Differential Psychology Flashcards
how do people vary
- basis = bodily differences
- physiological = medical stuff
- developmental
- phycological
- lifestyle
- demographic status
- experiences
- etc.
Cronbach’s two disciplines of scientific psychology (1957)
- experimental psychologists manipulate conditions
- correlational psychologists seek to identify and understand free-standing patterns of nature
VARIANCE
the expected squared deviation of a random variable from its statistical mean
sd = square root of variance
Meehl’s theory of validity
interpreting test scores means assuming that tests measure theoretical networks’ constructs - networks and theories imply that states should be observable under specific conditions - these conditions aren’t always present, and this creates variance
how to study individual differences
- define the dimensions of difference
- operationalise the dimensions - creates issues as individuals are bias but so are outsiders and groups
how to know if something is measured well?
- if repeated later, we get the same result
- assessment process is free from bias
- when we measure a person/object, the result has the same meaning when we apply it to others
Anatomy of a differential study
- start with a question
- identify needed data
- figure out how to collect a good sample
Population representation
rare to be able to test a whole population so we must select a representative sample - if we select randomly, this could be randomly unrepresentative
LBC21 - population representation
- impractical to measure the whole of Scotland on education so selected Edinburgh - these individuals are better educated than Scotland as a whole
- used NHS registration to give access to 99% of the old population in edinburgh
- wanted people who sat SMS-32
representative variables in LBC21 - ‘old age outcomes’
- longevity (making it to 79 can affect a person)
- health / well-being
- financial resources
- social support
- physical and cognitive ability for self-care
all of these could affect a person’s performance in interviews / cognitive tests
there are also other factors that can effect the way early life IQ presents itself in old age like:
- childhood lifestyle/ soci-economic status
- offspring / relations
- job satisfaction
- personality / religion etc.
- memory of events
Quantitative change
implies a number/objective change
implies a difference in magnitude
requires a common ‘measuring rod’ to be reliable
qualitative change
implies capacities have become eroded / improved
there is no common measuring rod
pervasive in life transitions
cross sectional design
e.g. assess 79 y/o and 83 y/o and compare
advantages
- faster and cheaper
- no P death or drop out
disadvantages
- varying levels of sample security
- no way to evaluate personal development directly
- cohort effects
Longitudinal design
e.g. assess 79 y/o, wait 4 years, assess them again at 83 y/o
advantages
- can measure changes in individuals
- can evaluate prior influences on individual differences
disadvantages
- expensive and time consuming
- Ps might drop out or die
- practice effects on repeated tests
- limited to one cohort
- age/time effects tangled
- measures can become outdated
Cohort- sequential design
e.g. assess 79 y/o and 83 y/o, wait 3 years, assess them all again, wait another 3 years, assess them all again and so on.
