Managing Missing Data with Multiple Imputation: A Common - Sense Approach for Enhancing Data Integrity

Dr. Alaina Steele, Dr. Daniel Hepworth

Abstract: Missing data is a prevalent issue for those conducting research, with implications that extend beyond statistical inconvenience, especially for studies using longitudinal data, as retaining study participants over time can be challenging. Effectively addressing missing data presents a significant methodological challenge that is exacerbated by the absence of a universally accepted approach within the research community (Schober & Vetter, 2020). However, multiple imputation has emerged as a favored approach for its ability to address missing data without compromising the integrity of the study (Rubin, 1987). Despite this, the improper application of this method is fraught with issues. For example, employing multiple imputation under the assumption that data are missing at random (MAR) without first assessing the validity of that assumption may jeopardize the integrity of study findings. This paper offers researchers a practical approach that seeks to enhance data integrity through the application of diagnostic techniques and the inclusion of auxiliary variables during the imputation process, that strengthen the confidence in the MAR assumption.

Keywords: Missing data, multiple imputation, auxiliary variables, longitudinal study, data integrity

How to Cite?: Dr. Alaina Steele, Dr. Daniel Hepworth, "Managing Missing Data with Multiple Imputation: A Common - Sense Approach for Enhancing Data Integrity", Volume 14 Issue 5, May 2025, International Journal of Science and Research (IJSR), Pages: 5-9, https://www.ijsr.net/getabstract.php?paperid=SR25429001939, DOI: https://dx.doi.org/10.21275/SR25429001939