PSYC314 Lecture 8: Intelligence Tests - Lecture Notes

Lecture Overview

• Different Approaches to Intelligence Testing
• Individual versus group testing
• Definitions: Intelligence versus Achievement versus Aptitude
• Intelligence Tests
  • Stanford-Binet Test Battery
  • Wechsler Intelligence Test Batteries (e.g., WAIS-IV)
  • Tests for special groups (e.g., Raven's Progressive Matrices)

Individual vs. Group Tests

• Intelligence tests can be:
  • Individually administered
  • Group administered
• Depends on the purpose/goal of testing
• Individual tests give extra clinical information:
  • (a) How does the person answer
  • (b) Test behavior (do they give up easily on tasks)
  • Maximize Motivation
  • Make allowances for fatigue and handicaps
• Essential for:
  • Young Children (esp. preschoolers)
  • ANY clinical assessment involving individuals with mental health concerns, acquired head injury, neurodiversity (e.g., ADHD, ASD diagnoses), learning disorders, intellectual disability, and/or developmental delays

Group Testing Advantages

• Ease and efficiency of scoring and administration.
• Less skill and training are required on the examiner's part
• Quite reliable and standardization samples are usually large

Group Testing Disadvantages

• Hard to maintain motivation and rapport and assess
• Hard to monitor factors such as anxiety
• Limited response choice, e.g., multiple-choice items.
• The assumption is that the tests are equally applicable to all subjects.
• Some subjects get bored because the test is too easy, or frustrated because the test is too hard.

Computerized Testing

• Computerized tests tend to lag well behind recent research findings about certain tests
• Ethical concerns have not yet caught up to the technology
• 20% of the population dislike computers
• It still misses non-verbal and emotional behaviors such as anxiety that can moderate test performance

Definitions

• Intelligence tests: attempt to predict future performance; predict abilities that are general and broad.
• Aptitude tests: typically predict more specific abilities such as math, science, or music.
• Achievement tests: attempt to assess what a person has learned following a specific course of instruction.
• Aptitude tests: attempt to evaluate a student’s potential for learning rather than how much a student has already learned.

Measuring Intelligence: Stanford-Binet Intelligence Scale

• Oldest of the modern tests of intelligence; the very first test, developed by Alfred Binet, used some key principles:
  • Age differentiation: Binet looked for tasks that could be successfully completed by 2/3 to 3/4 of children in a particular age group, a smaller proportion of younger children, and a larger proportion of older children
  • General mental ability: conceived of intelligence as a unitary factor, not separate mental abilities, which can be represented by a single score

Stanford-Binet Intelligence Scale

• 1905 Scale: 30 tasks or tests of increasing difficulty; no measuring unit – just categorized people very roughly
• 1908 Scale: grouped items according to age; could now describe an individual in terms of “mental age”

1916 Stanford-Binet Intelligence Scale

• Developed by L.M. Terman of Stanford University
• First time the concept of “intelligence quotient” was used:
• Formula: $IQ = \frac{MA}{CA} \times 100$
  • If child performs to age – $IQ = 100$
  • If 8yr has MA of 10 – $IQ = 125$
  • If 16yr has MA of 20 – $IQ = 125$

Stanford-Binet Intelligence Scales, 5th Ed

• 5th edition; released in 2003
• Relates to Carrol’s Model (but only measures 5 of the 9)
• 10 subtests
  • 5 factors x 2 domains (non-verbal and verbal)
  • Fluid reasoning, knowledge, visual-spatial processing, working memory, quantitative reasoning
• FSIQ, VIQ, NVIQ, 5 Factor scores (m=100, SD=15)
• 10 subtest scores (m=10, SD=3)

Stanford-Binet Intelligence Scales, 5th Ed (cont.)

