General / Measurement

Definition

Scientific Theory

Good Theory

Basic Issues

Definition: usual pattern
of behaviours / feelings / thoughts
across time & situations (enduring)
that differentiate people (distinctive)

Lewin: behaviour = f(personality, situation)

Allport - individual differences

  • dynamic organisation (loose knit, depends on context)
  • of psychphysical systems (mind/body, nature/nurture)

Cattell - human universals

  • population-based characteristics = nomothetic (what we share)
  • to predict behaviour

Issues

  • Unconscious: vs conscious
  • Self: what is it
  • Nomothetic (laws) vs idiographic (individuals)
  • Gender diff
  • Situation influence
  • Culture: nurture (vs nature)
  • Usefulness: application

Deductive: general -> specific

  • Top-down
  • Apply theories to situations

Inductive: specific -> general

  • Bottom-up
  • Observations -> hypo -> revise

Differences

  • Types: limited distinct categories - qualitative - most general
  • Traits: gradual transitions - quantitative - most specific
  • Factors: broader than traits - quantitative

Individuals

  • Study 1 person

Types

Qualitative

  • Someone - depth & experience
  • Interviews, archival data
  • Uses transcription, coding

Quantitative

  • Everyone, trends
  • Mean & SD
  • Controlled (min. context)

2. Experiment

  • Manipulate IV & measure DV - compare groups - usually natural
  • Strengths: specific variable, cause & effect, objective
  • Weaknesses: artificial, not all measurable, demand char., moderators

3. Case studies

  • Intensive, systematic - individual, small community - to gather ideas
  • Strengths: not artificial, flexibility, depth
  • Weaknesses: no causation, not objective, can't generalise

Thematic

  • ID themes/patterns across data
  • Strengths: > individual, interpretation, categories
  • Weaknesses: subjective, removed from individual

Mixed method: breadth + depth

1. Correlation

  • Degree to which 2 variables related - use correl. coefficients & factor analysis
  • Strengths: wide range of variables, easy, large samples
  • Weaknesses: no causation, self-report reliability, no depth

Purpose

  1. Explain phenomena
  2. Predict new info

Evaluating

  1. Verifiable / testable
  2. Comprehensive: broad range of behaviours
  3. Applied value
  • Parsimony: as few as possible
  • Heuristic value: suggest new research
  • Empirical validity
  • Breadth of info: diff sources of info

Person / Situation

  • Both important, work together
  • Situation: other people, environment
  • Person-situation debate: stable traits predict behaviour vs only situation important
  • -> both predict equally well
  1. Have diff experiences
  2. Respond diff to same situation
  3. People choose situation
  4. People change situation

People change

  • Usual tendencies: behaviour most of time (patterns), when you have choice
  • Measure over shorter periods

Types

Self-report

  • e.g. MMPI, NEO-PI, MBTI, PRF
  • Pros: easy/cheap, objective, reliable, self-view
  • Cons: not rich, easy to fake, relies on self-knowledge

Q-sort

  • Sort cards into piles - self, others, diff situations
  • Pros: active, gives rankings, use for diff targets
  • Cons: see self-report

Other ratings

  • Pros: less biased perspective, 'visible' traits, children
  • Cons: depends on knowledge & lack of bias

Biological

  • e.g. EEG, PET, fMRI
    • diseases & genetic conditions
  • Pros: not self/other-report
  • Cons: difficult/$$$, relationship to personality complex

Behavioural observations

  • e.g. experience sampling, code facial expressions
  • Pros: captures actual behaviour
  • Cons: hard to interpret

Interviews

  • e.g. free-form (Kinsey), structured (Type A)
  • Pros: deep, flexible
  • Cons: interviewer/respondent bias, $$$/time-consuming

Expressive behaviour

  • e.g. nonverbal, verbal
  • Pros: captures unique behaviour (incl. emotions), good for some (charisma)
  • Cons: hard to capture/code/interpret, culture important

Document analysis

  • e.g. study personal writing, diaries - individual
  • Pros: depth, across time, deceased, objective
  • Cons: only some aspects, not complete or always honest

Projective tests

  • e.g. Draw-a-Person, Rorschach inkblot, TAT, IAT
  • Pros: unconscious, insights
  • Cons: subjective, low reliability & validity

Demographics/lifestyle

  • Pros: age/gender/occupation/culture
  • Cons: can be misleading, doesn't tell much about person

Internet analysis

  • Pros: huge data, cheap, current topics, networks, unobtrusive
  • Cons: self-presentation bias, ethics

Parameters

Reliability = consistency

Validity = measuring what it should

Choosing test items

  • Should discriminate
  • Should be interrelated
  • Should have useful distribution
  • Reliability/validity vs length

