What is the term for the phenomenon where losing subjects may skew research results, particularly if certain characteristics influence dropout rates?

The phenomenon in question is known as attrition. Attrition refers to the loss of participants from a study over time, which can lead to biased outcomes if the characteristics of individuals who drop out differ from those who remain. When certain groups of participants are more likely to leave the study, it potentially skews the results, as the final sample may no longer represent the original population. This is particularly problematic in longitudinal studies, where understanding changes over time is crucial. For instance, if individuals with specific characteristics—such as lower socioeconomic status or worse health outcomes—are more likely to drop out, the remaining sample may show more favorable results than what would be observed if everyone had stayed.

Mortality could also refer to participant dropout, but in research contexts, it is often used to describe loss due to death rather than attrition from other factors. Maturation refers to natural changes in participants over time, which can affect results but is not specifically about loss of subjects. Demand characteristics involve participants altering their behavior based on perceived expectations of the study, which is unrelated to dropout rates. Thus, attrition is the most accurate term relevant to the impact of lost subjects on research outcomes.