In designed experiments, the term treatment pertains to manipulation (or observed factors) intended or expected to affect the experimental units, in conditions where the effects of that treatment (or treatments) can be objectively isolated from other factors that determine the observed responses of the experimental units, identified, and measured. The term treatment is used interchangeably with the phrase “a level of independent variable (predictor variable)” or a combination of such levels.  For example, in a study designed to compare the influence of two types of diet: A and B, on the rate of growth of mice, A and B (the two type of diets) can be refer to as treatments.