This method determines whether a detectable difference exists between two samples. The difference may be overall in a specific attribute, such as flavor, texture, color, odor, etc. The difference may result from a change in ingredients, processing, packaging, or storage. Samples should be different for one attribute only, and the same for all others. The cause of the difference is not identified. The panelist is asked to determine which two of the three samples are the same. The degree of statistical significance will not indicate either the magnitude or the direction of a detected difference. This test may also be used in selecting panelists and monitoring their performance.