PSYC37H3 Lecture Notes - Lecture 4: Concurrent Validity, Factor Analysis, Face Validity
Document Summary
Think of a time when you wrote an exam, and when you received your grade you felt it was a poor representation of your actual knowledge Reasons might be about: the test itself, the test-taker, the environment, how the test was scored, these are all sources of error variation. Reliability defined: reliability refers to the accuracy, dependability, consistency, or repeatability of test results. Refers to the degree to which test scores are free of measurement errors. The less error the test has, the more reliable the test is. Very hard to measure the true score in psych, but it"s the goal: assumption: errors of measurement are random. As opposed to systematic: rubber yardstick analogy. Randomly stretches or shrinks each time you take a measurement. Leads to variability in the measure: note that the test, itself, doesn"t change, it"s the obtained score/measure that varies, sampling theory suggests that the distribution of random errors is bell-shaped, or normally distributed (recall: normal curve)