PSY224 Trait models of personality Objectives Following this lecture/reading you should be able to:– Demonstrate an understanding of how personality traits may be identified Describe & critically evaluate the trait theories of Cattell, Eysenck and Costa & McCrae Address the criticism of circularity Reading IndividualDifferences Ch. 5 Some of the recommended reading: e.g., Block (1995), or Eysenck (1992) & Costa & McCrae (1992) Traits Plenty of words describe enduring dispositions. e.g., happy, demanding, mean, aggressive, neurotic, anxious… Aristotle Key assumptions: – stability of behaviour over time – stability of behaviour across situations – traits are “inside us” – they influence our behaviour traits & states Traits: enduring ways of behaving States: moods/emotions. Vary – over time – as a function of situation & life events So a trait should correspond to an „average state‟ Why not just use adjectives from the dictionary to define traits? …too many to use! (~18 000 words) …need hard evidence that each describes consistent behaviours …& impossible to tell whether some words mean the same as others. What is needed A method of discovering the main personality traits …which should show up through behavioural ratings, questionnaires, & perhaps objective tests …be stable over time …and predict performance in a variety of situations. Factor analysis Used to analyse responses from behavioural rating-scales and/or questionnaires. Looks for items that tend to vary together from person to person. Responses to 6 items Susan Bob Mary Ian 7 6 5 4 3 2 1 0 Insurance Hurts Suspicion Optimist Enjoys Talkative useless loved ones parties Item Factor analysis shows how many groups of items (TRAITS) there are, and which items belong to which group(s). It is claimed that factor analysis can reveal causal influences (“source traits”). e.g., factoring behaviours related to drinking… Guilford & Guilford (1934) Factored responses to 36 items Why THESE 36? Didn‟t find the one („extraversion‟) factor they‟d expected …but 4 factors: introversion, emotionality, impulsivity, self-interest …and by 1956 had found 9 more Cattell (1943) Personality Sphere. Tried to measure all possible trait-descriptors. – 4500 adjectives (Allport & Odbert, 1938) – Removing synonyms reduced this to 171 – Claimed to measure these 171 variables – Factor analysed & found about 46 factors – Some of which are measured by the 16PF – But his work can‟t be replicated. Cattell‟s factors A Warm/reserved L Suspicious/trusting B [Intelligent] M Imaginative/practical C Unemotional/emotional N Shrewd/Forthright E Assertive/co-operative O Guilty/self-assured F Cheerful/sober Q1 Radical/conservative G Conscientious/expedient Q2 Self sufficient/affiliative H Socially bold/shy Q3 Controlled/impulsive I Self-reliant/sensitive Q4 Tense/tranquil Hans Eysenck Wanted to identify neurotics But found 2 factors (extraversion & neuroticism) Later added „psychoticism‟ (1976) MMQ, MPI, EPI, EPQ, EPQR questionnaires. Eysenck‟s 3 factors are easy to replicate Extraverts. Del boy are sociable take risks don‟t persevere are unreliable can lose their temper are assertive are carefree Neurotics: Piglet Feel anxious Feel depressed Feel guilty Feel tense Feel moody Get emotional Psychotics: Francis Urquart Aggressive Cold & impersonal Self-centred Impulsive Lacking empathy Creative Tough-minded Norman (1962)/Costa & McCrae Five factor model: Extraversion, Neuroticism, Openness to experience, Agreeableness, Conscientiousness. (Acronym: OCEAN) Based on (flawed?) ratings of behaviour Some notable problems in replicating the factor structure P=low agreeableness/conscientiousness? Why such different results from various studies? Different ways of sampling/selecting items for ratings or self-report Quality of data varies – e.g., Tupes & Christal (1961) Quality of factor analysis Insufficient interest in properly validating factors before publishing a model. Circularity Beware of doing the following:- – Observing behaviour (e.g., Claire is apprehensive, easily upset, keeps her room compulsively tidy) – Inferring that Claire is high on a trait of Neuroticism – Saying that this high score „explains‟ her apprehension etc. …ctd To EXPLAIN behaviour using traits, it is necessary to show that what‟s measured by the questionnaire reflects some more fundamental process c/f an acid turns litmus red. x turns litmus red. Thus x is an acid. The concept of „acidity‟ is only useful if acids have broader properties than turning litmus red. Which they do... So what are the broader implications of personality traits? see next lecture! (after break) To conclude:– Factor analysis has been useful in discovering the main traits There is universal agreement for 2 main traits (extraversion, neuroticism) Opinions differ about the rest! Circularity is an important issue, so Check traits are broader in scope than the methods used to measure them.