Quiz 1 Flashcards
(32 cards)
what are the 3 major strategies in assessment?
1) interviewing
2) observing behaviour
3) psychological testing
When are reliability estimates sufficiently high?
.90 for clinical descision making and .70 for research
what are 3 practical issues regarding the context of the test
1) reading level/education of client
2) length of test
3) training required by psychologist
what are 3 questions related to the adequacy of norms
1) is norm group representative of client
2) is the norm group large enough
3) are there sub-groups or national norms
What are 2 main issues to the degree of error in a test (reliability)
1) human variation (ability is less variable than personality)
2) test methods are imprecise
What are 4 primary methods of obtaining reliability
1) test-retest
2) alternate forms (accuracy of test at the time)
3) interscorer
5) internal consistency (kuder-richardson) for pure measures of single variable
3 issues regarding test-retest reliability
1) practice effects
2) elapsed time between test
3) stability in what is being measure (e.g., anxiety would be inappropriate)
2 strengths of alternative forms relaibility
1) eliminates most carryover or practice effect
2) can examine changes after treatment or changes over time
* avoids many problems from test-retest reliability
2 weakness of alternative forms reliability
1) are the 2 forms actually the same?
2) person may change in between the 2 administrations
2 strengths of internal consistency
1) best technique to determine the reliability of a trait with high degree of fluctuation
2) administered once so it’s a test of the test items not temporal stability
2 weaknesses of internal consistency
1) doesn’t account for “warming up” or fatigues/boredome of the person
2) splitting the test reduces # of items so responses cannot be stablizsed around the mean (longer tests will be more reliable)
2 methods for interscorer reliability
1) have the same client’s responses be scored by 2 indvidiuals
2) have 2 indivdiuals test the client (client is tested twice) and compare their scores
which type or reliability is a) most suitable for highly stable traits and b) ones that are fluctuating and c) make predictions
a) stable: test-retest preferable
b) unstable: internal consistency
c) test-retest
Which is most reliable? Multiple choice or true/false
Multiple choice because T/F format has 50% possibility of the answer
which reliability is the highest and lowest?
a) highest: split-half relability
b) lowest: alternate forms
Is validity or reliability more crucial to test construction?
Validity
What are the 2 steps necessary for test construction:
1) construct must be theoretically evaluated
2) specific operations (test questions) must be developed
What is content validity?
Face validity; relevance of instrucment to construct
What is criterion validity?
Coparing test scores to outside performance
What 2 types of criterion validity are there:
a) concurrent: measure taken at same time (simple and cost effective)
b) predictive: measure taking one after another (appropriate for selection and classification)
What is synthetic validity
Extent to which attributes the test measures is relevaent to skills predicted
Is criterion validity stronger in IQ or perosnalities?
Criterion validity stronger in IQ test because there is more variation in personality
what is construct validity (2 types)?
Degree to which different aspects of a construct should relate to each other (discriminant and convergent validity)
what is sensitivity
% of true positive instrument can identify