• Routing procedure (to cut down on administration time and examinee fatigue)
  • Determines starting point
• Ages 2 to 85 years
• Level 1 is for pre-schoolers and Level 6 differentiates between gifted and very gifted
• Nonverbal IQ domain does not require expressive language
• Expert panel examined fairness related to gender, race, ethnicity, disability, religion

Stanford-Binet Intelligence Scales - Subscales

• NON-VERBAL:
  • Object Series/Matrices
  • Picture Absurdities
  • Quantitative Reasoning
  • Form Patterns
  • Working Memory
• VERBAL:
  • Knowledge/Vocabulary
  • Verbal Analogies
  • Quantitative Reasoning
  • Position and Direction
  • Last Word

Measuring Intelligence: Wechsler Scale

• Wechsler’s IQ scales have represented the most widely used measure of intelligence for decades.
• Introduced in 1939, the scales progressively replaced the Stanford adaptation of Binet’s test.
• The Wechsler test could also be used to measure adult IQ and was valid on large and representative samples.
• Wechsler designed a specific version of his test for children (aged 5-16) called the Wechsler Intelligence Scale for Children (WISC) and the Wechsler Preschool and Primary Scale of Intelligence (WPPSI) for preschool children.

Wechsler’s Concept of Intelligence

• Intelligence:
  • “The capacity of the individual to act purposefully, to think rationally, and to deal effectively with his environment”
  • “Intelligence is a global construct (g) which can also be categorized by the sum of many specific abilities”. -- Wechsler, 1944

WAIS-IV

• The WAIS-IV is an individually administered assessment of intelligence and cognitive abilities suitable for adults aged 16 – 90 years
• Average testing time for the Full-Scale IQ is 60 – 90 minutes
• The WAIS-IV is recommended for the assessment of adult intelligence in clinical, educational, and research settings

Development of the Wechsler Scales

• Wechsler-Bellevue I 1939 ages 7-69
• WAIS 1955 ages 16-64
• WAIS-R 1981 ages 16-74
• WAIS-III 1997 ages 16-89
• WISC-V 2014 ages 6-16
• WPPSI-IV 2012 ages 2-6
• WAIS-IV 2008 ages 16-90
• WAIS-V TBR 2024
• The Wechsler Intelligence Scale for Children-Fifth Edition (WISC-V): new test structure to reflect current models
• The Wechsler Pre-school and Primary Scale of Intelligence-Fourth Edition (WPPSI-IV)

The WAIS-IV Sub-tests

• Full Scale IQ
  • Verbal Comprehension Index (VCI)
    • Core Subtests:
      • Similarities
      • Vocabulary
      • Information
    • Supplementary Subtest:
      • Comprehension
  • Working Memory Index (WMI)
    • Core Subtests:
      • Digit Span
      • Arithmetic
    • Supplementary Subtest:
      • Letter-Number Sequencing (16-69yrs)
  • Processing Speed Index (PSI)
    • Core Subtests:
      • Symbol Search
      • Coding
    • Supplementary Subtest:
      • Cancellation (16- 69yrs)
  • Perceptual Reasoning Index (PRI)
    • Core Subtests:
      • Block Design
      • Matrix Reasoning
      • Visual Puzzles
    • Supplementary Subtests:
      • Figure Weights (16-69yrs)
      • Picture Completion

Factor Indices: Verbal Comprehension Index (VCI)

• Verbal conceptualization, knowledge, and expression
• Answering oral questions that measure factual knowledge, word meanings, reasoning, and the ability to express ideas in words
• Involves retrieval of verbal information from long-term memory and reasoning with verbal information
• (Kaufman & Lichtenberger, 1999, p. 121) (Lichtenberger & Kaufman, 2009, p. 219)

Verbal Comprehension Index (cont.)

• In order to successfully complete the VCI subtests, one needs to be able to:
  • Listen and understand what is communicated (i.e., receptive language).
  • Retrieve verbal information (e.g., factual/general knowledge, word meanings) from long-term memory/storage.
  • Use verbal reasoning and concept formation skills.
  • Use words and sentences to communicate (i.e., expressive language).

Verbal Comprehension Index

• The VCI subtests assess skills and knowledge that are normally acquired through education and exposure to the mainstream culture.
• Subtests require the ability to reason with previously learned information.
• This Index is often conceptualized as measuring 'Crystallized Intelligence' (Kaufman & Lichtenberger, 2009).