Bias

Ethnic

  • Should consider context & culture -> compare within culture
  • Cultural strength can be seen as weakness
  • Difficult to see prejudice

Gender

  • Environment / expectations often ignored
  • But some gender differences

Response set

  • Unrelated to characteristic measured
  • Acquiescence/naysay: agree/disagree with everything -> reverse-score
  • Social desirability: present favourably to please -> choose between equally desirable
  • Random responses: can't read, subvert, tired -> lie scales
  • Use various assessments

Not good

  • Not evidence-based
  • e.g. astrology, handwriting analysis

Ethics

Internal consistency

  • 2 halves of test highly correlated
  • Measured by Cronbach's alpha (.80)

Test-retest reliability

  • Consistent on diff occasions

Construct validity = whether it measures concept (vs other concept)

Convergent: related to what it should be (2 related tests have high correlation)

Discriminant: not related to what it shouldn't be related to (2 unrelated tests have low correlation)

Criterion-related = whether it predicts outcome

Content validity = whether it fully measures concept

  • Can ID strengths/weaknesses
  • But harmful if biased, poorly constructed, used incorrectly,

Culture

Influence

  • Culture: customs, values, behaviours of nation, ethnic group, class, time period
  • Diff social norms in diff cultures
  • Most research from WEIRD (Western, educated, industrialised, rich, democratic)
  • Cross-cultural diff from geography (e.g. mammals -> agriculture, rice -> collectivist)
  • People + cultural products
  • Mutual constitution model: culture shapes individual; individual shapes culture

ATSI

  • Pre-colonisation: no collective Indigenous ID
  • Tribal ID: traditional
  • Many IDs: local, pan-Aboriginal; community vs legislative definitions; biological (descent) + meaning; Aboriginality is fluid concept
  • Ethnicity: distinct/diff ethnic group - Australia's First People
  • Descent + heritage/persistence + dispossession/resistance; seen as 'objects', deficit discourse; essentialism ignores heterogeneity
  • 'Authentic': remote > urban, full-blood vs half-caste
  • ID: land central

Cultural systems

Language

  • Idiolect: unique version of native language (part of personality)
  • Dialect: regional/cultural characteristics
  • Language affects personality, e.g. use active verbs -> field-independent
  • Bilingual: personality/behaviour according to salient ID
  • Social interaction: e.g. tu/vous
  • Gender: e.g. no gender-neutral pronouns

Differences in traits

Many types because

  1. Theoretical: some more suitable
  2. Methodological: to compensate for bias

But Barnum effect: take vague descriptions to be true & valid
-> use multiple methods

Geography

Ethnicity/race

  • Ethnic - habits, customs; race = physical
  • Flawed to use race as grouping

SES (Socioeconomic status)

  • = social class, income, education

Generation

  • = everyone born in arbitrary 15-20y period
    • More individualistic now

Individualism: favours self

  • e.g. US, later generation
  • Focus on personal attributes
  • Focus on person -> internal attribution

Collectivism: favours others/society

  • e.g. Japan
  • Focus on group affiliations
  • Focus on situation -> external attribution

Testing

Testing culture's personality
Hard because

  • Reference-group effect: compare self to others in same culture
  • Perception of national character: typical member of culture - better - measures stereotypes
  • Big Five: some have no O, China adds interpersonal relatedness

Cultural-bias

  • Using race is flawed (culture <> physical appearance)
  • -> Culture-free: e.g. Ravens - NOT culture-free
  • -> Culture-fair: control/rule out cultural effects, e.g. response time - NOT culture-free
  • Compare biased & culture-fair subscales to see if testing appropriate
    -> consider culture part of personality

Extraversion

  • US/Norway/Switz/Austria/Canada/Serbia/Croatia vs Taiwan/Malaysia/Zimb/SK/Bang/France
  • Linked to individualism, low power-distance, economic prosperity, moderate climate
  • Higher over generations

Agreeableness

  • Greece/Congo/Japan vs Argentina/Ukraine/Japan
  • Collectivist, low alcohol consumption
  • Lower over generations
  • Ref-group effect

Conscientiousness

  • Germany/Switzerland vs Mexico/Indonesia/Greece
  • Low in 'event-time' cultures - more spontaneity
  • Measured behaviourally -> big ref. group effect

Neuroticism

  • Japan/Russia/Spain/France/Belgium/Argentina vs Scandi/Indonesia/Congo/Slovenia/Ethiopia
  • Uncertainty avoidance: rules that minimise ambiguity; low SES
  • No ref group effect
  • Higher over generations (disconnection, extrinsic values?)

Openness

  • Chile/Belgium/Bangladesh vs Ukraine/Japan
  • Moderate climates, high SES
  • Few studies
  • Lower over generations