Methods - personality 333
Page 2:
Personality is a construct
Construct: An idea about a psychological attribute that goes beyond what might be assessed through any particular method of assessment
Data: Observations or measurements, usually (but not always) quantified and obtained in the course of research
Variables: A condition in an experiment or a characteristic of an entity, person, or object that can take on different categories, levels, or values and that can be quantified
Page 3:
Importance of being clear about variables in research
Conceptual definition: A description of something in terms of what it means
Operational definition: A description of something in terms of the operations (procedures, actions, or processes) by which it could be observed and measured
Page 4:
S Data (Self-Report)
Obtained by asking individuals directly
Often in the form of surveys
Straightforward and easy to obtain
Face validity
Page 5:
I Data (Informant Report)
Obtained by asking someone who knows the person of interest
Can use reworded S Data measurements
May come from any relevant source
Relevance depends on the reason why data is gathered
Page 6:
L Data (Life Outcomes)
Verifiable, concrete, real-life facts that may hold psychological significance
Can be thought of as the "residue" of personality
Page 7:
B Data (Behavioral Observations)
Obtained by observing the person's behavior
Inherently interpretative
May be hard to categorize
Natural vs. Laboratory B data
Page 8:
Advantages and disadvantages of different sources of data for personality assessment
S Data: Self-Reports
Advantages: Large amount of information, access to thoughts and feelings, simple and easy
Disadvantages: Error reports, bias, some S data are true by definition, causal force
I Data: Informant Reports
Advantages: Large amount of information, real-world basis, common sense
Disadvantages: Limited behavioral information, lack of access to private experience, error, bias, causal force
L Data: Life Outcomes
Advantages: Objective and verifiable, intrinsic importance, psychological relevance
Disadvantages: Multi-determination
B Data: Behavioral Observations
Advantages: Wide range of contexts, appearance of objectivity
Disadvantages: Difficult and expensive, uncertain interpretation
Page 9:
Not all data can be easily categorized
Behavioroid: A type of data where participants report what they think they would do under various circumstances
A piece of data may have elements of two or more types of data
Page 10:
Reliability: The tendency of an instrument to provide the same comparative information on repeated occasions
Measurement error: The variation of a number around its true mean due to uncontrolled, essentially random influences
Improving reliability through care, standardization, measuring something important, aggregating, and using the Spearman-Brown formula
Page 11:
Internal consistency and stability across time as measures of reliability
Test-Retest reliability and Inter-rater reliability
Page 13:
Validity: The degree to which a measure actually measures what it's intended to measure
Must be reliable
Conceptual definition vs. operational definition
Page 14:
Types of validity: Construct validity, Criterion validity, Convergent validity, Discriminant validity, Face validity
Page 15:
Considerations for validity: Cultural differences, variation in meaning of constructs, interpretation differences, response sets, acquiescence, reverse-coded items, social desirability
Page 16:
Reliability vs. validity metaphor using target shooting
Page 17:
Generalizability: The degree to which a measurement or conclusion can be found under diverse circumstances
Threats to generalizability: Undergraduate samples, lack of ethnic and cultural diversity, gender bias, cohort effects, shows vs. no-shows and response rates
Page 19:
Projective tests: Assessment in which individuals project from the unconscious onto ambiguous stimuli
Projective hypothesis: Individuals supply structure to unstructured stimuli based on their own unique patterns of conscious and unconscious needs, fears, desires, impulses, conflicts, and ways of perceiving and responding
Types of projective tests: Rorschach inkblot, Thematic Apperception Test
Page 20:
Validity of projective tests is mixed
Weaknesses of B data
Projective tests still heavily used in psychotherapy settings
Page 21:
Objective tests: Personality tests consisting of a list of questions to be answered by the participant
Used in many areas inside and outside of psychology
Need to be reliable and valid
Single-trait vs. omnibus tests
Page 23:
Different methods for developing personality assessments: Rational (Theoretical) method, Factor analytic method, Empirical method
Page 24:
Rational method: Items directly related to what is being measured, may come from theory or participant's thoughts/experiences/views
Example: Woodworth's Personality Data Sheet (WPDS)
Page 25:
Four conditions for validity of measurement in the rational method
Many rationally constructed personality tests fail to meet one or more criteria
Page 26:
Factor analytic method: Statistical method for finding order amid seeming chaos
Identifies groupings within a list of items
Page 27:
How factor analysis works
Subjective process of labeling factors based on theoretical motivations or item contributions
Page 28: The Factor Analytic Method
Limitations of factor analysis:
