INT Lecture 8 - Sex Differences in Intelligence

Key Questions

• Differing views on whether women and men differ in general intelligence
• Examination of the specific mental abilities, where the differences arise

Scottish Mental Surveys 1932

• 79,376 11-year-olds participated in a survey, representing 86% of their birth year demographic
• Survey included 39,343 girls and 40,033 boys
• The study aimed to test the entire population and look at intelligence
• The large sample size allows for detecting even tiny sex differences
• Data consisted of the Moray House Test No 12 and two non-verbal (picture tests)
• Scores across three tests were summed to create an overall IQ score (mean of 100 and SD of 15)
• In middle-scoring groups (90-110 IQ), there are more girls
• In a lower-scoring group (just under 60 IQ), the percent difference between girls and boys is 17.2%
• There is a larger number of boys in the higher scoring group of 130-140 IQ
• The percent difference between girls and boys in the 130-140 group is 15.4%

Summary of 11-year-old data

• Analysis was undertaken using a large sample of around 40k females and 40k males
• This allowed for the detection of even minor sex differences in intelligence
• No difference was detected in average intelligence (n = ~40k) in either females and males (n = ~40k) at age 11
• Sex differences were evident the spread of IQ scores
• More girls had IQ scores nearer average
• Boy's had IQ scores are more at either the lower or higher levels of intelligence

National Longitudinal Survey of Youth 1979

• The study focused on post-adolescent American adults
• Opposite-sex siblings were selected from a large sample (n = 12,686)
• Opposite-sex siblings were selected to control for family background and genetic factors
• The study found 1292 sibling pairs with the mean age of 18.4 years old and SD of 2.1
• Intelligence was measured using the Armed Forces Qualification Test
• The test consisted of four subtests: arithmetic, word knowledge, paragraph comprehension, and mathematics knowledge
• A general intelligence score was calculated

National Longitudinal Survey of Youth 1979 Results

• Males showed a very slight but statistically significant advantage on the g factor
• Effect size was tiny (Cohen’s d = 0.064)
• Mean IQ scores were significant, the difference was less than 1 IQ point
• Difference between males and females was negligible
• US adults, like the Scottish 11-year-olds, showed more males at the lower and higher extremes but overall, the means were the same

Cognitive Abilities Test (CAT) 2001

• Large dataset representative of UK school children aged 11-12
• Participants included 324,000 children from 1305 schools, around 50% of children from 11-12 years
• Overall intelligence test score favoured girls by about three-quarters of an IQ point
• The greater proportion of girls in CAT 2001 clustered around the middle
• There was a greater proportion of boys present at the extreme scores

CAT3 Nine Tests

• These tests are broken down into 3 cognitive domains:
  • Verbal reasoning
  • Quantitative reasoning
  • Non-verbal reasoning

The Cognitive Abilities Test (CAT) 2001, Verbal reasoning:

• Verbal reasoning section includes:
  • verbal classification
  • sentence completion
  • verbal analogies

The Cognitive Abilities Test (CAT) 2001, Quantitative reasoning::

• The quantitative reasoning section includes:
  • number analogies
  • number series
  • equation building

The Cognitive Abilities Test (CAT) 2001, Non-verbal reasoning:

• The non-verbal reasoning section consists of:
  • figure classification
  • figure analogies
  • figure analysis

CAT Test Results:

• Girls performed better than boys 0.15 of a standard deviation in verbal ability or just over 2 IQ points
• Girls scored higher, while boys were more often situated on the lower end in verbal-based tasks
• Boys had a slight advantage on the rest with quantitative and non-verbal results only being 0.03 of SD
• Very little effect shown for either boys or girls on quantitative and non-verbal

Meta-Analyses on Cognitive Test Results

• Overview of multiple studies shows women score better in reading comprehension, perceptual speed and associative memory
• Males tend to demonstrate greater mathematics and visuospatial reasoning skills
• NAEP monitored academic achievements in a large range of 9- to 17 year olds, with sample sizes from 70-100k
• Results clearly suggest females display a greater level of writing capabilities in comparison to males

Raven's Progressive Matrices (RPM)

• RPM is assumed to be a measure of general intelligence, without language
• A review of 120 studies before 1980 found no differences between males and females on Raven's Progressive Matrices
• 2 meta-analyses found that in adults, men score higher than women by approximately 4.5-5 IQ points on the RPM
• Meta-analysis of 57 studies from 30 countries, covering 195 samples of approximately 80,000 people, was undertaken
• Results were presented as generalisations about male and female differences in RPM scores
• The male:female ratio in each sample didn’t match the country's male:female ratio, possibly skewing the results
• This potentially skewed the recruitment effect, making some conclusions about the results questionable
• Males see an increase in scores in the RPM from the age of 15 onwards
• It is claimed the male-female difference emerges after 15 years of age favoring males
• While this is considered significant, the difference is considered small with a score of 0.2 or less
• Other studies have shown no male-female differences with larger samples as seen in RPM
• scores contain a significant visuospatial reasoning component, moderate in size

Sex Differences in Specific Cognitive Abilities

• Men perform better when concerning spatial tasks
• Women score better under verbal tasks showing greater proficiency in comprehension and language

Biological cognitive abilities and sex

• Polycystic ovary syndrome (PCOS) is an endocrine disorder associated with elevated levels of free testosterone
• Women with PCOS have demonstrated significantly worse performance in tests of verbal fluency, verbal memory, manual dexterity, and visuospatial working memory
• These women did not perform any differently in basic mental rotation, spatial visualization, spatial perception, or perceptual speed
• Women with elevated testosterone levels typically perform worse on cognitive tests

Stereotype threat

• Results displayed in performance accuracy for SAT's, specifically math sections
• "It's important to keep in mind that if you are feeling anxious while taking this test, this anxiety could be the result of these negative stereotypes that are widely known in society and have nothing to do with your actual ability to do well on the test."

Summary

• There is little evidence of any average differences in intelligence between girls and boys or young adults
• Females clustered around the average IQ score
• Males are at both the higher and lower extremes, demonstrating more variable scores
• Evidence of sex differences in specialised cognitive abilities shows men are better spatial and women are better verbal
• Some evidence suggests that effects on cognitive ability level derive from biological and/or cultural stereotype threats