ABSTRACT
Both reliability and validity are essential parts of the psychometric properties of a measuring instrument. 1 The reliability of an instrument is concerned with the consistency of measurements: from time to time, from form to form, from item to item, or from one rater to another. On the other hand, the validity of an instrument is usually defined as the extent to which the instrument actually measures “what it is designed to measure” or “what it purports to measure,” that is, it assesses the relevance of an instrument for addressing a study’s purpose(s) and research question(s). Both reliability and validity are context-specific characteristics: for example, researchers are often interested in gauging if a measure remains reliable and valid for a specific culture, situation, or circumstance (e.g., a psychological test might be highly reliable and valid in a population of Caucasian adults but not in one of African American children). The conceptualization and specific definitions of reliability and validity have changed over time, as reflected in the various editions of Educational Measurement (Cronbach, 1971; Cureton, 1951; Feldt & Brennan, 1989; Haertel, 2006; Kane, 2006; Messick, 1989; Stanley, 1971; Thorndike, 1951). Such changes have also been reflected in the most recent edition of Standards for Educational and Psychological Testing (American Educational 398Research Association, American Psychological Association, & National Council on Measurement in Education, 2014). Table 29.1 contains a list of desiderata regarding reliability and validity of instruments that should be followed in any empirical research report.
