Linda Suskie's Assessing Student Learning documents a wide variety of common assessment errors. They result from the subjective nature of grades in all but the most factual subjects. Many failures point to the need for more objectivity and a better system of accountability, including leniency, generosity, and severity errors; halo, contamination, similar-to-me, and first-impression biases; and that most common of errors, rater drift—that is, the unintentional redefining of scoring criteria as the marker grows tired.