PSYC2017 Personality and Social Psychology Lectures 1-5 PDF
Document Details
Uploaded by BrighterWilliamsite5570
2023
Hannah Woodbridge
Tags
Related
- PSYC2017 Personality and Social Psychology Lectures PDF
- PSYC2017 Personality and Social Psychology Lectures 1-9 PDF
- FMLY 1010 Chap 4 Socioemotional Development In Infancy PDF
- Personality and Social Psychology Lecture Notes PDF
- Personality Summary Notes 2234 PDF
- PSYU2234 Social & Personality Psych Lecture Notes PDF
Summary
These lecture notes cover psychological tests, including validity and reliability, in the context of personality and social psychology. The topics are structured for an undergraduate course.
Full Transcript
PSYC2017 PERSONALITY AND SOCIAL PSYCHOLOGY SEMESTER 2 | 2023 | HANNAH WOODBRIDGE ○ PSYCHOLOGICAL TESTS ● sychological measurement involves assigning numbers to P a person tofaithfully representtheirattributes ○ INTERPRETATION IMPORTANCE ● To make anindirect a...
PSYC2017 PERSONALITY AND SOCIAL PSYCHOLOGY SEMESTER 2 | 2023 | HANNAH WOODBRIDGE ○ PSYCHOLOGICAL TESTS ● sychological measurement involves assigning numbers to P a person tofaithfully representtheirattributes ○ INTERPRETATION IMPORTANCE ● To make anindirect assessmentof latent attributes ● Latent psychological concepts= unobservable ○ Eg. depression, anxiety, personality ● To make decisions about individuals ○ Eg. education, employment, stability ● ● ● ● ● easurements are precise and accurate M Assesses a single attribute Psychologicalconstructsexist ○ Concept (discovery) vs construct (creation) Attribute can be represented by numbers, similar to physical properties ○ Problem:what is neuroticism squared? Constructs arestableacross time and place and have predictable variability ● ● QUALITIES SAMPLING ● ● ● S upposed to represent population Avoids systematic and non-systematicbiases Issues with non-respondents/dropouts/volunteers /response rate ● Methods of scoring ○ Objective ■ Eg. standardised questionnaires ○ Subjective/assessor’s judgement ■ Eg. vignette, projective testing Standardisation= transforming scale scores into universal indexes (Eg. IQ) ○ Used forcomparisonbetween individuals, groups and other scales PSYC2017 Validity= degree to which a claim is correct ○ Does an instrument measure what it claims to measure? ○ The appropriateness, usefulness or meaningfulness of test scores and their interpretation ■ Eg. ‘quality of life’ measure Validity is affected by levels ofbiasesorstatisticalerrors in the test construction and conclusions ○ Objects studies are abstract andlatent constructs CONSTRUCT VALIDITY ● ● SCORING ● Are the observed attributes real? ○ Cultural biases ○ Procedural biases ○ Interpretation biases Does testing help or hurt? ○ Labelling ○ Not a holistic approach External validity VALIDITY ASSUMPTIONS ● ● ● Determine population norms (normative data) ■ Ie. population attributes ■ Norm has limitations (eg. sample size and type) Examples: Z-scores and t-scores onstruct validity= the degree to which constructshave a C coherent theoretical foundation ○ Relevant tointernalandexternalvalidity Operationalisation= the process of generating construct or variable (operational) definitions that allow for empirical assessment ○ Eg. mental health operationalised by presence of symptoms outlines in the DSM-V FACTORIAL VALIDITY ● F actorial validity= type ofconstruct validityfocusedon the measurement instrument ○ Relavent to onlyinternal validity 1 CONVERGENT VALIDITY ● ● onvergent validity= type ofconstruct validitywhich C demonstrateshighlevels ofcorrelationbetween: ○ Items (individual questions) that make up the same or related constructs ○ Other tests that measure the same or related constructs Example (anxiety): other questions in the test for anxiety/other tests for anxiety DISCRIMINANT VALIDITY ● ● ● ● riterion-related validity= degree to which testcorrelates C to (or predicts) atheoretical representationof the construct ○ Method of assessing validity ○ Correlation calculated bycorrelation coefficient ○ Main types: ■ Concurrentvalidity: criterion is in the present(eg. school maths average vs current students maths scores) ■ Predictivevalidity: the criterion is in thefuture(eg. aptitude test vs. future job performance) Example (anxiety): future mental health issues/undergoing current treatment for anxiety S tructural validity= Scoresrepresentthe contentarea they claim to represent ○ A strurally valid assessments represent the whole domain ■ Eg. questions from all modules in the final exam to represent the whole unit CEILING/FLOOR EFFECTS ● isriminant (divergent) validity=oppositeofconvergent D validity,lowerlevels ofcorrelationbetween: ○ Items (individual questions) that make up unrelated constructs ○ Other tests that assess unrelated constructs Example (anxiety): a test for cognitive ability CRITERION-RELATED VALIDITY ● STRUCTURAL VALIDITY T he ceiling/floor effects= the problem of indiscriminability ○ no meaningful viability, everyone is scoring the same ■ problem with