reliability - RELIABILITY OBJECTIVE To increase...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
RELIABILITY OBJECTIVE: To increase understanding of the concept of reliability and to provide first hand experience with the calculation of interscorer and test-retest reliabilities. GENERAL INFORMATION - RELIABILITY The term reliability as used in the field of psychological testing refers to consistency in measurement. Classical Test theory or "true score" theory proposes that a score on a test consists of two parts. The first part is the "true" amount of whatever is being measured (e.g., intellectual ability on an intelligence test, or depression on a depression scale). The second part consists of unsystematic error of measurement (e.g. time of day the test was given, the particular sample of questions chosen for the test, the lighting in the room, a faulty pen, etc.). All of these extraneous variables may affect test takers differentially. This part of the score is referred to as the error portion of the score. A reliability coefficient is a numerical index of reliability that expresses the ratio between the "true" score on a test and the total variance and basically gives an indication of how free the test is from errors in measurement. In test theory, a "true score" on a test can be thought of as the average (hypothetical) score that would be obtained if an individual were to take the test an infinite number of times. In other words, the true score is the remaining part of the score once it has been stripped of random error. In testing, one goal is to determine true score difference. What we know is: X = T + e where: X represents an observed score, T represents a true score, and e represents an error score (the part of the score due to random, nonsystematic influences on the test). We only know an individual's observed score (X), which is their measured score on a test. With one equation and two unknowns, we cannot directly compute their true score, which is what we really want to know. Stated another way, we only know a person's measured or observed score; we want to know their true score. But, by definition, we cannot ever know a true score. So we must somehow estimate error and then estimate a true score.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 11/16/2011 for the course PSYCH 3900 taught by Professor Staff during the Fall '10 term at Valdosta State University .

Page1 / 6

reliability - RELIABILITY OBJECTIVE To increase...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online