Chapter 2 Flashcards
Introduction
- Raw score: number of questions answered correctly on an intelligence test
- Classic ageing curve: mean score in a graph rises through childhood and adolescence, reaches a plateau in late teens, declines at some point in adulthood
• Longitudinal study
: measure individual when they are young and then again when they are old
• Cross-sectional study
measure a young and an old individual at the same point in time with same measurement
• Cohort effect
: a difference due to generational differences in background and upbringing rather than ageing per se in cross-sectional studies
• Confounding variables
matching the groups on the cohort measures you feel could distort the findings
- When confounding variables in cross-sectional age studies are controlled, the age difference is typically diminished but not removed
- Impossible to remove all cohort effects
• Overlapping longitudinal study:
: testing several age cohorts on one occasion, then testing them on regular intervals thereafter younger ages will become the ages the older cohorts were earlier in the studycross-sectional and longitudinal used for control for cohort effects
• Seattle Longitudinal Ageing Study (Schaie et al.)
in 1956, group of people between 20 and 70 tested and then retested at 7-year intervals, periodically new participants added part of difference between age groups due to cohort effect people aged 60 in 1970 had lower scores than people aged 60 in 200 intellectual skills were not as favored in 1970
Problems with Longitudinal studies
- Drop-out effect: many participants who feel like their intellectual performance gets worse drop out because they regard the testing as a competition, leaving a rest with higher intellectual performance
- size of drop out effect among elder participants can be very large 26-92%
• Berlin Aging study
if only participants who attended several sessions of longitudinal study are considered, then their performance on cross-sectional and longitudinal comparison was similar if all participants were considered (including drop outs): cross-sectional comparisons significantly bigger age changes longitudinal studies might underestimate ageing decline
• Practice effect
participants who remain in longitudinal study improve performance on tests might become test wise: at ease with test procedures, increased general awareness etc…
• Key longitudinal studies:
Seattle, Berlin, Duke Longitudinal Study
Decline
- By age of 75 mean score in measure of intelligence (e.g. Wechsler Adult intelligence scale) is 1 standard deviation lower than young adult mean
- Decline starts in late 20ies
- Old age not special time for intellectual decline
- Decline does not follow same patterns in all individuals
- Decline only in some measures of intellectual skills
- Research based on raw scores of intelligence, not IQ
• Intelligence quotient:
: converting the raw score into a measure of how intelligent a person is relative to the rest of their age group
• Percentile format: percentage with higher or lower scores
• IQ scale : 100: average, 15= SD
• Individual remains roughly same IQ throughout life Scottish study, Deary et al.
• IQ always relative to own age group can therefore differ from younger age groups
• Correlations between IQ test in young and old age r=0,7 still some variability
• No method of measuring intelligence perfect: longitudinal= drop-out and practice effects, cross-sectional=cohort effects
Fluid and crystallized intelligence
• in later life: fluid intelligence declines, crystallized intelligence increases or remains stable supported by Horn and Cattell (1967
• Dewey’s paradox of ageing
maturity good, age bad
• General intelligence (g):
global measure of intellectual ability, actually set of sub-skills
• Fluid intelligence:
ability to solve novel problems
• Crystallized intelligence:
pre-existing knowledge
• test batterie:
set of tests that test the same thing