Psych Assessment PDF

PSYCH ASSESS the more items the higher reliability test can be reliable but not valid test cannot be valid unless it's reliable Reliability consistency accuracy same result from test taker in order to say that it's reliable Types...

PSYCH ASSESS the more items the higher reliability test can be reliable but not valid test cannot be valid unless it's reliable Reliability consistency accuracy same result from test taker in order to say that it's reliable Types of Reliability Test Retest test at 2 diff times pretest posttest time sampling error pearson r temporal consistency stable traits Parallel Forms compare 2 forms of test that have same attributes 12 versions of test but still same item sampling error pearson v formed consistency Internal consistency doitems in a test same measure Split Half Reliability divide into half then score separate a THIS pearman brown formula KR20 dichotomous varying degrees of difficulty Her ex true false KR 21 dichotomous same degree of difficulty  ex speed test Cronbach's coefficient Alpha polytomous several possible answer ex likert scale Inter Rater judges evaluate same behavior observer differences Kappa statistics external consistency note negative relationship outliers extreme sc positive relationship Jet Ability Test Achievement Test previous learning past Aptitude Test potential for learning new skills future Intelligence Test general potential present structured Test self report Projective Test ambiguous Personality Test overt covert dispositions Threats to Internal Validity History occurrence of events events happened to test taker before takingtest  Maturation internal physical changes using subject for long period ex longitudinal Testing effects of pretest to post test practice effect Instrumentation inconsistent use of measurement instrument test administration test materials Statistical Regression tendency of extreme to regress to mean score gave pretest to all but posttest to some Selection no random assign quasi experiment Subject Mortality loss of subjects Selection Interaction family of threats note lewis terman stanford binet john back HTP raymond b cattell 1675 CFIT fluid crystallized florence goodenough draw a person test james cattell mental testing Validity test is appropriate does the test measure what it should measure ex intelligence test does it measure intelligence Types of Validity  criterionValidity criterion test criterion data Predictive Validity time elapse ex entrance exam GPA Concurrent Validity simultaneous v1 ex givetest 4do IIIa Incremental ex projective tests ContentValidity experts judge validity of test Construct Validity test measures what it purports measu abstract variable ex love based on theoretical perspectives Convergent Validity it measures w other related variabl late to it ex test of love fit Divergent Validity discriminant no correlation on variab notbe ex test of love fine Ig'd Face Validity only look at appearance of test note  random error can't be determined systematic error fix errors can be fixed expected contradictory statements always false analytical statements always true falsifiable statements disapprovable by research Theories in Reliability Classical Test Score Theory assumes that each person has true scores that is obtainable if there_ no errors in measurement standardize the test in order to minimize random errors Domain Sampling Model the move items you represent the construct Item Response Theory each item have diff item difficult Reactivity altering behavior due to awareness Hawthorne Effect know they're being studied John Henry Effect control group competition w exp group Experimental Designs Within subjects repeated measures  subjects serve more than I variable Between subjects subjects serve only 1 variable 2 Independent GroupsDesign I independent variable w Kasia 1 dependent Multiple Groups Design more than 2 levels I dependent of.fiffgqendenti Matched Groups Design I independent w 2 levels I dependen the subjects are matched Mixed Design I is within other is between Factorial Design 2 or more independent variable note reliability range estimate acceptable in clinical setting 90 9 reliability range estimate acceptable in industrialsetting 70 alternative hypothesis there is significant differences null hypothesis there is no significant differences Item Analysis set of methods used to evaluate items Item Difficulty people who got the item correct item easiness optimal difficulty halfway between 100 and level of success expected by chance  30 1.70 tend to maximize info about differences among individuals Item Discriminability whether people who have done well item also done well on whole test on Extreme Group Method compare high scorers to low Point Biserial Method correlation on performance on if Eichotoms item and on test Discriminability Analysis identifying total test scores of those who answered correctly in specific item Non Experimental Approaches Phenomenology lived experience low degree of manipulation of antecedent conditions low imposition of units Case studies descriptive record of experiences low ac low to high in Field studies non exp approaches in field natural observation participant observer stu Archival studies existing records reexamined for new purpos Qualitative Research words than numbers note strictness error test takers score on low end of scale leniency error test takers score on high end of scale central tendency error test takers score on center average scal homogenous group similar characteristics all boys or girls Single Blind Experiments only the subject don't know which group they're in control demand characteristics Double Blind Experiments both subject experimenters don't know avoid experimenter bias ANOVA to know the significant difference between 3 or more groups Norm Referenced Test compare your scores to the populatio Class A Tests anyone quiz Class B Tests Rem Resy personality intelligence class C Tests Resy individualized intelligence projective