psy 4331

Psychology: Scientific study of ___ It aims to understand common principles that explain everyone’s thoughts, feelings, behaviors.

Affect, Behavior, and Cognition.

As a Science: Has a ___

scientific method, relies on theory (testable and falsifiable), integrates deeply with other fields.

dispositional trait approach

conceptualization of individual differences

measurement of individual differences

A Personality Trait or Characteristic

Any aspect of human psychology that shows variations among individual people

A tendency in thought, feeling, behavior. Conceptually cohesive. Something that is stable over time and context.

A Personality Trait or Characteristic

Population Level

 Human nature - like all others, traits and mechanisms possessed by (nearly) everyone

Group Differences

Understanding how and why co-vary with various ways with various ways humans can be grouped

Individual Differences/Uniqueness

Nomothetic approach – Statistical comparisons of individuals, Idiographic approach – focused on a single subject, psychobiographies

Funder’s 2nd Law

There are no perfect indicators of personality: there are only clues, and clues are always ambiguous. Which data tells us about a person's personality, and in what ways?

Self-report

Relies on self-expertise. Classic research suggests that self-reports are pretty good. Methods include face-valid items, surprisingly easy to create (Questionnaires, surveys, open-ended questions etc.).

 “Definitional Truth”

Some self-reports are inherently true

Informant Report

Other people can accurately report about you. Cognitive errors such as salience bias (predisposes individuals to focus on or attend to items, information, or stimuli that are more prominent, visible, or emotionally striking) recency effects (the tendency to remember information that was presented most recently).

Life Outcomes

Objective information about one’s life. the outcomes of how a person moves through life. Measured via things like archival records (Educational attainment, legal records, salary, job). Most “objective” than human reports. Verifiable. Reflects personality in situ with real world outcomes, personality mutually reinforced with context.

Direct Self-ratings

are usually on a “Likert” scale (e.g. 1-5)

 Indirect Self-reports

have the “intent” and “meaning” of the measure obfuscated. They’re more transparent and they have a lower likelihood of having answers that promote social desirability.

 Open-ended Self-descriptions

Has the best of both worlds, you get direct and indirect, but requires specialized knowledge to extract personality.

Experience Sampling Methods (ESM)

has people make periodic reports throughout the day. There are two different subtypes, Ecological Momentary Assessments (EMA) and Electronic Activated Recorder (EAR). Can provide additional data through questions like “Where are you right now?”

Traces

are a logical extension of “behavioral residue”, the digital residue of one’s activity. Logs of where you’ve been, what you’ve been doing, rather than going into your dorm room. Includes things like looking at people's likes, shares, etc.

Words

are EVERYTHING that you say and they directly relate back to your personality. There are two key components to “word” data. Content words are what you talk about (nouns, verbs, etc.). Function words are how you talk about it (tone, pronouns, prepositions, etc.).

Images

are the visual information we share online (instagram, snapchat, etc.). The content tells us a lot about the person. What their interests are, what their environment is, how they want others to see them (impression management). What can you tell about people based on their face? Zero-acquaintance judgements can report snap judgements on levels of extraversion/narcissism were passingly accurate.

Networks

are our connections with other people. Can be determined by social networks, who you interact with, who you are near (through bluetooth). “Birds of a feather”, homophily, local clustering, and is true in disorders as well. Digital data is a very rich source and bypasses some limitations of BLIS but is not perfect. Captures an “expanded” view of personality in the digital age.

Psychometrics

are tightly woven into personality psychology and is the field of study concerned with the theory and technique of psychological measurement.

Classical Test Theory (CTT)

 the basic psychometric theory that attempts to map constructs to data. The construct under study has a “true” score: T. We attempt to measure the construct with X. We try to get X as close to T as possible. X = T + error

 What types of errors are there?

Misunderstandings, idiosyncratic interpretations, distractions, moods, time pressure, time of day/year/etc.

Reliability

 “How consistent/dependable is the measure”?

Test-Retest Reliability

Does the same measure give us the same result each time?

Internal consistency

Are these different measures reflecting the same underlying thing?

Inter-rater reliability

Informant reports/annotations. Is there agreement in measurement?

Validity

“Does this actually measure what it’s supposed to?”

Types

are configural, person-centered. You either/or – no middle ground. Examples of this are galen’s four humours/temperaments, phrenology, jung’s dominant psychological functions (rational vs. irrational, myers briggs personality types, A/B personality theory)

Dimensional traits

are a continuous spectrum, variations around universal psychological properties. It is self-centered and relational. Characterizing individuals by composition (levels of X) rather than configuration (X type of person). e.g. “a neurotic person” vs. “a person scoring high in neuroticism”.

Lexical Approach 

Used “measurement of character” (1884). Important descriptors should have unique words and others followed his work across languages. Allport & Odbert (1936) shaped the future through his research on personality traits.

Five Factor Personality Model 

Research from Raymond Cattell built on Allport and Odbert’s research and wanted to find meaningful constructs of PErsonality Traits. This was the first, very early usage of factor analysis. Had clusters of intercorrelations and S and I data types.

 Raymond Cattell

He found 16 Factors and called them “Primary Traits”. He also found that those traits were hierarchical and differentiated between “primary” and “global” traits. (16 primary, 5 global). Cardinal (e.g. Extraversion) → Primary (e.g. Liveliness) → Global (e.g. Enthusiastic, animated, spontaneous).

Hans Eysenck

made the dimensional, biological theory of personality and PEN method. It’s highly reliable, highly replicable across studies and labs.

Five Factor Personality Model Weaknesses 

 3 of the 5 are just lower level, “primary” traits. Meta-analyses suggest 3 factors, not 5. Why 5? What’s the theory? Where’ the ontogeny? Where’s the biological link? 

The Five Factor Personality Traits

Neuroticism ←→ Extraversion, Openness, Agreeableness, Conscientiousness

Crituques of the Five Factor Personality Triats

 Is it just how good their or socialization/social desirability is? Does the lexical approach boil down to traits that people think are “good” vs. “bad”?

Where are assessments used?

from scientific research (measurement of theory-relevant constructs), Personnel selection (wartime, military recruitment, industry jobs, screening applicants and fit), Diagnostics (clinical/ medical/ psychiatric applications, screening for pathology, assisting with clinical diagnosis)

Factor Analytics Method

An intrinsic, data-driven method with the goal of discovering associations in personality clues. We use it when we’re looking to discover patterns in how clues relate to each other. Discover latent factors that make all related clues similar.

Rational Method

A criterion-reference, theory driven method. Criterion means something that we want to measure/quantify. Goal: measure a specific construct. Should be directly, obviously, and rationally related to construct. Requires accurate self-assessment. Willingness to report. Items must mean the same thing to all test-takers. All items must be valid indicators.

Empirical Method

A criterion-reference, data-driven method. Goal: to measure a specific construct. somewhere between the Factor analytics and Rational Methods. “Let reality speak for itself”, Gather lots of items – item content “doesn’t matter”. Statistically determine which items differentiate groups.

Measurement Set Size

More clues = more data = less error = greater ability to accurately assess personality. 

Spearman-Brown formula

using cronbach’s a shows that a values increase as the items increase. An excellent measure is when 0.9 < a, is excellent and a < 0.5 is unacceptable. (e.g. 10 items; a = .6, 20 items; a = .75, 40 items; a = .86)