an experimental design having multiple independent variables in which the number of measurements or observations obtained is different for each condition under study. Although balanced designs generally are preferred for their greater power in statistical analyses, unbalanced designs nonetheless may arise due to participant attrition or other unavoidable factors. For example, if a researcher is investigating how sleep and diet influence academic performance and only 35 of the 40 college undergraduates recruited to participate were able to do so through the full term of the project, certain data will be missing for the five people who left the study prematurely and the design will become unbalanced.