ABSTRACT
Unlike reliability theory, establishing the validity of a scale involves no particular statistical techniques – just common sense, critical evaluation and a healthy level of scepticism. Face validity is, however, necessary in order to encourage participants to take a questionnaire or test seriously and to work hard when attempting it. Content validity is most common in areas such as educational attainment, where learning outcomes are clearly defined, and teachers/students therefore know precisely what the assessment process will cover. The scales that are used to test the construct validity of a novel scale must clearly be the most theoretically relevant ones that can be found. Gathering data from concurrent validity studies is quick and easy, whereas true predictive validity studies involve tracking down individuals months or years later and are thus both time-consuming and expensive. If adding an additional scale to a test battery improves its performance, that scale provides incremental validity.
