a comparison of mean values for more than two different levels or time points of an independent variable to determine whether they follow a particular mathematical pattern, such as linear, quadratic, cubic, or quartic. For example, a researcher might have a specific prediction about how clients’ average degree of motivation to attend therapy will differ over the course of treatment (e.g., will decrease, then increase, and then decrease again), and he or she could use a polynomial contrast to determine whether this is indeed the case. See also repeated contrast.