Chapter 9 Auxiliary Variables

The MAR assumption can be made more plausible by including additional variables in the imputation model (Baraldi and Enders (2010)). Therefore, it is advised, to include extra variables that have a relationship with the missing data rate in other variables, i.e. have a relationship with the probability of missing data or that have a relationship (correlated) with the variables that contain the missing values (Collins, Schafer, and Kam (2001)). These additional variables can help dealing with missing data as well and are called auxiliary variables.

References

Baraldi, A. N., and C. K. Enders. 2010. “An introduction to modern missing data analyses.” J Sch Psychol 48 (1): 5–37.

Collins, L. M., J. L. Schafer, and C. M. Kam. 2001. “A Comparison of Inclusive and Restrictive Strategies in Modern Missing Data Procedures.” Psychological Methods 6 (3): 330–51.