Abstract
Generalizability theory (G theory) evolved out of the recognition that the concept of undifferentiated error in classical test theory provided too gross a characterization of the multiple sources of error in a measurement (Cronbach, Gleser, Nanda, & Rajaratnam, 1972). The multiple sources of measurement error in test scores can be readily described: for example, one of many possible test forms might be administered on one of many possible occasions by one of many possible examiners. The choice of test form, occasion and examiner are all potential sources of error. G theory attempts to assess each source of error to characterize the measurement and improve its design.2
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
