User talk:Rdfphd

Iterative Norming

Iterative Norming is a form of normative reference testing. Normative reference testing is often used in Psychology, Education and other fields where measures of performance are compared with a standard sample (or normative sample) of individuals. Normative comparisons typically compare the scores of an individual with a collection of scores obtained from others who represent a normative cohort. Cohorts may reference age, gender, geographic location, race, economic status or other factors. When standardization samples are large they may be stratified by one or more of these factors. Standardization samples that are referred to as normative referenced are static samples, taken at one point in time. Consequently, it is necessary to "re-standardize" reference groups from time to time. Statistical methods used with normative reference testing allow evaluators to identify performance as abnormal but do not provide sufficient information to identify the source of abnormality.

Iterative Norming incorporates the scores of all test subjects into the normative sample so that the re-standardization is ongoing. There are several advantages to Iterative Norming. First, all of the individuals who participate in the testing are included in the normative sample. This removes exclusion bias from the standardization group. Second, as the normative sample expands, it is possible to identify sub-groups within the standardization sample. Third, more complete representation of persons who receive the tests for decision making purposes allows for the use of a wider range of inferential statistical analysis such as Bayes and Neural Network applications. Forth, the expanded standardization, combined with a wider range of statistical models, provides opportunity to identify specific disorders and disabilities rather than simply identify abnormal performance.