1 / 46
Click the card to flip 👆
Terms in this set (46)
transformed scoresex: % shape of distribution changesstandardized scoresex: z-scores shape does not change, label changescomposite scoresmade up of sub-scores ex: GREmeanaverage; the arithmetic average of a distribution, obtained by adding the scores and then dividing by the number of scoresmedianthe middle score in a distribution; half the scores are above it and half are below itmodethe most frequently occurring score(s) in a distributionoutliersextreme values that don't appear to belong with the rest of the data; skews the datavariancestandard deviation squaredstandard deviationa measure of variability that describes an average distance of every score from the meanrangethe difference between the highest and lowest scores in a distributionchi-squareInvolves categorical variables. Looks at 2 distributions of categorical data to see if they differ from each other.correlationsstatistical descriptions of how closely two variables are related (use with numerical data/values) positive, negative, neutralReliabilityconsistency of measurementTrue variancevariability in scores (except natural to be there)error variancevariability due to errorWhat are sources of variance?people, test, test administration, scoringWhat are types of reliability estimates?test-retest, parallel/alternate forms, internal consistency, inter-rater/inter-observerTest-retestgive test, wait adequate amount of time, then give same test ---> then correlate both resultsParallel/Alternate Formsgive test, wait, then test again with slightly different versionInternal Consistencylevel to which all the test items clump together (split-half reliability) measuring one thinginter-rate/inter-observerscoring by multiple people, consistently correlating the scoresValidityActually measuring exactly what you intend to measure: ACCURACYWhat are types of validity?face validity, content validity, criterion validity, construct validityFace Validityhow well a test appears to measure what it is supposed to measure -superficial -can be misleadingContent Validityare the behaviors sampled by the test a representative sample of the attribute being assessed?Criterion Validitydegree to which a test predicts performance on a measure of interestCriterionanother relevant test, behavior, outcome, diagnosis, etcPredictorthe assessment toolValidity coefficientcorrelation between the criterion and the predictorConstruct Validitydegree to which a test is measuring the hypothetical construct or trait it says its measuring -want construct to correlateWhat is the evidence of construct validity?1. predicted age change 2. pre-test/post-test changes 3. variation in distinct groups 4. convergent validity 5. discriminant validityPredicted age changeif construct theoretically changes with age, test scores should change with agepre-test/post-test changesif an intervention changes the construct, then test scores should theoretically change after the interventionvariation in distinct groupsif group membership theoretically changes scores, then different groups should show different scoresconvergent validitytest scores should CORRELATE positively with scores on other previously validated tests, measuring same/similar construct [find tests measuring an existing construct and compare]discriminant validitytest scores should NOT CORRELATE with scores on previously validated tests measuring other constructs