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 values with an estimate derived from iterated multiple regression. The treated data matrices were subjected to multiple regression analyses, and the resulting regression equations were compared to the equations obtained from the original, complete data. The two regression based estimation procedures provided the most accurate regression equations, followed by the method of inserting means. Discarding cases with incomplete records was the least accurate method. Although the results supported the practice of using covariate information to estimate missing data, the increases in accuracy were minimal under the conditions investigated.

The effect of preceptor role effectiveness on newly licensed registered nurses' perceived psychological empowerment and professional autonomy

“…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).…”

To Invest or Not? The Role of Coworker Support and Trust in Daily Reciprocal Gain Spirals of Helping Behavior

“…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).…”

Performance Evaluation of Imputation Methods for Incomplete Datasets

“…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.…”