Evaluators using instruments such as questionnaires and tests should recognize that scores on measures in the affective and cognitive domains are assumed to have errors of measurement. A person's obtained score (e.g., 52 on an algebra test) is likely to be higher or lower than the person's true score (e.g., 55.25). The higher the reliability of an instrument, the closer the obtained score will be to the true score. Highly reliable instruments have a relatively low standard error of measurement, which is a mathematical index of both the reliability and the size of measurement error.