The Impact of Missing Data on the Ethical Quality of a Research Study

• Published in: Handbook of Ethics in Quantitative Methodology pp. 357 - 381
• Main Authors: Enders, Craig K., Gottschall, Amanda C.
• Format: Book Chapter
• Language: English
• Published: United Kingdom Routledge 2011; Taylor & Francis Group
• Edition: 1
• Subjects: Psychological methodology; Cost Utility Model; Missing Data Mechanism; MCAR Mechanism; Multiple Imputation; Missing Data Analysis; Mar Mechanism; Missing Data; Va Ri; Auxiliary Variables; Missing Data Handling; MNAR Mechanism; Incomplete Variables; IRT Model; MAR Assumption; MNAR Analysis; MNAR Model; Ethical Continuum; Imputation Phase; Automated Essay Scoring; Missing Data Rate; Jar; Ethical Issues Related; Missing Data Issues; Pattern Mixture Model
• DOI: 10.4324/9780203840023-18
• ISBN: 9781848728554; 1848728557; 9781848728547; 1848728549

The purpose of this chapter is to explore the impact of missing data on the ethical quality of a research study. In doing so, we borrow heavily from the work of Rosenthal (1994) and Rosenthal and Rosnow (1984). The overarching principle of Rosenthal's (1994) work is that ethics is closely linked with the quality of a research study, such that high-quality studies are more ethically defensible than low-quality studies. Missing data pose an obvious threat to quality at the analysis stage of a study (e.g., when a researcher uses a missing data handling technique that is prone to bias), but ethical issues arise throughout the entire research process. Accordingly, we explore the linkage between quality and ethics at the design and data collection phase, the analysis phase, and the reporting phase. In doing so, we also apply Rosenthal and Rosnow's (1984) cost-utility model to certain missing data issues (see also Rosnow & Rosenthal, Chapter 3, this volume). In this framework, the costs associated with a study (e.g., potential harm to participants, time, money, resources) are weighed against its utility (e.g., potential benefits to participants, science, or society). As it relates to ethics, a study is more defensible when its benefits exceed its costs.