-
Be able to identify problems with early measures of intelligence as noted by Wechsler.
- Most early measures of intelligence not appropriate for use with adults
- •i.e., items lacked validity for older population
- •other problems noted by Wechsler:
- •difficulty establishing rapport
- •emphasis on speeded tasks handicapped adults
- •mental age norms did not apply to adults
- •Wechsler objected to use of single intelligence score
- •believed intelligence was assoc.with varied abilities
-
What concepts or innovations did Wechsler introduce to the measurement of intelligence?
- Noteworthy innovations:
- (1) nonintellective factors considered in item selection
- •wide range of tasks offered
- (2)point scale method used to group items
- (3)inclusion of a performance scale
-
What is a point scale? How is it beneficial to intelligence measurement?
- In place of the age scale, items with the same content were placed together into any one of 15 separate tests to create point scales. For example, all vocabulary items were placed together in one test; all matrix items placed together in a second.
- permits the calculation of specific scores for each of the 15 tests.presumably reflects g, one can obtain scores related to each specific content area.
- Point scale concept •major innovation in test construction Binet used age scale format•problem– no partial credit given for items
- •grouping similar items = individual score for each content area
- •allows for analysis of examinee’s strengths and weaknesses
- •point scale considered standard in contemporary test construction
- •Stanford-Binet switched to point scale for 4th edition
-
What is a performance scale?
- Performance scale concept
- •Wechsler disagreed with Binet’s emphasis on verbal content
- •Binet had some nonverbal items –tended to be concentrated at younger age levels
- •nonverbal intelligence score notpossible
- •Wechsler test included entire separate scale devoted to nonverbal intelligence
- •Major benefits
- •could now directly compare verbal and nonverbal intelligence through use of one test
- •scales expressed in comparable units
- •scales standardized on same sample
- •ability to identify processing deficits
-
Major Flaw Wechsler
- •poor standardization
- •non-representative sample of Caucasians, primarily from New York area
- •led to revision of test in 1955 – Wechsler Adult Intelligence Scale (WAIS)
- •WAIS-R (1981)
- •WAIS-III(1997)
-
Psychmetric properties of Wais
- Standardization sample:
- •2200 adults divided into age bands
- •200 subjects per age band (16-69)
- •100 subjects per age band (70-90)
- Reliability
- •test found to be highly reliable
- •avg. split-half reliability estimates = .98 for FSIQ, .97 for VIQ, .94 for PIQ
- SEM’s:•FSIQ = 2.29
- •VIQ = 2.50 •PIQ = 3.75
- •95% confidence interval = 2 SEM’s above and below mean
- •1 SEM = 68% confidence interval
- •e.g.,FSIQ of 105•95% of time, subject’s score will
- occur within a range between approx. 100 - 110
|
|