questions or problem with scoring system RELIABILITY ● eliability= the degree ofstabilityof measurement R output acrosstime(t emporal) orcontext ○ Absenceoffluctuationthat areunaccounted for (eg. by random error) ○ If you canexplain variance= reliable CLASSICAL TEST THEORY (CTT) ● ● ● Classical test theory (CTT) ○ People (objects) have a true score on a construct ○ Errors are assumed to be random X (observed score) = T (true score) + E (error) Variance = measure of change ○ σ 2 𝑋 = σ 2 𝑇 + σ 2 𝐸 Rough estimate of reliability = reliability index (r) INTERNAL VS EXTERNAL 2 2 ■ r =σ ■ Proportion of variance 𝑇 / σ 𝑋 EXTERNAL VALIDITY ISSUES WITH CCT ● E xternal validity= degree to which results can be generalisedacross differentcontexts ○ Ecological validity= the degree to which test score reflect thenatural world ● Issues with CTT ○ Does the construct exist? ○ How stable is the true score? ○ What is random error? INTERNAL VALIDITY RELIABLE MEASUREMENT ● Internal validity= degree of confidence aboutcausal relationsbetween the measured constructs ○ Naturalistic designs= have good EV but bad IV ○ Experimental designs= have good IV but bad EV CONTENT VALIDITY ● ● SE of measurement ○ The range of observed scores which the true score would be expected to be ○ 95% confidence interval ■ +/- 1 SE = 88.3% sure within that interval ontent validity= at face value, the items appearto be C related to the construct in question PSYC2017 2 SOURCES OF MEASUREMENT ERROR INDIVIDUAL ERROR ● ● Idiosyncratic ○ Mood ○ Fatigue ○ Memory ○ Language Generic ○ Faking questionnaire ○ Self-deception ○ Acquiescence ■ yes to everything ○ Nay-saying bias ■ no for everything ○ Midpoint/extreme responses ■ Floor/ceiling effects ○ Random responses INTER-RATER RELIABILITY ● Inter-rater reliability = raters score similarly for the scale, instrument or construct in question ○ % of agreement (Cohen’s kappa) FACTOR ANALYSIS ● ● ● F actor analysis= advanced multivariate statistical techniques used to uncover latent constructs (dimesion, factor or component) from a set of observed attributes ○ Make the data smaller and more meaningful ○ Holistic assessment tool ○ High correlation between items = they are potentially relate to the same factor Factor loading = the correlation between the item and the factor (good = above .4) Example application of psychometric knowledge: ○ Profiling and psychobiography ○ Psychobiography and psychohistory MEASUREMENT TYPES OF FACTOR ANALYSIS ● ● ● Content-related ○ Lack of clarity ○ Biased questions Format-related ○ Arranged in a particular way Administration-related ○ Distracting setting RELIABILITY COEFFICIENTS ● ● INTERNAL CONSISTENCY RELIABILITY ● Internal consistency reliability = the degree of consistency of items that measure the same construct ○ Cronbach’s alpha (between 0 and 1) TEST-RETEST RELIABILITY ● T est-retest reliability= assesses the stability oftest scores over time (temporal reliability) ○ Same test administered to the same people at different times ○ Determined by ■ Pearson correlation (but if increase by same amount each time, correlation is still 1) ■ T-test ISSUES WITH TEST-RETEST RELIABILITY ● Issues: ○ ○ ○ ptimal time-interval O Drop-outs Temporal instability of constructs PSYC2017 E xploratory factor analysis (EFA) = used to identify or create a possible latent construct ○ Principle components analysis (PCA) is NOT a type of EFA ■ assumes that all variance in the items can be explained by the same latent construct Confirmatory factor analysis (CFA) = used to confirm or validate an already hypothesised, theorised or identified latent construct TYPES OF FACTORS ● ● rthogonal = dimensions that are considered to be O independent from each other ○ Eg. neurotisism and extraversion Oblique = dimensions that are considered to be related to each other ○ Eg. fluid intelligence and crystalised intelligence ASSUMPTIONS ● ● F actorability = the mathematical and conceptual suitability of an item to be included in a FA model ○ Eg. extraversion is not factorable within a set of neurotisicm items ○ Bartlet’s test Simple structure = items form distinct factors (or spacial clusters), based on the degree of their associations ○ Strong incluster associations, but weak between cluster association ■ Exhibit high levels of convergence and discriminant validity ○ Sphericity is significant when all inter-item correlations are significantly different from zero 3 ○ ross-loading (the opposite to simple structure) C = cross over between factors ● Tend to be orthogonal (independant) ○ ○ They are finite dimensions Measued using psychometrics and statistics ○ Questionnaires, interview, life stories, brain scanes, twin studies etc. STRUCTURE OF PERSONALITY ● ● S phericty is significant when all inter-tem correlations are significantly different from zero ● ● ROTATION ● ● rthogonal rotation = rotation of the axis 90 degrees, to O the point which the axis fit the factors ○ Resulting factors are uncorrelated ● blique rotation = creates lines