Verbal Comprehension Index (cont.)

• Crystallized intelligence is commonly defined as the sum of everything an individual has learned (i.e., a fund of information, relationships, and mental skills developed through education, experience, and practice).
• The subtests of the VCI are less sensitive to impairments following head injury (e.g., traumatic head injury) and disruptions caused by mood disorders (e.g., depression and anxiety).

Factors that can affect VCI subtest scores: (Kaufman & Lichtenberger, 2009)

• Exposure to written text (e.g., newspapers, books, internet).
• Range of interests.
• Alertness to the environment.
• Exposure to the mainstream culture.
• Foreign language background (i.e., English as a second language).
• Intellectual curiosity and striving.
• Richness of early environment.
• Negativism (e.g., arguing against the items of the test).
• Overly concrete thinking.
• Development of conscience or moral sense.
• Understanding of social norms.

Perceptual Reasoning Index

• Application of reasoning with nonverbal, visual stimuli including the ability to “analyze and synthesize abstract visual stimuli” (Wechsler, 2008, p. 13-14).
• Nonverbal reasoning and visual-motor coordination: integrating visual stimuli, reasoning nonverbally, and applying visual-spatial and visual-motor skills (Kaufman & Lichtenberger, 2009).
• Involves solving problems not normally taught in schools using abstract, novel stimuli.
• Many subtests are timed or time-limited.

Perceptual Reasoning Index (cont.)

• Successful completion of the PRI subtests requires the ability to:
  • Apply reasoning skills on visual, nonverbal stimuli.
  • Apply reasoning skills on visual quantitative information.
  • Apply reasoning skills on conceptually related concrete and abstract visual stimuli.
  • Integrate visual elements to create a model.
  • Visual discrimination and attention to detail.

Factors that can affect PRI subtest scores: (Kaufman & Lichtenberger, 2009)

• Executive functions.
• Visual-perceptual problems.
• Working under time pressure.
• Obsessive concern over detail.
• Ability to respond when uncertain.
• Color blindness (e.g., matrix reasoning subtest).
• Overly concrete thinking.
• Flexibility in thinking.
• Motivational level.
• Persistence.
• Impulsivity in responding.
• Ability to self-correct.

Working Memory Index

• Measures cognitive abilities involved in the registration and holding of information (i.e., STM) and the mental manipulations of information being held (i.e., working memory).
• Involves auditorily presented verbal and verbal-quantitative stimuli.
• Involves number ability and sequential processing.
• Requires a good non-distractible attention span for success. (Lichtenberger & Kaufman, 2009, p. 225)

Working Memory Index (cont.)

• In order to successfully complete the subtests of the WMI, one needs to be able to:
  • Attend to the verbal stimuli (e.g., number sequences)
  • Discriminate between auditory stimuli.
  • Store the verbal stimuli in memory.
  • Retrieve the verbal stimuli from memory.
  • Engage in mental flexibility (i.e., divide attentional resources between two or more streams of thoughts or goals).
  • Apply mathematical skills involving addition, subtraction, multiplication, and division.
  • Note that the subtests of the WMI are particularly sensitive to conditions that disrupt attention, memory, and concentration.

Processing Speed Index

• Measures processing speed with nonverbal, visual stimuli (i.e., visual processing speed, motor processing speed, and visual-motor processing speed).
• Also requires visual perception and discrimination, attention to detail, multitasking (mental flexibility), fine motor coordination, organization skills, working memory, executive function abilities. (Lichtenberger & Kaufman, 2009, p. 229)

Processing Speed Index (cont.)

• Successful completion of the subtests of the PSI require:
  • Visual-motor processing speed
  • Fine-motor coordination (i.e., those required in handwriting).
  • Visual perception and visual discrimination.
  • Visual scanning.
  • Attention to visual detail.
  • Mental alertness.
  • Mental flexibility.

Factors that can affect PSI subtest scores: (Kaufman & Lichtenberger, 2009):

• Compared to VCI subtests, the PSI subtests are more disrupted by head injury.
• Mood Disorder (e.g., Anxiety, Depression)
• Distractibility.
• Learning Disabilities.
• ADHD.
• Visual-perceptual problems.
• Working under time-pressure.
• Fine motor coordination involved in handwriting.
• Motivation level.
• Persistence.
• Obsessive concern over accuracy and detail.

WAIS-IV Norms

• Metrics for Standard Scores
  • Scaled Subtest Score: Mean = 10, S.D = 3, Range = 1-19
  • IQ Score: Mean = 100, S.D = 15, Range = 40-160
  • Index Score: Mean = 100, S.D = 15, Range = 40-160

WAIS-IV Score Descriptive Classification (L & K (2009))

• 130 to 140+: Very Superior, 130+ Upper Extreme/Normative Strength, 131+
• 120 to 129: Superior, Above Average/ Normative Strength, 116 To 130
• 110 to 119: High Average, Average Range/ Normal Limits 85 to 115
• 90 to 109: Average, Average Range/ Normal Limits 85 to 115
• 80 to 89: Low Average, Below Average/ Normative Weakness 70 to 84
• 70 to 79: Borderline, Below Average/ Normative Weakness 70 to 84
• ≤ 69: Extremely Low, Lower Extreme/ Normative Weakness ≤ 69

Evaluation of the WAIS-IV: Reliability

• Average WAIS-IV Reliability
  • Subtest Internal Test-Retest
  • Vocabulary .94 .89
  • Similarities .81 .87
  • Information .93 .90
  • Comprehension .87 .86
  • Arithmetic .88 .83
  • Digit Span .93 .83
  • Letter-Number Seq. .88 .80

Evaluation of the WAIS-IV: Reliability

• Average WAIS-IV Reliability
  • Subtest Internal Test-Retest
  • Visual Puzzles .89 .74
  • Figure Weights .90 .77
  • Picture Completion .84 .77
  • Block Design .87 .80
  • Matrix Reasoning .90 .74
  • Coding -- .86
  • Symbol Search -- .81
  • Cancellation -- .78

Evaluation of the WAIS-IV: Reliability

• Average WAIS-IV Reliability
  • IQ or Index Internal Test-Retest
  • Full Scale IQ .98 .96
  • VCI .96 .96
  • PRI .95 .87
  • WMI .94 .88
  • PSI .90 .87

Evaluation of the WAIS-IV: Validity

• WAIS-IV Manual presents multiple Criterion Validity studies.
• WAIS-IV Manual presents numerous sources of evidence for Construct Validity (e.g., Factor analytic studies have found support for the 4-factor structure).

Downward Extensions of the WAIS-IV

• Wechsler Intelligence Scale for Children
  • 6–16 years of age
  • First developed in 1949
  • Revised in 1974, 1991, 2003, and 2014
  • Currently in fifth edition (WISC-V)
• The WISC-V
  • Represents the best of psychological testing
  • Standardization, reliability, and validity are all improvements over previous versions
  • Follows model of WAIS-IV
  • Still uses pattern analysis and comparison of index scores, and such practices are still questionable
  • Validity has been studied extensively with various special groups

Downward Extensions of the WAIS-IV (cont.)

• Wechsler Preschool and Primary Scale of Intelligence
  • 2.5 years to 7 years, 7 months
  • First developed in 1967
  • Revised in 1989, 2003, and 2012
  • Currently in fourth edition (WPPSI-IV)
• Extended the age of testability from previous versions
• More flexible based on the specifics of the person being tested
• Same hierarchical structure as WISC-V
• Compatible with measures of adaptive functioning for diagnostic purposes
• Research finds strong psychometric properties

Other Intelligence Tests

• Why are other tests of ability required?
  • Disadvantage examinees with physical or sensory disabilities
  • Language issues
• But these are still considered inferior from a psychometric point of view – why?
  • Smaller and more narrow standardization sample
  • Less documentation re validity (and reliability)
  • Less well-developed administration manual and scoring procedures
  • IQ scores not interchangeable with those from Binet or Wechsler

Other Intelligence Tests: Tests for Younger Children

• Bayley Scales of Infant Development 4th ed. (2020)
  • Comprehensive assessment tool for determining developmental delays in children
  • Predicts intellectual disability but has little predictive ability of intelligence in the normal range
• Kaufman Assessment Battery for Children 2nd ed. (2004)
  • Sound psychometric properties (not sensitive to giftedness)
  • Note: Getting dated

Bayley Scales of Infant Development

• Bases assessment on normative maturational developmental data
• Normative Sample = 1262 (1– 42 Months)
• Gender, race, SES, rural/urban, geographic region
• Domains:
  • Cognitive
  • Language
  • Motor
  • Socioemotional (Parent Report)
  • Adaptive (Parent Report)
    *Test Characteristics
  • Play tasks:
    • Cognition: attention to familiar and unfamiliar objects, looking for a fallen object, and pretend play
    • Language: recognition of objects and people, following directions, and naming objects and pictures
    • Motor: grasping, sitting, stacking blocks, and climbing stairs
      *Scaled Scores
      *Composite Scores
      *Percentile ranks
      *Growth Scores
      *High reliability
      *Good construct validity

Tests for special populations

• Leiter International Performance Scale – 3rd Edition (Leiter-3) (2013)
  • A completely nonverbal measure of intelligence (3 to 75 years)
  • Applications: Examinees who are cognitively delayed, non-English speaking, hearing impaired, speech impaired, or on the autism spectrum
• Peabody Picture Vocabulary Test 3rd ed.
  • Examinee required to select a picture that matches a word definition given verbally by the examiner
  • Does not require a verbal response, fine motor skill, reading ability but may require hearing
  • May underestimate IQ as per Binet/ Wechsler

Tests for special populations (cont.)

• Columbia Mental Maturity Scale 3rd ed. (CMMS)
  • Examinee required to select a picture from 3 to 5 pictures that doesn’t belong
  • Does not require a verbal response, fine motor skill, reading ability
  • Multiple choice - issues of ‘chance’/ random error

Nonverbal group ability tests

• Raven’s Progressive Matrices (1998)
  • Best known and most popular
  • Parallel version available
  • Better than Binet/ Wechsler in terms of racial disadvantage
  • Worldwide norms
• The Culture Fair Intelligence Test
  • Released in the early 1970s
  • Aims to remove cultural bias
  • Norms for the Australian population

Intelligence testing – the controversy

• No intelligence test is perfect
  • Caution is always warranted in interpretation
  • Other sources of information must be secured when drawing conclusions about a person
• Intelligence testing continues to be an issue of hot controversy within psychology
• Several books have been written dealing exclusively with the merits (or otherwise) of the entire enterprise of intelligence testing.
• At the heart of this debate lies
  • the inability to adequately agree on what defines intelligence
  • to clearly articulate what it is that intelligence tests are measuring.

Intelligence testing – the controversy (cont.)

• Some fundamental issues in relation to intelligence that fuel the controversy:
  • Is intelligence a “thing”?
  • Can intelligence be captured in a single number?
  • Can intelligence be measured?
  • Is intelligence innate?
  • Is intelligence heritable?
  • Is intelligence normally distributed?
  • Are ethnic differences meaningful?
  • Can intelligence be changed?

Pros and Cons of Intelligence Testing

• Advantages:
  • Accurate in predicting short-term future behavior
  • Correlated with academic and occupational success
  • Provides valuable info about cognitive strengths and weaknesses
  • Ability to compare performance to age-related peers
  • Provides baseline for determining degree of change
• Limitations:
  • “Labeling” and misrepresenting IQ score as “fixed” ability
  • Focus on analytical and scientific models of thought
  • Look at “what” rather than “how”
  • Limited usefulness in assessing minority groups

Intelligence testing – the controversy (summary)

• In summary, the very nature of intelligence is controversial, and by extension, intelligence testing has also come under attack.
• The most concern is probably in relation to the misuse of intelligence tests.
• The widespread use of intelligence tests as predictors of academic achievement appears unlikely to be overturned as a result of this controversy.
• The ethics of testing we have discussed previously are highly pertinent to this discussion.