3- Psychometric Theories Flashcards
1- What are psychometric theories?
- “Intelligence is what an IQ test measures”
- Theories derived from studying individual
differences in performance on cognitive tests (IQ
tests). - The IQ tests are given to Σ individuals then scored.
Scores are then analyzed – intercorrelated,
creating correlation matrix. - Based on IQ tests, and quantitative data
- The analysis methods are inherent in the theory of intelligence
- The IQ tests scores are then analyzed to find underlying dimensions of cognitive ability (i.e. factors, e.g., verbal ability, reasoning, etc.).
- correlation
- factor analysis
- These underlying dimensions usually form the basis of the resulting theory of intelligence.
- Allows for multi-dimensional perspectives
- Correlations and factor analyses allow us to break down intelligence into
components that are associated with each other (e.g., verbal
reasoning/comprehension) - Caveats :
1. The results of such an approach depend crucially on the
sampling of tests used
2. The resulting psychometric theories are intertwined with the selected mathematical techniques (i.e., factor analysis) used to analyze the correlational data.
Typical questions
What is the structure of human intelligence
What is the importance of general intelligence?
Main theories
* Spearman’s g
* Thurstone primary abilities
* CHC model - most “recent” one
2- Spearman’s g: Spearman’s two factor theory
Spearman
English psychologist
one of the first to develop a
theory of intelligence based
upon psychometrics.
Performed his own experiment
Spearman (1904)
Two types of measures
1- Measures of sensory discrimination:
*discrimination of pitches,
*shades of gray,
*Weights
All sensory measures (correll
average 0.55)
2- Measures of “intelligence”:
*school achievement
*school achievement corrected for age
*teachers’ impressions of students and
*“common sense” as evaluated by an interview
Claimed that these two had a 1:1 correlation
All intelligence measures (correll
average 0.25)
All intelligence measures and sensory (correll average 0.38)
Intelligence is unitary : presents
his theory, the “law of the
universal unity of the intellectual
function”
Believes in interpersonal
difference in intelligence :
“people differ in the amount of g
they possess”
Lay perceptions of intelligence
Spearman’s theory had a big influence on the general public :
“g” = “mental energy”
something people are born with
Individual vary in “g” (i.e., they are “smart” or “dumb”)
“g” (i.e. one’s intelligence) displays itself in many different intellectual tasks because it is innate, it is relatively immune to remediation or improvement.
But what is g?
no clear / definite answer
g = what is common to all tests of
intellectual ability (??)
intelligence is logical reasoning
(Spearman, 1923)
g is neural plasticity or the blood
(Spearman 1927)
g is mental energy (Brody &
Brody, 1976; Gardner & Clark,
1992).
Spearman’s two factor theory of intelligence (1903)
1- g
* general intelligence
* genetic, inborn, stable.
* necessary for all intellectual tasks
g is unchangeable
2- s (specific factor)
* specific ability related to task at
hand
* skills that can be developed –
explains why performances vary
across tests
g contributes to s, not the other way around
* tasks may have various “g loading”
s is modifiable
3- Thurstone’s primary mental abilities 1938
L. L. Thurstone. Psychologist and
psychometrician at the University of
Chicago
Disagreed with Spearman about the
existence of a single, overarching g
factor (Gardner & Clark, 1992).
* Did not believe unitary ‘g’ exists
* Believed in several group factors for different ‘parts of intelligence’
In Thurstone’s view, the mind was
dominated by (p. 82) several “group”
factors: factors responsible for certain
aspects of mental activity (e.g., verbal
ability or numeric ability).
Once these factors were considered,
there would be no need to postulate a
“g”
Thurstone : interest
in factors & factor
analysis
Thurstone (1938) collected data from 240 students who completed 56 ability tests
Results: he finds 7 to 9 factors
Verbal (ex: vocabulary)
Spatial (ex: paper folding)
Memory (ex: digit span)
Reasoning (ex: number series).. sometimes split into deduction and induction
Numbers: (ex: arithmetic reasoning)..sometimes split into numerical and arithmetic
Word fluency: (ex: rapid word finding)
Perceptual speed: (ex: comparing symbols quickly)
Thurstone believed that different
tests lead to different factors:
“ the content of the test is unimportant,
its loading in g is important. If the factors are not the same, its bc their tests loading in g isn’t the same”
“ Human intelligence should be
decomposed into a number of
largely independent
components.”
“ A good IQ tests will reveal
these independent faculties.”
He uses a confirmatory factor
analysis technique
Factor analysis’s role = reveal
the independent factors then,
knowing these factors, we can
develop tests assessing them…..
