PSYC 510 Exam 1 - Questions and Answers

PSYC 510 Exam 1 - Questions and Answers What is reliability? -degree of consistency of a measure (gives same result each time) What is validity? -degree of accuracy of a measure (is it measuring what its intended to measure) What is observed vs. true score? -Observed (what we see) 1)true score: actual degree the trait characterizes the p. 2)measurement error: score we did not intend to assess What are 2 types of error? -random: fluctuates each time something is measured (low reliability) -systematic: present each time something is measured (validity issues) What are the 3 forms of reliability? -across time (T1 vs. T2) -across different measurement forms -consistency across items within a measure What is interrater reliability? -degree to which raters ratings are correlated -Cohen's kappa: index of agreement across raters What is cronbach's alpha? -mean correlation of all items (to test consistency among items within a measure) What is construct validity? -measure actually indicates a person's true score on a hypothetical construct What are 5 types of evidence to support validity? -generalizability -content validity -structural validity -external validity -substantive validity What is generalizability? -extent to which a measure provides similar results across time, setting, pop. What is content validity? -extent to which content of a measure adequately assess all aspects of construct being measured -content must be relevant & representative What is structural validity? -extent to which dimensionality of a measure reflects the dimensionality of the construct What is external validity? -results of study hold up under new conditions -extent to which scores on a measure correlate w/ relevant factors external to the measure -correlate w/ scores on well-validated measures of same or related construct What is substantive validity? --testing theoretically derived propositions about how the construct in question should function in specific contexts & how it should influence indiv. behavior/ thoughts/ feelings What is discriminative validity? -evidence that a measure is not assessing something it is not supposed to assess -related to construct and content validity What is the Multitrait-multimethod matrix (MTMMM) - 3 types of info? -common method for assessing discriminate validity -correlations b/w several construct that are unrelated in theory (minimum 3) -> each measured by diff methods (# of methods matches # of constructs) What is differential validity? -measure is more valid in assessing a construct for members of 1 group than members of another group Self-report pros & cons -pro: direct method for obtaining info (mood, beliefs); easy to collect -con: ppl misremember, not aware of connection b/w attitudes and behavior, accurately assess internal states, dependent on verbal skills Behavioral measures pros & cons -pros: used w/out ppl's awareness; study automatic behaviors; p. interest -cons: do not know why of behavior; highly situationally dependent; must systematically classify observations; expensive Physiological measures - pros & cons -pros: most direct means of quantifying bio responses; precise measures; assess behavior not under ppl's control -cons: need sophisticated electronic/ mechanical equipment; train operators; much info collected, distinguishing signal from noise; intrusive device; constrain ppl's freedom of movement; p. find task long & boring -> low quality data; difficult to establish construct validity What are explicit vs. implicit measures?