Psych 3402 - Statistics Exam 2 Study Guide

a subset of the population - sample every individual that the research pertains to parameters are the numbers that relate to a population - population A method of investigation used to demonstrate cause-and-effect relationships -multiple groups (minimum of 2 groups) -group make-up is identical (to control for extraneous variables) -one group is unchanged, other group(s) are manipulated - experimental method each manipulation (multi-leveled), something changed by the experimenter - independent variable The measurable effect, outcome, or response in which the research is interested. - dependent variable nominal, ordinal, interval, ratio - Four scales of measurement numbers used purely as labels; eg. social security numbers, gender (female 1, male 2). - nominal elements can be ordered on the amount of the property being measured. Examples: taking first, second or third place; Likert scales. - ordinal also, rank-order, but the distance between the value has meaning (zero is an additional point, but not meaningful). Examples: Fahrenheit scale. - interval rank-order, fixed distance between units, and has fixed origin, has a meaningful zero but has no negative. Example: Length of a piece of wood. - ratio a computed measure of how much scores vary around the mean score - standard deviation Represents the average score a person would obtain on an infinite number of parallel forms of a test, assuming that the person is not affected by taking the test (i.e., assuming no practice or fatigue effect). - true score The discrepancy between an examinee's observed test score and his or her true score. X - T = E -At infinite measurements, the error should equal zero. - error Used to determine if multiple forms of a test are consistent; correlate the scores of each form - parallel forms errors that affect an individual's score through purely chance happening. Examples: guessing, distractions, administration errors - random error errors that consistently affect an individual's score because of some particular characteristic of the person or the test that has nothing to do with the construct being measured. Examples: content sampling, scoring errors, fluctuations in the examinee's state - systemic error ________________ errors reduce the consistency and usefulness of test scores - random errors ____________ errors do not result in inconsistent measurement. - May cause test scores to be inaccurate and reduce their practical utility - systematic errors Used to determine if a test is consistent over time - Administer test, wait, then re-administer. - Correlate the scores - test-retest reliability Used to determine if multiple forms of a test are consistent - Correlate the scores of each form - parallel forms reliability Used to determine if the individual items on a test are consistent with each other Cronbach's Alpha: