Personality and Consequential Outcomes: Predictive Validity Across Achievement, Health, and Relationships

Predictive validity of personality: achievement, health, and relationship outcomes

  • Purpose of the lecture: to examine how well personality traits predict consequential life outcomes across three domains: achievement (educational, occupational), health (longevity and health behaviors), and relationships.
  • Core motivation: despite personality being interesting in its own right, it has practical value if personality assessments reliably and validly predict important outcomes.
  • Core theoretical anchor: the lexical hypothesis
    • Important traits are encoded in language; lexical terms describing trait characteristics capture meaningful individual differences.
    • The “importance” of a trait, in historical terms, is its predictive utility for social interactions (e.g., who will help, who can solve problems, who would be a good mate).
  • Predictive validity vs predictive power
    • Predictive validity: does a trait measure predict outcomes that are theoretically relevant to the trait?
    • Predictive power: does the trait predict practically important outcomes in real-world settings?
  • Modeling framework for trait-outcome links
    • Direct effects: general-to-specific inferences where a broad trait predicts a specific instance of behavior or outcome.
    • Analogy: a city’s general climate predicts what you should wear that day.
    • Example: higher conscientiousness predicting higher effort-related performance on tasks
      Y = eta0 + eta1 X + eta_2 Z + \, \epsilon
      where X is conscientiousness, Y is a performance criterion, Z represents other covariates.
    • Indirect effects (statistical mediation): a trait predicts an outcome via a mediator variable.
    • Example: situation selection mediates the link between a trait and an outcome.
    • Mediation path: X → M → Y; total effect c = direct effect c′ + (a × b).
    • Notation: path a is X → M, path b is M → Y, path c is total effect from X to Y, and c′ is the direct effect of X on Y controlling for M.
    • Interactive effects (person × environment interactions): traits predict outcomes differently depending on environmental context.
    • Example: extraversion predicting leadership success more strongly in environments with high interpersonal demands.
  • Historical context: measurement and prediction in psychology
    • Early tests aimed to predict educational and occupational outcomes by measuring cognitive abilities and later, personality.
    • Binet & Simon: origin of cognitive ability assessment (Stanford–Binet) to identify children needing special education; later evolved into SAT/college admissions.
    • Yerkes (WWI): army selection and placement; catalyzed later industrial psychology emphasis on selection and prediction.
    • Schmidt & Hunter (late 1990s): meta-analysis of predictors of job performance; cognitive ability and personality factors examined together.
  • Meta-analytic findings on prediction of job performance
    • Predictors examined: cognitive ability, personality traits (Big Five), work experience, references, structured vs unstructured interviews, integrity tests.
    • Weakest predictors: years of education (variable effects, generally weaker).
    • Moderate predictors: job experience, references; structured interviews outperform unstructured.
    • Personality findings:
    • Conscientiousness: strongest Big Five predictor of job performance across occupations; average correlations around
      r ≈ 0.31r \,\approx\, 0.31
      (Barrick & Mount; Schmidt & Hunter) – considered medium-to-large depending on guidelines.
    • Integrity tests: among the strongest predictors of job performance, blending conscientiousness and agreeableness aspects.
    • Cognitive ability: overall strongest single predictor; correlation around r≈0.51r \approx 0.51 for job performance. Importantly, personality adds predictive power beyond cognitive ability (additive validity): e.g.,
      • Cognitive ability alone: r=0.51r = 0.51
      • Cognitive ability + conscientiousness: approximate multiple correlation R=0.60R = 0.60
      • Cognitive ability + conscientiousness + integrity tests: R=0.65R = 0.65 (large effects by some guidelines)
    • The Big Five overview (Barrick & Mount, Hertz & Donovan): conscientiousness predicts broad performance, especially effort-related criteria; extraversion predicts in certain roles (e.g., management, sales); openness, agreeableness, and low neuroticism show some role depending on job type (customer-facing, highly social roles).
    • Reference checks and interviews
    • Structured references and structured interviews yield better predictive validity than unstructured ones.
    • Structured reference checks can explicitly assess conscientiousness, neuroticism, leadership, and interpersonal skills; predictive validity for performance observed in the neighborhood of 0.26$ to 0.51 for those ratings.