Quality of information limited by quality of items
Factors don't always make sense
Difficulty and subjectivity in deciding how items are conceptually related
Uses of factor analysis:
Reduce list of traits to essential few
Refine personality tests
Page 29: The Empirical Method
Data-driven approach to test construction
Basic assumption: Certain kinds of people will have distinctive ways of answering questions on personality inventories
Steps in the empirical method:
Gather lots of items
Administer items to people already divided into groups
Compare the answers of the different groups
Items may seem absurd or unrelated to personality
Page 30: Minnesota Multiphasic Personality Inventory (MMPI) Sample Items
Examples of items from the MMPI:
I like mechanic magazines
I have a good appetite
I think I would like the work of a librarian
My father was a good man
There seems to be a lump in my throat much of the time
My daily life is full of things that keep me interested
I enjoy detective or mystery stories
I am sure I get a raw deal from life
Page 31: The Empirical Method (continued)
Same information as on page 29
Page 32: MMPI Scales
Different scales of the MMPI:
Hypochondriasis
Depression
Hysteria
Psychopathic Deviate
Masculinity-Femininity
Paranoia
Psychasthenia
Schizophrenia
Hypomania
Social-Introversion
Page 33: The Empirical Method (continued)
Implications of ignoring item content/low face validity:
Items can seem contrary or absurd
Responses are difficult to fake
Tests are only as good as the criteria by which they are developed and/or cross-validated
Considered B data though it is dependent on self-report
Can also be used with factor analysis (and the rational approach)
Page 34: How Do We Study Personality?
Observing the self and others as a starting point
Research designs:
Case study
Correlational
Experimental
Multi-factor studies
Page 35: Case Studies
In-depth study of one person over a long period of observation
Typically includes unstructured interviews
Usually used for extraordinary situations
May undermine current thinking or initiate new models
Provides rich detailed information
May not generalize (idiographic)
Page 36: Relationships Among Variables
Psychologists examine how different variables relate to each other
Two forms of examining relationships between variables:
Associative (correlational)
Causal (experimental)
Statistical significance vs. practical significance
Page 37: Correlational Research
Correlational research establishes the relationship (not necessarily causal) between two or more variables
Variables are measured, not manipulated
Page 38: Correlational Research (continued)
Predictor variable (x) is used to predict another variable in a correlation/regression analysis
Criterion variable (y) is the variable being predicted by another variable in a correlation/regression analysis
Page 39: Correlational Research (continued)
Correlation coefficients express the strength and direction of the relationship
Range between -1 and +1
Depicted by r (e.g., r = -.35)
+/- expresses direction
Number depicts strength
Page 40: Correlation
Visual representation of correlation coefficients
Shows the strength and direction of the relationship
Page 41: Correlation Does Not Imply Causation
Correlational data cannot determine causation
Relationships may be due to:
Causation
Reverse causation
Bidirectional
Third-variable problem
Page 42: Correlational Research (continued)
Advantages of correlational research:
Establishes a relationship and can make predictions
Can be quick and easy
May be the only way to study a relationship in the real world
Disadvantages of correlational research:
Cannot determine causation
Page 43: Experimental Research
Experimental research involves manipulating one or more variables to test for causal influence on another variable
Variables are categorized as independent (IV) or dependent (DV) variables
Within-subjects vs. between-groups design
Page 44: Experiments Can Make Causal Claims
Three criteria for causality: covariation, temporal precedence, and internal validity
Experimental control and random assignment for between-groups experiments
Page 45: Experimental Research (continued)
Advantages of experimental research:
Controls for extraneous factors to rule out alternative explanations
Can make causal claims
Disadvantages of experimental research:
Confounding variables
May not generalize to "real-life"
Placebo effects
Experimenter/observer expectancies may bias results
Page 46: The Myth of Experimental Research as the Gold Standard
Experiments have limitations:
Not always possible
May not generalize beyond the study
Effects might not be long-lasting
Difficulty in determining the exact cause and effect relationship
Oversimplification of the relationship
Page 47: Multifactor Studies
Studies with two or more predictor/independent variables
Variables can be manipulated, measured, or a combination
Subject (participant) variables help examine the complexity of life
Experimental personality research examines the relationship between a personality factor and an experimental manipulation on a dependent variable
Page 48
Main effects: A finding where the effect of one predictor variable has an effect on the dependent variable, independent of other variables
• Interactions: A finding where the effect of one predictor variable differs depending on the level of another predictor variable