Ethical Boundaries in Data Cleaning: Issues, Methods, and the Threat of Data Manipulation

Abstract

Data integrity is essential to credible scientific research, yet practices such as data cleaning can lead to ethical concerns when conducted without clear methodological justification. This paper examines the ethical limits of data cleaning by reviewing common data issues, valid cleaning techniques, and the risks of manipulation that may compromise research validity. Using a qualitative literature review, the study finds that data cleaning is ethically acceptable only when supported by statistical reasoning, transparent documentation, and reproducible procedures. In contrast, removing or altering data to reinforce hypotheses or improve results constitutes manipulation and violates research ethics. The failure to distinguish these practices risks misinformation, weakened knowledge development, and diminished academic credibility. The findings underscore the need for strict data integrity practices across all research environments.