| A | B |
|---|
| trait | any distinguishable, relatively enduring way in which one individual varies from another |
| state | the transitory exhibition of a trait, indicative of a relatively temporary disposition to behave in a particular way |
| reliability | dependability or consistency in measurement |
| measurement of reliability | through the standard deviation of a test that is squared |
| variance | the standard deviation of a test that is squared |
| total variance | derived from standard deviation |
| true variance | test variance plus error variance |
| error variance | the part of the test that does not show consistency |
| test-retest | gold standard of reliability |
| positive reliability | test takers tend to score in the same rank-order in general, even though individual scores may change |
| correlation of stability | correlation of two administrations |
| intervening variable | something not given to all test takers that can influence the score |
| parallel forms | have equal means and variances |
| alternate forms | do not necessarily have equal means or variances but equivalent |
| internal consistency coefficients | compares different items within the same test against each other |
| ICC split half method | compares half of test against the other half |
| ICC inter-item consistency | correlation among all items on a scale |
| homogenous | measures only one trait |
| heterogenous | measures more than one trait |
| Kuder-Richardson | statistical effort to check for multiple split halfs of the same test |
| coefficient alpha | can be used for continuous formats |
| Spearman-Brown formula | allows a test developer to estimate internal consistency reliability from a correlation of two halves of a test |
| dichotomous | items that can be scored right or wrong |
| Likert | scored on a scale with no right or wrong answers |
| inter-scorer reliability | degree of agreement (consistency) between two or more scorers |
| dynamic | measuring something that changes easily |
| static | usually unaffected by situational or cognitive experiences |
| speed test | not enough time to finish/questions easier |
| power test | long enough to attempt all questions but some questions will be too difficult |
| criterion-referenced | attaining a cut-off score is the goal |
| Standard Error of Measurement | a mathematical estimate of the standard deviation of a normal distribution of test scores that would be presumably obtained by a person who took the test an infinite number of times |
| validity | a judgment of how well a test measures what it purports to measure in a particular context |
| face validity | the appearance of the test seems to conform to what you are intending to measure |
| content validity | the test represents all the possible domain it is measuring |
| criterion-related validity | judgment of how adequately a test score can be used to infer an individual's most probably standing on some measure of interest |
| concurrent validity | an index of the degree to which a test score predicts some criterion measure |
| cross validation | used primarily for predictive validity |
| criterion contamination | a state in which a criterion measure is itself, based in whole or in part, on a predictor measure |
| incremental validity | an index of the explanatory power of additional predictors over and above the predictors already in use |
| construct validity | extent the instrument measures a psychological concept such as anxiety or personality attributes |
| convergent | should have high correlations with instruments measuring the same construct |
| discriminant | should have low correlations with instruments measuring different constructs |
| selected response | requires test taker to select from alternate responses (T/F, multiple choice) |
| constructed response | test taker must supply the answer (short answer, essay) |
| rational choice | examiner decides which are good distractors |
| empirical | collect the most used wrong answers and use them for distractors |
| response set | a variety of ways that the student continues to answer a question without actually thinking through the question or answer |
| rearrangement items | sort a group of options into a fixed number of categories |
| ranking items | rearrange items of a series |
| factor analysis | statistical tool to determine if items on a test appear to be measuring the same things |
| item-discrimination index | mathematical formula that measures if high scorers do well on an item and low scorers do not |
| item-response theory | made up of many models that evaluate how a test taker's performance on a test interpreted relative to predicted test performance |
| intelligence | an innate ability to deal with relational, cognitive, emotional, and/or kinesthetic functions |
| predeterminism | an innate degree of general mental ability |
| Charles Spearman | two-factor theory |
| g factor | a general intelligence factor measured by all IQ tests |
| s factor | a specific intelligence factor that does not correlate with all IQ tests |
| Thurstone | tried to find 7 primary ability factors, but each orrelated with the other so g was once again confirmed |
| Gardner | developed 9 types of intelligence |
| emotional quotient | unrelated to IQ yet correlated with academic and social success |
| Cattell and Horn | explored crystallized and fluid intelligence |
| crystallized intelligence | what skills and knowledge we learn/increses with age |
| fluid intelligence | gintelligence that is nonverbal, culture free, independent of learning/declines with age |
| Carroll | three-stratum theory of cognitive abilities |
| top stratum of Carroll | general intelligence |
| second stratum of Carroll | 8 abilities and processes |
| third stratum of Carroll | includes subset factors of each ability found in the second stratum |
| level factors | breaking down an ability to other factors |
| speed factors | how fast the process can be completed |
| Cattell-Horn-Carroll Theory | underlying theory for most IQ tests and other assessment tests used in schools |
| termites | group of children with 130 IQ studied over a long period of time |
| Termen | studied children with high IQ's over a long period of time |
| convergent thinking | one answer |
| divergent thinking | multiple answers/creative thinking |
| Flynn Effect | IQ rises 3/3 per decade so IQ tests must be renormed every 10 years |
| looking glass theory | people perform to expectations of teacher |
