Anonymiced Shareable Data: Using mice to Create and Analyze Multiply Imputed Synthetic Datasets

Abstract

Synthetic datasets simultaneously allow for the dissemination of research data while protecting the privacy and confidentiality of respondents. Generating and analyzing synthetic datasets is straightforward, yet, a synthetic data analysis pipeline is seldom adopted by applied researchers. We outline a simple procedure for generating and analyzing synthetic datasets with the multiple imputation software mice (Version 3.13.15) in R. We demonstrate through simulations that the analysis results obtained on synthetic data yield unbiased and valid inferences and lead to synthetic records that cannot be distinguished from the true data records. The ease of use when synthesizing data with mice along with the validity of inferences obtained through this procedure opens up a wealth of possibilities for data dissemination and further research on initially private data.

Using synthetic data to improve the reproducibility of statistical results in psychological research.

“…In cases where this model can be specified directly, conventional software for MI can be used to draw the synthetic values x ðmÞ syn (e.g., the packages "norm" or "jomo" in R; Quartagno et al, 2019;Schafer & Olsen, 1998; see also Volker & Vink, 2021). Otherwise, the synthetic data can be generated from a sequential model:…”

“…We assume that masked copies x1∗ and x2∗ of the original x 1 and x 2 have been created and added to the data set. In DA-MI P , the masked copies x1∗ and x2∗ are treated as additional predictors, and the synthetic data are simulated from: In cases where this model can be specified directly, conventional software for MI can be used to draw the synthetic values boldxsyn(m) (e.g., the packages “norm” or “jomo” in R; Quartagno et al, 2019; Schafer & Olsen, 1998; see also Volker & Vink, 2021). Otherwise, the synthetic data can be generated from a sequential model: which can be implemented in the “synthpop” package in R (Nowok et al, 2016) or similar software by adding the masked copies as additional predictor variables in the synthesis model.…”

Editorial of the Psych Special Issue “Computational Aspects, Statistical Algorithms and Software in Psychometrics”

“…Volker and Vink [24] outline a workflow for generating synthetic data with the multiple imputation software mice. It was demonstrated in a simulation study that the analysis results obtained on synthetic data yielded unbiased and valid statistical inference.…”

“…It was demonstrated in a simulation study that the analysis results obtained on synthetic data yielded unbiased and valid statistical inference. Volker and Vink [24] argue that the ease of use when synthesizing data with mice, along with the validity of inferences obtained, demonstrates rich possibilities for data dissemination.…”

Informed strategies for multivariate missing data

“…The implement of PPC in MICE (version 3.13.15) is straightforward. A new argument where is included in mice function which allows us to replace the observed data by randomly drawing values from the predictive posterior distribution (Volker and Vink, 2021). Here is an example of generating replications of the observed data.…”