Lecture 2.1 Personality Assessment, Measurement & Research Design (Chapter 2)

Sources of data

Self-report (S-data), Observer (O-data), Test (T-data)

2 forms of Self-report (S-data)

Structured and unstructured

Structured (s-data0

Response are SET

Includes:

Dichotomous (forced choice, eg true/false or I am a smart person \/this cause is easy

Likert rating

Unstructured (s-data)

Responses are NOT set

Inculdes: opened questions

Pros and cons of structured (s-data)

Pros: standardization, use of stats

Cons: limits in responses, possible limited accuracy

Pros and cons of Unstructured (s-data)

Pros: Detailed, no limits to responses

Cons: MAY NOT BE standardized, use of stats may be limited — BUT research can be qualitative

Limitations of self-report data (s-data)

honesty in responses

not having self-knowledge/objectivity

There are other s-data approaches

event sampling —> ecological momentary assessment

Self-report data that occurs OVER TIME to assess variables that might change in ‘real-time’

Kranzler et al., 2017

sample: 47 adolescents/young adults ages 15-21 who had self-injured 2 times or more in the past 2 weeks

method: used a mobile app over a 2 week period to track intensity of thoughts about self-injury + occurrences of self-injury + state-level emotions before and after engaging in self-injury

findings: increased negative emotions and decreased positive emotions predicted intensity of self-injury thoughts + increased negative emotions predicted engagement in self-injury + after self-injury they reposted decreases in intended negative emotion and increased in relief

implications: self-injury is driven by the need for emotion regulation (self-injury is used to cope with difficult/painful emotional experiences + focusing on emotion regulation is like critical in intervention

what is this study (Kranzler et al., 2017) an example of

Event Sampling → Ecological Momentary Assessment

observer (O-data)

involves gathering data from other individuals (not the self) (eg. Teacher ratings of children’s behavior in the classroom
Here, the teacher (an observer) provides assessments of traits like attentiveness, aggression, sociability, or impulse control. The data are not self-reported by the child, so they qualify as observer (O-data).)

Pros and cons of observer (O-data)

Pros: access to unique data & multiple informants

Cons: objectivity AND respondents may not be able to infer internal processes (eg, feelings)

Where can O-data be collected

Naturalistic + artificial

Naturalistic setting

Observations that occur in a natural setting

artificial setting

observations that occur in settings created to resemble a real-life setting

Test (T-data)

utilizes standardized testing situations to determine aspects of perosonality

Forms of T-data

projective tests

Mechanical Recording

Physiological data

t-data limitations

participants may guess the trait being assessed and create an impression

participants and researchers may view the testing situation differently

The influences of the researchers on the part

Evaluating Personality Measures

Reliability, Validity, Generalizability

What is reliability

Whether data reflect the true level of what is being measured (consistency of measurement)

Types of reliability

Test-retest, Internal consistency, Inter-rater

Test-retest

scores of a measure correlate on repeated measures

Internal Consistency

Items on 1 measure correlate with each other

Inter-rater

Ratings on 1 observer correlate with those of another

Response sets (impact reliability)

response tendency that is UNRELATED to item content (tendency to respond in a particular style regardless of the content of the questions.)

tendency to respond in a particular style forms

Acquiescence, Extreme responding, social desirability

Acquiescence

“yea-saying”

“Yes… yes… yes…” even if items contradict.

Extreme responding

Choosing extreme points (e.g., “Strongly agree” / “Strongly disagree”) rather than moderate ones.

social desirability

Answering in a way that makes the person look good or socially acceptable rather than true:

“I always help others and never feel angry.”

Validity

Agree ti which a test measures what is claims to measure (accuracy)

Types of validity

Face, predictive (criterion), convergent, discriminant, Construct

Face

whether it appears to measure what it should

Predictive (criterion),

Whether the test predicts criteria it is suppose to

Convergent

Whether the test correlated with other, similar, tests

Discriminant

refers to what the measure should NOT correlate with

Construct

Include all types of validity — broader in scope

Generalizability

whether a measure retains validity over different contexts/samples (Does this measure still work (i.e., remain valid) across different people, settings, and situations?)

Research designs in personality

Experimental methods, Correlation studies, case studies

Experimental methods

used to determine causality

How many requirements for Experimental methods, and what are they?

2

1. An independent variable is manipulated to affect the dependent variable

2. participants are equilivent (random assignment)

Correlation Studies

Used to understand if 2 (or more) variables share a relation

Correlation Studies (Correlation Coefficient)

Indicates directions and degree of relation (range -1 to +1)

Correlation Studies (Limitations)

Directionality (no causation)

Third variable

case studies

Focus on one case (or a small number of cases)
Rich, qualitative detail
✔Often involves multiple types of data, such as:

• interviews

• observations

• personal history

• medical/clinical records

• psychological tests

Example; Ted Bundy