  • Distinctive role of cognitive ability and conscientiousness in predicting achievement
    • Across occupation types, cognitive ability is a robust predictor of job performance; personality adds extra predictive power in many cases.
    • Conscientiousness emerges as a broad predictor across multiple job types, particularly for effort-related criteria rather than purely technical skills.
    • A 2000s meta-analysis (Hertz & Donovan) confirms similar patterns: conscientiousness broadly predictive; agreeableness, openness, and low neuroticism predictive in roles involving interpersonal interaction; extraversion linked to management/sales; neutral or mixed results across settings.
  • Education and attainment
    • In predicting GPA and academic performance:
    • Cognitive ability is the strongest predictor across U.S. programs; conscientiousness adds predictive value beyond cognitive ability.
    • Openness and agreeableness show more modest associations.
    • In Australian data, Poropat (2009) found conscientiousness adds to GPA prediction beyond cognitive ability; openness/agreeableness do not add once cognitive ability is accounted for.
    • Highest level of education attained
    • Openness is the strongest predictor of years of full-time education (curiosity, information reward sensitivity, interest in ideas and study).
    • Openness also predicts indicators of educational engagement and breadth/depth of reading.
    • On college major choice, personality patterns can show systematic, though not deterministic, tendencies (e.g., psychology students often higher in agreeableness and openness).
  • Health outcomes and longevity
    • Longitudinal work spanning decades shows conscientiousness as a robust predictor of longevity and health behaviors.
    • Classic findings: consistently higher conscientiousness in childhood/adulthood relates to lower mortality risk across decades; patterns replicated (e.g., Jackson, Your Friends Know How Long You Will Live; gender differences noted in one study but not consistently across replications).
    • Health-promoting behaviors as a key mechanism
    • Conscientiousness predicts engagement in healthier behaviors (less alcohol/drug use, healthier eating, safer driving, safer sexual behavior).
    • Interventions: changes in conscientiousness over time predict changes in health behaviors and, in turn, current health.
    • Longitudinal work by Takahashi et al. suggests changes in conscientiousness across time relate to changes in health behaviors and health status.
    • Type A personality and hostility
    • Type A (Friedman & Rosenman) linked to cardiovascular risk, but psychometric support for the Type A cluster is weak.
    • Trait hostility (a facet of low agreeableness) has been associated with higher cardiovascular risk; this has more robust psychometric standing than the broader Type A construct.
    • Meta-analytic summary (narrative): conscientiousness most robust predictor of all-cause mortality; optimism linked to lower risk; hostility linked to higher risk; effect sizes are small in some samples but potentially clinically meaningful when aggregated across populations.
  • Relationships and couple dynamics
    • Assortative mating: couples tend to be similar on openness and agreeableness, suggesting similarity rather than complementarity; neuroticism shows no clear assortative pattern.
    • Environmental channeling and selective breakup as alternative explanations for similarity over time
    • Longitudinal couples studies (German sample; ~1000 couples over 16 years) show:
    • Baseline similarity across several traits (extraversion, agreeableness, openness, conscientiousness) but not neuroticism.
    • No strong evidence for convergence over time; if anything, slight divergence rather than convergence.
    • Actor and partner effects in relationship outcomes
    • Actor effects: your own personality predicts your own relationship outcomes (e.g., higher agreeableness/conscientiousness linked to higher relationship satisfaction; low agreeableness/conscientiousness linked to higher infidelity risk for the actor).
    • Partner effects: your partner’s personality affects your relationship outcomes (e.g., partner’s agreeableness/conscientiousness linked to your satisfaction).
    • Relationship dissolution (divorce) and personality
    • Higher neuroticism, openness, and extraversion linked to increased likelihood of dissolution; higher agreeableness and conscientiousness linked to lower likelihood of dissolution.
    • Some studies also point to enduring dynamics (stability of the trait–relationship link over time) as a key mechanism rather than emergent distress (changes in the link over time).
    • Interactions and dynamics in relationships
    • Eaton & Funder’s conceptual model emphasizes bidirectional influences among actor/partner traits, behaviors, and impressions, affecting the social environment and potentially personality over time.
