Related Experiment Videos
Missing data: an introductory conceptual overview for the novice researcher
Maher M El-Masri1, Susan M Fox-Wasylyshyn
1Faculty of Nursing, University of Windsor, Health Education Centre, Ontario, Canada. melmasri@uwindsor.ca
Summary
Handling missing data is crucial for accurate research. This paper introduces novice researchers to common missing data patterns and statistical techniques like imputation and deletion.
Area of Science:
- Research Methodology
- Statistical Analysis
Background:
- Missing data is prevalent in research and can significantly impact study outcomes.
- Improper handling of missing data can lead to biased results and erroneous conclusions.
- Novice researchers often require guidance on addressing missing data effectively.
Purpose of the Study:
- To provide an introductory conceptual overview of missing data issues for novice researchers.
- To explain common patterns and challenges associated with missing data.
- To familiarize readers with various statistical techniques for handling missing data.
Main Methods:
- Conceptual overview of missing data patterns (e.g., MCAR, MAR, MNAR).
- Discussion of common missing data handling techniques.
- Explanation of statistical methods including deletion, mean substitution, regression imputation, and expectation-maximization.
Main Results:
- Identified various patterns of missing data and their implications.
- Described the mechanics and assumptions of different missing data handling techniques.
- Highlighted potential pitfalls and considerations for each method.
Conclusions:
- Understanding missing data is essential for maintaining research integrity.
- A range of statistical techniques are available, each with specific advantages and disadvantages.
- Researchers should carefully select methods based on data patterns and study objectives.