PSY 3307 Study Guide - Final Guide: Takers, Confidence Interval, Revised Neo Personality Inventory
3 Mar 2016
School
Department
Course
Professor
Document Summary
Spearman brown: calculated split half correlation of . 80 for test split randomly in half. Interested in knowing estimated reliability of full length version of the test. The whole test is twice as long as each split half the value of n=2: then sub those values into the equation. Answer = ______ is the best estimate of the reliability of the full-length test. Co-alpha used when test questions have a range of possible answers such as a rating scale, or scales made up of questions with only one right answer. 1. 00 would also be very reliable, negative correlation those who scored high on one test would score low on the second test and those who scores low on the first test would consistently score high on the second test. To better understand amount of error in a test score, use reliability coefficient to calculate standard error of measurement.
