Appl Math (Irvine)

In his 1987 classic book on multiple imputation (MI), Rubin used the fraction of missing information, |, to define the relative efficiency (RE) of MI as RE = (1 + |/|)|, where | is the number of imputations, leading to the conclusion that a small | (≤5) would be sufficient for MI. However, evidence has been accumulating that many more imputations are needed. Why would the apparently sufficient | deduced from the RE be actually too small? The answer may lie with |. In this research, | was determined at the fractions of missing data (|) of 4%, 10%, 20%, and 29% using the 2012 Physician Workflow Mail Survey of the National Ambulatory Medical Care Survey (NAMCS). The | values were strikingly small, ranging in the order of 10| to 0.01. As | increased, | usually increased but sometimes decreased. How the data were analysed had the dominating effects on |, overshadowing the effect of |. The results suggest that it is impossible to predict | using | and that it may not be appropriate to use the |-based RE to determine sufficient |.

27398259

PMC4934387

Journal Article

CC999999/Intramural CDC HHS/United States

CDC Public Access