PSYC 3250 Lecture Notes - Lecture 6: Standard Deviation, Classical Test Theory, Random Variable
8 Dec 2015
School
Department
Course
Professor
Document Summary
Things to note: the measurements vary considerably, measurements close to 1. 2 cm occur more frequently than do measurements at either extreme. We could average all of the measurements- and refer to the mean as the true score. Any deviation from the mean could be called error. We can define the true score as the expected value (long term average) of the variable. Errors that are consistently in the same direction. Average out to zero over an infinite number of measurements. E. g. skill specific to particular form of test. In most cases we are interested in lasting, general characteristic of people (e. g. level of risk taking) Thus, in most cases we are interested in the lasting and general characteristics of the individual and the rest of the sources of variability are sources of unwanted inconsistency. A perfect measure would consistently assign numbers to the attributes of people; in practice our measures are never perfectly consistent.