of best fit for each of the O factors ○ Resulting factors are correlated ○ More complicated method GORDON ALLPORT TRAITS THEORY ● ● RETAINING FACTORS ● ● T he comprehensibility rule = retain factors that are meaningful, clearly interpretable (valid) and measurable (reliable) Kasier criteria, scree plot, joliffe criterion, variance explained rule and parallel analysis are all methods for determining the number of factors to keep in a factor analysis ● ● ersonaility = organised psychobiologcal characteristics P that influence ones behavioir, attitudes, motivations and psychology ○ Probabilistic tendencies (more likely to do something) Shapers of personality ○ Biology and genetics ○ Evolution ○ Environment Personality is made up of dimesions PSYC2017 ersonaility (Allport) = “what a human really is… a P dynamic system of traits” ○ Not a construct, personality is real ■ It is concrete and physically observable ■ Heuristic realism = there is a real universe and we just have some access to it Traits (Allport) = neuropsychic structures that dispose a person towards specific kinds of actions ○ Rooted in the nervous system ○ Innate and environmental ○ Traits make diverse stimuli functionally equivalent ■ Making someone respond the same way to different situations APPROACHES PERSONALITY ● T raits (charcteristics) = general dispositions people possess which can be inferred from their behaviour ○ All humans have all traits to some intensity and importance ○ Structured; dimensitonally and hiearchaically arranged ○ Relatively stable over time but do drift as traits evolve/develop Disposition = more contextual than a trait State = a trait at any given time ○ Eg. mood = states of emotive trait Temperament = characteristic reaction pattern present from an early age (biological) ○ Eg. Calm, angry ○ Fuzzy between temperament and traits ● ● omothetic approach = attempts to establish universal N laws Idiographic approach = attempts to identify unique combination of traits which charcterisies an individual ○ Eg. psychologists use this in practice ○ Allport took this approach as he emphasiesd the uniqueness of the indivudal 4 TYPES OF TRAITS ● ● ● Allport identified over 4500 traits and 3 main groups ○ Cardinal = pervasive, outstanding, and dominating traits (meta-traits) that not all individuals possess ○ Central = the 5-10 traits that best describe an individual ○ Secondary = peripheral traits which are not always present Any trait can fall into any category, depends on its importance in the individuals life Nomothetic common traits (shared traits) ○ Universal norms ○ Group norms THE SELF ● ● PROS AND CONS TO THE THEORY PROS ● ● CONS ● ● Proprium (Alloport) = self (ego) ○ Integration of personal traits Stages of proprium development ○ Developing a sense of self ● ● MOTIVES ● ● ● otives are NOT infantile motivation or unconscious M needs (unlike Freud suggests) ○ Emphasis on the individual’s present state, current situation and feelings Functional autonomy = childhood has some influence on motives but does not directly influence your bahavious now Perseverative functional autonomy = motives that depend on a feedback mechanism or biochemical process resulting in the continuation of a behaviour ○ Eg. smoking Propriate functional autonomy = motives develop by an individual’s wishes or desired goals ○ Eg. interest in a musical instrument ersonaility (Cattell) = the characteristics of the individual P that allow for prediction of how they will behave in a given situation ○ Purely pragmatic, useful, practical Traits (Cattell) = the relatively stable and long-lasting building blocks of personality that posses predictive value ○ Environemental (environmental-mold traits) and gentic (constitutional traits) Multivariate abstract varience analysis (MAVA) ○ Developed MAVA based off heritability coefficients CLASSIFYING TRAITS ● ● ● ● PSYC2017 ctively ignored psychopathology (wanted to focus on A everyday life instead) The identified traits are impractical when it comes to measurement and understanding ○ Never developed any standardized measurement RAYMOND CATTELL TRAITS THEORY ● ● iewed humans as ration active agents (unlike Freudian V determinism) ○ Freudian determinism suggests that you can't help things because of your childhood or unconscious desires Founder of academic field of personality Ability traits = what we use to reach our goals ○ Eg. intelligence, memory, talents Temperament traits = how we reach our goals ○ Eg. easygoing, anxious, layedback ○ Highly heritable Dynamic traits = why we want to reach our goals (motivation) ○ Eg. competitive, cooperative, ambitious ○ Heirarchically arranged in a dynamic lattice ■ Erg → sentiment → attitude ■ Attitudes = overt; expression of interest in things ■ Sentiments = complex; latent ● Eg. interests, values, religiosity ■ Metaergs = attitudes and sentiments ■ Ergs = innate drives Surface Traits = observable and directly measurable ○ Eg. shyness, cheerfulness ○ Cluster together and have high correlation 5
