Measuring Intelligence Flashcards
The five groups of tests used in the Stanford-Binet Intelligence Scale (5th edition) are:
(a) Fluid Intelligence, Processing Speed, Working Memory, Perceptual Reasoning, Verbal Comprehension.
(b) Fluid Reasoning, Crystallized Intelligence, Working Memory, Quantitative Reasoning, Visual-spatial Reasoning.
(c) Working Memory, Fluid Reasoning, Knowledge, Verbal Comprehension, Perceptual Reasoning.
(d) Quantitative Reasoning, Verbal Comprehension, Working Memory, Fluid Reasoning, Crystallized Intelligence.
The answer was b. See Lecture 7, first slide entitled “Stanford-Binet Intelligence Scale (5th edition)”
The Peabody Picture Vocabulary Test:
(a) Is a picture-based test of reading.
(b) Can be used with people who have significant (but not total) motor impairment.
(c) Is appropriate for people with significant visual impairment.
(d) Is a measure of achievement and not aptitude.
The answer was b. Option a is not the correct answer because the test doesn’t measure reading (each word is spoken aloud by the tester). It’s no use for people with significant visual impairment because they have to be able to see the pictures in the response matrix. It is considered a measure of aptitude rather than achievement (it is not testing the formally-taught content in a specific course). However it can be used with people who have significant motor impairment - they just need some way of communicating which of the four pictures in the response array corresponds with the word spoken by the examiner (which could in principle even be something like eye blinks).
Statement 1: Carroll proposed that intelligence should be thought of as a three layer hierarchy of cognitive abilities.
Statement 2: Crystallized intelligence is generally robust to the effects of aging.
(a) Both statements are true.
(b) Statement 1 true; Statement 2 false.
(c) Statement 1 false; Statement 2 true.
(d) Both statements are false.
The answer was a. See Lecture 7 – slide entitled “Carroll’s hierarchical model”. Crystallized intelligence is described in the slide entitled, “Cattell-Horn model” and it says that it “doesn’t decline with age”.
Statement 1: Traditional tests of intelligence could be used as achievement tests.
Statement 2: Tests of intelligence designed for pre-schoolers can predict the intelligence of outlying individuals later in life.
(a) Both statements are true.
(b) Statement 1 true; Statement 2 false.
(c) Statement 1 false; Statement 2 true.
(d) Both statements are false.
The answer was c. See Lecture 7 slides. Statement 1 is false. Achievement tests refer to things like university examinations – where knowledge of formally taught material is assessed. Intelligence tests wouldn’t be considered to be achievement tests. Statement 2 is true. While pre-schooler intelligence tests don’t predict later intelligence for most people they can be predictive for those at the extremes (e.g. for people with profound cognitive disability).
Statement 1: Spearman used Factor Analysis to support the notion of a single factor theory of intelligence.
Statement 2: Thurstone used Factor Analysis to support the notion of a multiple factor theory of intelligence.
(a) Both statements are true.
(b) Statement 1 true; Statement 2 false.
(c) Statement 1 false; Statement 2 true.
(d) Both statements are false.
The answer was a. See Lecture 7 slides. Statement 1 is true – Spearman in fact invented Factor Analysis in order to inspect correlations between lots of different cognitive ability tests – and found that it was possible to pull out a single underlying factor (i.e. all the tests intercorrelated to some degree). Statement 2 is true. Thurstone also used Factor Analysis – but interpreted his data differently. He extracted multiple factors of intelligence. However these factors were intercorrelated to some degree (and hence his findings aren’t actually empirically inconsistent with Spearman’s findings).
Factor analysis of the Stanford-Binet Intelligence Scale (5th edition) suggests that, at most ages, the test has five factors just as the underlying theory predicts. This represents evidence for:
(a) Test-retest reliability.
(b) Incremental validity.
(c) Construct validity.
(d) Internal consistency.
The answer was c. Showing that the factor structure of the Stanford Binet maps onto the proposed theoretical model that the test is based on is an example of construct validity. Note that incremental validity can’t be the correct answer because it has nothing to do with the factor structure of the test (see Lecture 4) - and the other two options are examples of reliability not validity.
Which of the following has been claimed for Raven’s Progressive Matrices test?
(a) It requires verbal skills to complete.
(b) It is an example of an achievement test.
(c) It correlates highly with academic ability.
(d) It allows the intelligence of people who speak different languages to be compared.
The answer was d. The Raven’s doesn’t require verbal skills to complete (you just have to have some way of choosing between the response options). It is an intelligence test and hence not a type of achievement test (i.e. it’s not assessing formally-taught content). It does correlate with academic ability but these correlations are “not high”. However it does allow the intelligence of people who speak different languages to be compared.
Statement 1: Jensen argued that a key problem with factor analytic approaches to intelligence is that they tend to treat intelligence as no more than a theoretical construct.
Statement 2: Sternberg’s Triarchical Model of intelligence includes an emphasis on planning as a key component of intelligence.
(a) Both statements are true.
(b) Statement 1 true; Statement 2 false.
(c) Statement 1 false; Statement 2 true.
(d) Both statements are false.
The answer was c. See Lecture 7 slides. Statement 1 is false. Jensen was arguing in support of factor analytic approaches to intelligence when he said that intelligence is indeed a theoretical construct – where he went on to say that gravity is also a theoretical construct – but no one has any problem with that. Statement 2 is true. Sternberg’s Triarchical Model is an example of an information processing approach to intelligence, where planning is considered a key component of intelligence arguably overlooked by factor analytic approaches.
Spearman’s g refers to:
(a) What different intelligence tests have in common.
(b) The specific factors assessed by different intelligence tests.
(c) A type of non-parametric correlation coefficient.
(d) A statistic for predicting the effect of changing the length of a test on its reliability.
The answer was a. Spearman’s g maps onto the Full Scale IQ score – it is the intercorrelation found between all cognitive tests. The non-parametric correlation (answer c) is Spearman’s Rho not Spearman’s g. Option d refers to the Spearman-Brown formula.
Statement 1: Tests of vocabulary might reasonably form part of a measure of Crystallized Intelligence.
Statement 2: Carroll’s hierarchical model of intelligence includes Spearman’s g.
(a) Both statements are true.
(b) Statement 1 true; Statement 2 false.
(c) Statement 1 false; Statement 2 true.
(d) Both statements are false.
The answer was a. See Lecture 7 slides. Statement 1 is true. Level of vocabulary is regarded as one classic manifestation of an individual’s crystallized intelligence. Statement 2 is true. Carroll’s hierarchical model has Spearman’s g at its top level.
