PSYX 385 1st Edition Lecture 4Outline of Last Lecture I. 4 Types of Clues (Funder, 2011) (Cont’d S Data)a. Advantagesb. DisadvantagesII. (I) Informant Dataa. Advantagesb. DisadvantagesIII. Behavioral Dataa. Natural Datab. Laboratory Datai. Advantages (Cont’d in Next Lecture)Outline of Current Lecture I. Disadvantages of Behavioral DataII. (L) Life Outcomes: The Result of Personalitya. Advantagesb. Disadvantagesc. A reasonable approach to assessmentIII. Classical Test Theory and Psychometricsa. Understanding reliability and Validityb. Classical test theoryc. ReliabilityCurrent LectureDisadvantages of Behavioral Data1.) situation is contrived: observation changes behaviora. do demands related to Rorschach (remember Ink blot tests) match perceptual demands in life?2.) Link between physiological functions and personality are not fully clear(L) Life Outcomes: The result of personality (e.g., medical records, work history, etc.1.) Advantage:These notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute.a. High psychological relevance2.) Disadvantagesa. Multidetermination: personality is one factor among manyb. Retrospective: people can/do change3.) A reasonable approach to this assessmenta. Employ multiple cluesb. Conduct assessment with purposec. Beware of potential biasesd. Know reliability and validity evidenceClassical Test Theory and PsychometricsReliability and validity: Setting the Stage…1.) many decisions based on tests are ‘weighty’ and difficult to reverse (custody, fitness, etc.)2.) We need…a. Consistent measurement tools and a way to estimate that consistencyb. We need tools with established and demonstrated utilityUnderstanding Reliability and ValidityClassical Test Theory:1.) Observed test score…= T(rue)+E(rror)a. True= construct; error= everything elsei. All personality test scores include errorb. All replicated measurements on a test are equally goodReliability: consistency of replicated measurements1.) test-retest: across time2.) Inter-rater: across examiners3.) Internal consistency: within testa. E.g., split-half, Cronbach’s alpha4.) What do reliability estimates do for us?a. Represent conceptually the ratio of true score variance: observed varianceb. Set upper bound validityc. Allow calculation of Standard Error of Measurement (SEM):i. Estimating accuracy: SEM= the standard deviation multiplied by the square root of 1 minus the rel.5.) Are reliable tests necessarily useful?a. Validity: evidence whether test performs as intendedi. Sources of validity evidence:1. From test content and operations:a. Face validity: items appear valid (upon face value), face validity is not always accurateb. From internal structurei. FFM…should have 5 factors…2. From relations to other variablesa. Construct validity—guided by theoryi. Convergent: test relates to others of same constructii. Discriminant: test does not relate to tests of unrelated constructsiii. Criterion-related/predictive: test predicts outcome criterion (e.g, test of shyness predicts # of friends, test of achievement predicts