Psychometrics is the science concerned with evaluating the attributes of psychological tests and is typically applied in psychology and other behavioural sciences. Psychological tests are generally applied as a way of assessing un-observable psychological characteristics, mental states or processes that we believe to be reflected in an observable behaviours. Because it can only be assumed that these behaviours represent the mental state/process in question, one is making inferences from the observed about the un-observed .
The science of measuring mental capacities. The term psychometrics is derived from psych, meaning mental or mind, and metrics, meaning measure.
Psychological tests measure observable behaviours so that inferences can be made about un-observable psychological characteristics. The study of psychometrics was made popular by Facnis Galton who defined psychometry as "the art of imposing measurement and numbers upon the operations of the mind" . Galton was interested in creating simple, easy to administer tests of mental capability, which led to the creation of the modern intelligence test. Because it is assumed that the number of correct answers reflects intelligence, psychometricians became concerned with measurements that ensure there is a strong link between what is being measured (test questions) and what is being inferred (intelligence)[1,2]. Psychometric tools are often applied in education as there is an assumption that the tests items are an adequate reflection of learning, competence or skill . Therefore, in high-stakes educational exams, strict precautions are taken to ensure the reliability and validity of assessments.
Reliability is a ratio of the how much of the observed score is due to true score and how much is due to error. Sources of error that are commonly assessed in educational testing is inter-rater (differences between examiners) and test-retest (differences between times the test is taken).
Validity tests the theoretical link between the measurement and what is being measured. "If the scores from a measure seem to be actually measuring the mental state or mental process that we think they are measuring, we say that our inferences of scores on the measure is valid".
The use of psychometrics, or psychometric evaluations, has been criticized due to its positivistic roots and the reductionist nature of these tests . When constructing "simple, easy to administer" tests in educational venues, a great deal of effort is placed in ensuring the tests has high psychometric value. However, through this approach, the measurement may over-simplify and abstract the construct to be measured. For example, if creating a reliable and valid measure of professionalism, the construct of professionalism becomes reduced to the items that are included on the evaluation thus potentially over-simplifying and abstracting the concept.
 Furr R, Bacharach, VR. (2007) Psychometrics: An Introduction. Sage: Toronto
 Shea JA, Fortina GS (2002) Psychometric Methods in Norman GR, Van der Vleuten DPM, Newble, DI (Eds.) International Handbook of Research in Medical Education. Dordrecht: Klewer Academic: Great Brittain. P 97-126
 Green, J. (1996) Are questionnares so bad?. Nursing Inquiry; 3:60