CIRCULAR REASONING
4- Differences between Spearman (1904) and Thurstone (1938)
Spearman
-intelligence is a unitary capacity
* Believed g was the central
factor, first order
* g contributes to s ability,
but not the other way
around
First order = general g factor
Second order = s factors
Thurstone
-‘‘g’’ doesn’t exist
* Believed broad factors were first order
* Believed g was a second order factor, contributed
to by multiple skillsets
* 7 – 9 factors: Verbal, spatial, memory, reasoning,
numbers, word fluency, perceptual speed
First order = broad (7-9 factors)
Second order = general g factor
5- Vernon hierarchal model 1950 (not really one of the three theories, just parenthesis)
Hierarchical models where
“g” exists, along with the factors
Hierarchical factor theory of
intelligence/cognition
Multi-leveled model
The Structure of Human
Abilities (1950)
(see image)
* g is supported by underlying
factors
6- CHC model: part 1 Cattell
Developmental psychologist
Supervised by Spearman
Cattell tended to view intelligence as being highly shaped by the environment, rather than genetic factors; this was in contrast to Spearman who favoured genetic arguments
Unsatisfied with the notion of a general factor of intelligence
He wished to go beyond the idea of general intelligence (which was associated with g and with
Spearman’s genetic arguments),
He wished to describe specific
intelligences
Created theories of fluid (gf) and
crystallized (gc) intelligence:
Intelligence as a developmental process
-different abilities develop in different
manner, peak, maintain, or decline across the life span in different manner
Cattell introduced the notion that
deterioration in intelligence in adults was not uniform across all tests but differential, - vocabulary, information, verbal
comprehension showing the least decline across the life span (crystallized)
speeded tasks, abstract problem-solving,
unfamiliar performances showing the largest declines (fluid)
The former became referred to as
“crystallized” intelligence and the latter became “fluid” intelligence
Catell’s investment theory
“people invest their fluid
intelligence to acquire specific
skills, strategies, and knowledge
in all kinds of fields” (Cattell, 1963;
Cattell, 1987)
knowledge acquisition / learning,
requires relation perceiving (i.e.
fluid intelligence).
(basically use gf to develop gc)
7- CHC theory: part 2 Horn
Horn was Cattell’s doctoral student
He suggests other forms of
intelligence exist beyond Gc Gf
Based on the results of his factor
analysis, he added:
-general visualization (Gv)
-general speediness (Gs)
Each factor also has subfacets (next)
General visualization (Gv):
Vz [visualization]
S [spatial orientation]
Cf [flexibility closure]
Cs [speed closure]
DFT [adaptive flexibility]
“the processes of imagining the way objects may change as they movein space [Vz], maintaining orientation with respect to objects in space[S], keeping configurations in mind [Cf/Cs], finding the Gestalt amongdisparate parts in a visual field [Cf/Cs] and maintaining a flexibilityconcerning other possible structuring of elements in space [DFT]”
Gs (General Speediness)
Sc (speed copying)
Wf (writing flexibility)
P (perceptual speed)
Gs had almost been added previously by Cattell himself
* Thought Gs was the by product of how difficult the test is to a person, rather than a stable trait. So someone’s speed will vary in difference tasks, based on how difficult the task is to a person.
* He thinks Gs “is measured most purely in simple writing and checking tasks” – in simple tasks, basically
* Reminds the importance of measuring with both unspeeded and speeded tasks
Horn’s Views on g
Horn did not agree with the value of
using g to describe a person’s
overall intelligence
For him, the Gf-Gc theory “questions
the belief that the conglomerate
measured by combining subscores
from a collection of intellectual tests
is the best estimate of intelligence”
Rejects the notion that there existed
a general intelligence, whether
comprised of a single g or two g ’s
(i.e., g f- g c).
He used this to argue for why IQ
tests tend not to correlate as well as
we might expect (ex. WJ and WAIS)
Horn adds more factors:
Believed in subscore method, adding:
* Glr, long term memory retrieval
* Gsm, short-term retrieval
* Ga, auditory processing
* Gq, quantitative, math abilities
* Grw, reading/writing abilities
* CDS, correct speed decision
See image of the Cattell-Horn extended gf-gc model
8- CHC part 3 Carroll’s three stratum model (1993)
- Analyzed 2 000 available datasets (1983 – 1985)
wanted to provide extensions to and
expansions of existing theoretical models - Kept 460 which met critical criteria
- Analyses – exploratory, not confirmatory
Carroll was insistent upon the use of
exploratory rather than confirmatory
techniques because he wished to allow the
“data to speak for themselves” - Further develops the CHC
three-stratum model
included a stratum above - essentially the
same as “g” (next slide)
At the narrow ability level,
Carroll’s ( 1993 ) analyses
revealed about 65 primary,
first-order factors
At the broad ability level,
Carroll’s ( 1993 ) analyses
revealed only 8 broad
abilities
The Cattell-Horn model used
Ga for auditory processing,
whereas Carroll used the
designation Gu
Glr became Gr and Gsm
became Gy (general
memory and learning)
Gs = cognitive speediness;
Gt = processing speed
- Stratum 1: ~65 narrow primary factors
- Stratum 2: 8 broad abilities
Stratum 3: General ability= g
9- CHC: McGrew’s analyses
McGrew used joint confirmatory factor analyses (that
analyzed many results from several intelligence tests)
Found 9 broad abilities (Stratum 2) (all of the ones from the CattellHorn model except CDS) and 36 narrow abilities (Stratum 1) (almost
half of the 65 in the CHC model)
McGrew’s newest version of CHC theory is currently being used to revise the factorial structure of IQ tests and their subtests
CHC theory is continuing to gain momentum as a theory of intelligence, despite a very slow start (it did not start to become popular until decades after it was
proposed)
* Gf is the best predictor of
‘g’: Confirms the popular view of intelligence as problem-solving rather than memorization
See image of current CHC model: most popular model now: best fits the data from the WAIS
See all the different models of the CHC
10- Summary
Early studies on IQ, leading to psychometric theories of intelligence : Show all tests in an IQ battery tend to correlate - tendencyfor all scores to be in the same range for one person.
This correlation was found in many different studies (very robust
finding) and was interpreted by some (like Spearman) as being a general intelligence factor (“g”)
Others (like Thurstone) have tried to define underlying processes or “factors” to explain intelligence.
The distinction between fluid and crystallized intelligence, different aspects of intelligence show different developmental courses, and differing susceptibility to brain injury. This lead to the Gc Gf
distinction.
Currently the CHC model offers the best fit with current psychometric data (though, again, the logic is somewhat circular)
