Multiple imputation: a primer

By Joseph L. Schafer. 1999

How researchers see this paper

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

Citations in the corpus

Informing sequential clinical decision-making through reinforcement learning: an empirical study (2010, methodology): “1 below, we apply multiple imputation (Little and Rubin 1987; Rubin 1996; Schafer 1999) to use all the available data (individuals with both complete and incomplete data) in order to reduce bias.”
Developing prediction models for clinical use using logistic regression: an overview. (2019, background): “Further information on multiple imputation techniques can be found through the following references (3,11,13,14).”
Multiple minority stress: The role of proximal and distal stress on mental health outcomes among lesbian, gay, and bisexual people of color (2019, result): “It is suggested that a missing data rate of less than 5–10% is inconsequential and does not bias the results (Bennett, 2001; Dong & Peng, 2013; Schafer, 1999), and that patterns of missing data have more of an influence on results than the proportion of missing data (Tabachnick & Fidell, 2013).”
The role of information asymmetry in the choice of entrepreneurial exit routes (2014, result): “…of our analysis with three imputations used to simulate plausible values for ourmissing controls, which showed a non-nestedmissingness pattern, (Schafer, 1999) were not substantially different from the original results we obtained (which are reported in Section 4.3; all of the hypothesized…”
Big Math for Little Kids: The Effectiveness of a Preschool and Kindergarten Mathematics Curriculum (2015, methodology): “These methods yield efficient and consistent estimates in the presence of data that are missing at random (Enders, 2010; Schafer, 1999), which we verified was the case for this study.”
A model to estimate the lifetime health outcomes of patients with Type 2 diabetes: the United Kingdom Prospective Diabetes Study (UKPDS) Outcomes Model (UKPDS no. 68) (2004, methodology): “Standard methods for combining the results of multiple imputations were then employed [31], including a bias correction to adjust for the fact that a (relatively) small number of multiple imputa1758 P.”
Effects of Comprehensive Lifestyle Modification on Diet, Weight, Physical Fitness, and Blood Pressure Control: 18-Month Results of a Randomized Trial (2006, methodology): “Otherwise, they were imputed by using a single imputation hot-deck procedure (26) that used values from participants in the advice group who were similar at baseline with respect to variables that predicted future missing data.”
The proportion of missing data should not be used to guide decisions on multiple imputation (2019, background): “Varying guidance exists; in the literature, 5% missingness has been suggested as a lower threshold below which MI provides negligible benefit [16].”
Loneliness in the UK during the COVID-19 pandemic: Cross-sectional results from the COVID-19 Psychological Wellbeing Study (2020, unknown): “As less than 5% of the data were missing pairwise and listwise deletion were implemented [69,70].”
Joint trajectories of behavioral, affective, and cognitive engagement in elementary school (2017, methodology): “NORM uses an iterative method based on EM algorithm to impute missing data by drawing values from the conditional distribution of the variables, given their observed data (see Schafer, 1999 for technical details).”
Missing data: our view of the state of the art. (2002, methodology): “Reviews of MI have been published by Rubin (1996) and Schafer (1997, 1999a) .”