Upgrade to remove ads
Only $2.99/month

Intellectual assessment exam #1

Terms in this set (53)

Crystallized intelligence (Gc)
acquired skills and knowledge dependent on exposure to a particular culture, formal and informal education (knowledge I've attainted from experience in the world, education, definitions of words)
Fluid intelligence (Gf)
nonverbal, relatively culture-free, and independent of specific instruction, not dependent on life experiences (how well does my brain function in processing new information)
Reliability
consistency of a measurement of a given score, determined by true score variance/total variance (how consistently does the test represent intelligence)
Internal reliability
consistency across test items, how well the items in the test correlate with each other (does this test measure one thing? homogeneity, do all items tap same construct)
Test-retest reliability
consistency over time, evaluates stability of a measure (does the test give consistent results at different times, ex. test stability of personality test, assumption that construct that we're measuring is consistent overtime)
Alternate form reliability
consistency across alternate forms (evaluates relationship between different forms of a measure, correlations between two different tests)
Interrater reliability
consistency across raters (evaluate level of agreement between raters on a measure)
Test score reliability
internal, test-retest, alternate form, inter-score/interrater reliability
Internal consistency
inter-item correlation in multiple test items (high for IQ tests, low for MMPI-2)
Standard error of measurement
estimate of the amount of error in one's obtained score
Test score validity
content, construct, and criterion-related evidence
Validity
does the test measure what it says it measures (like reliability, both refer to scores not tests)
Content validity
how adequately do test items represent the constructs, validity based on evaluation of the content covered by items of the test
Lawshe and content validity
items are essential if more than half of raters indicate that items are essential, other options: useful but not essential, not necessary for job performance
Construct validity
convergent and discriminant/divergent validity
Convergent validity
degree to which two measures of constructs should be related
Discriminant/divergent validity
degree to which two measures of constructs should not be related
Criterion-related validity
measures how well one measure predicts an outcome for another measure (judgment of how adequately a test score can be used to infer an individuals' most probable standing on some measure of interest - the criterion)
two types of criterion-related validity
concurrent and predictive validity
Concurrent validity
index of degree to which test score is related to criterion measure obtained at same time (concurrently)
Predictive validity
index of degree to which test score predicts some criterion or outcome measure in the future
WAIS-IV and criterion validity
has satisfactory criterion validity
Face validity
how relevant test items appear to be (high- self-report personality measures, low- projective personality measures)
Inductive decision making
bottom-up, open-ended and exploratory process, begins with observations, symptoms, and possible patterns (report of test score, low performance on processing speed index, but otherwise above average intellectual functioning)
Deductive decision making
top-down, structured and confirmatory, starts with a (general concept or idea) hypothesis or previous diagnoses then gather data to confirm or disconfirm the hypothesized diagnosis
Representation bias
some observations, signs, and symptoms are more strongly related to certain diagnoses (clinicians often make a leap from a specific symptom[s] to a representative disorder without consideration of the specificity of the symptom)
Availability bias
information that is most readily accessible or most easily recalled influences the decision at hand (primacy bias - could be first information presented by the person, recency bias - most recent information)
Hindsight bias *
overestimating probability of a diagnosis due to prior diagnosis
Regret bias
let's go ahead and diagnose and treat because risk of missing it is dangerous
Confirmatory bias
give greater attention to confirmatory evidence while undervaluing disconfirming evidence (notes concern with only considering client report and current test scores)
Diagnostic bias
there is always a diagnosis
Informed consent
unless testing is mandated by law, part of routine educational, employment activity, person is being evaluated to determine if they are competent to make decisions, client acknowledges consent, therapeutic assessment
Processing speed
symbol speech, coding, opt: cancellation
Processing speed tests
rate of test taking, perceptual discrimination, speed of mental operation, psychomotor speed, attention and concentration, short-term visual memory, visual-motor coordination, numerical ability, cognitive flexibility, self-monitor
Working memory
digit span, arithmetic, opt: letter-number sequencing
Working memory tests
short-term memory, ability to sustain attention, numerical ability, encoding ability, auditory processing skills, cognitive flexibility, self-monitor
Verbal comprehension
similarities, vocabulary, information, opt: comprehension
Verbal comprehension tests
application of verbal skills and information to new problems, process verbal information, think with words, crystallized knowledge, cognitive flexibility, self-monitor
Perceptual reasoning
block design, matrix reasoning, visual puzzles, opt: figure weights, picture completion
Perceptual reasoning tests
ability to think in terms of visual images and manipulate them with fluency, relative cognitive speed, interpret or organize visually perceived material with a time limit, cognitive flexibility, nonverbal ability, form abstract concepts and relationships without use of words, fluid reasoning, self-monitor
Confidence Interval
band or range of person's true score
we cannot measure true score so estimate
reliability
true score (variance)
consistency/dependability
Split half or spearman-brown reliability coefficient
divide at halfway point, obtained by correlating two halves of items from the same test (varies for different populations)
Coefficient alpha (Cronbach's alpha)
mean of all split-half correlations, every possible combination
Kuder-Richardson reliability coefficient
used for items with dichotomous answers (yes/no, T/F) or heterogenous tests were split-half methods must be used, generally lower in value than split-half coefficients
Type of internal reliabilities
split half or spearman-brown reliability coefficient, coefficient alpha or cronbach's alpha, and kuder-richardson reliability coefficient
test-retest reliability is highest for which WAIS-IV index score?
Verbal comprehension (overall WAIS-IV provides stable FISQ score over time)
Which domain the has lowest test-retest reliability in the WAIS-IV?
Processing speed (.76 for age 30-54)
Improving diagnostic decision-making through knowledge & experience
see a wide variety of cases, broad training, experience training with mentors and feedback
Improving diagnostic decision-making in assessment setting
obtain complete history, consider base rate, confirmatory and disconfirmatory data, consider all evidence and weight given to each separate finding when integrating results
verbal tests
verbal comprehension and working memory
nonverbal tests
perceptual reasoning and processing speed
YOU MIGHT ALSO LIKE...