Week 3: Traits: Taxonomies Flashcards
The big 5 (NEO-PI-R) Costa & McCrae
Extraversion, neuroticism, openness, agreeableness, conscientiousness
Only robust dimension rival to Eysenck.
Eysenck’s trait-taxonomy
Evidence for just 3 dimensions of personality (measured by the EPQ-R)
Extraversion (introversion)
Neuroticism (emotional stability)
Psychoticism (socialised/superego)
Limitations of the FA approach
Vertical structure of the hierarchy (primary factor may break down into several sub-factors with small correlations between them ie impulsivity)
Horizontal structure of the hierarchy (highly loaded traits can make it hard to figure out where they belong horizontally)
Bandwidth-fidelity dilemma
Bandwidth-fidelity dilemma
Broad dimensions or narrow traits for better behaviour prediction?
- Reliability in measurement
Broad typically more reliable as reliability is improved by aggregating over a large number of correlated items - Criterion-related validity
Ones & Viswesvaran argue that large validity coefficients for narrow bandwidth traits are possible only if the criterion measure is also narrow and homogenous. However, real world is complex therefore should use broad. - Theoretical
Ones & Viswesvaran evaluate the merits of broad vs narrow from 3 different perspectives: reliability in measurement, criterion-related validity, theoretical
Trait taxonomies and FA
Traits refer to ℅ variation of a number of thoughts, feelings and behaviours.
FA is a powerful tool for studying these covariates; assumes that correlations between a set of observed variables due to a small number of factors/latent variables (often assumed to have causal efficiency)
Eysenck’s hierarchical approach
Level 1: individual thoughts, feelings, beh
L2: habitual thoughts, feelings, beh
L3: traits (primary order factor)
L4: dimensions (second order factor) –> evidence of just 3 dimensions of personality: E-I, N-ES, P-S
Cattell’s 16 PF
Primary factory
Attempts to replicate have not succeeded.