) suggest that missing data up to 10% on a specific variable is considered small and the variable should be retained in the analysis. Furthermore, Raymond and Roberts (1987) estimated that a variable should be retained with 40% or less missing data. Removing cases with missing data from data analysis procedures can result in reduced sample size, compromised statistical power, and inaccurate parameter estimates (Barnard & Meng, 1999;Patrician, 2002;Tabachnick & Fidell, 2001).…”
Section: Discussionmentioning
confidence: 99%
“…Because the missing data appeared to be nonsystematic and was well below missing-at-random rates (e.g., 40%; see Raymond & Roberts, 1987), we imputed variable means for individual missing data points (Paul, Mason, McCaffrey, & Fox, 2008).…”
Section: Participants and Proceduresmentioning
confidence: 99%
“…All MDTs deteriorate as the 39 percentage of missingness grows and it is almost inappropriate to apply any of them Da t a S e t s S e r ie s 1 P e f or m a nc e of M e a n I mp u t a t i on when the missing percentage is greater than 50. Raymond et al [16] found that 1 when data are missing at random, MI performed better than LD. In our results, we found in two instances that LD outperformed MI.…”
mentioning
confidence: 99%