    • A coworker dynamics study (20-day experience sampling) showed that people low on agreeableness were more reactive to others’ behavior; high Agreeableness individuals were less reactive to others’ behavior.
    • Practical synthesis for relationships
    • Overall pattern: conscientiousness and agreeableness tend to support better relationship outcomes (lower dissolution; higher satisfaction), while higher neuroticism is linked to more negative dynamics.
  • The replication crisis and robustness of findings in personality outcomes
    • Replication crisis summary: many psychological findings fail to replicate; concerns about questionable research practices, publication bias, and low statistical power.
    • General replication project (Open Science Collaboration, 2015): across 100 findings, about 39% replicated; original effects often had larger effect sizes than replications (typical original ~0.25–0.35; replicated ~half that magnitude).
    • Life outcomes of personality project (Soto, 2019): replication of 78 previously reported associations between personality and life outcomes; 87% replicated with about 75% of original effect sizes.
    • Generalizability concerns and sample diversity: subsequent replication work generally finds high replication rates across sex and age groups, with some caveats about racial/ethnic diversity.
    • Implications for theory and practice
    • Predictive validity remains a credible concept; predictive power may be smaller in replication than in initial studies, but findings are still meaningful, especially when considering population-level effects.
  • Predictive validity vs practical implications
    • Predictive validity: conscientiousness consistently related to effort-related performance and health behaviors; cognitive ability often strongest predictor for performance, but personality adds incremental validity.
    • Predictive power: even small correlations can be meaningful across large populations (e.g., small effect sizes accumulate to meaningful public health or organizational outcomes).
    • Policy and intervention considerations
    • Interventions to alter conscientiousness or agreeableness are controversial but explored in literature on behavior change and development.
    • Interventions may need to be tailored to individual differences (the “One Size Doesn’t Fit All” viewpoint) for effectiveness in education, health, and work contexts.
  • Reading for this week and notes on interpretation
    • The primary reading for the week is Soto (2019) on life outcomes and personality; earlier references and slides may differ due to an update in the online reading list.
    • The broader takeaway: while there is robust evidence that personality predicts meaningful life outcomes, researchers emphasize robustness, replication, and mechanism (direct, indirect, and interactive pathways) when interpreting these associations.
  • Important methodological considerations highlighted in the lecture
    • Distinguish correlation from causation: observational associations do not confirm causal direction.
    • Consider reverse causation and bidirectional effects in mediation models.
    • Use longitudinal data and cross-lagged designs where possible to test directionality.
    • Recognize state vs trait components in personality assessment; repeated measures help separate stable traits from temporary states.
  • Examples and concrete findings mentioned in the lecture
    • Job performance predictors ( Schmidt & Hunter )
    • Cognitive ability: r \approx 0.51</li><li>Conscientiousness:</li> <li>Conscientiousness:r \approx 0.31</li><li>Combinedmodels:cognitiveability+conscientiousness:</li> <li>Combined models: cognitive ability + conscientiousness:R \approx 0.60;addinganintegritytestraisesto; adding an integrity test raises toR \approx 0.65</li><li>Occupationaloutcomesandjobtype</li><li>Conscientiousnesspredictsbroadperformance;effectstrongerfor“effort−related”criteriathan“skill−related”criteria.</li><li>Extraversionpredictsperformanceinmanagementandsalescontexts;lowneuroticismalsopredictiveinsomeclient−facingroles.</li><li>Educationaloutcomes</li><li>GPA:cognitiveabilityisthestrongestpredictor;conscientiousnessaddsbeyondcognitiveabilityinU.S.data;OpennessandAgreeablenessshowsmallerornoadditionalpredictivevaluebeyondcognitiveabilityinsomecontexts.</li><li>Highesteducationlevel:Opennessstrongestpredictor;associatedwitheducationengagement,breadth/depthofreading,andchoiceofcollegemajor(e.g.,psychologystudentsoftenhigherinagreeablenessandopenness).</li><li>Healthoutcomes</li><li>Longevity:conscientiousnessrobustlylinkedtolifespanacrossdiversesamples;connectionmediatedbyhealth−promotingbehaviors.</li><li>Healthbehaviors:conscientiousnesspredictsengagementinhealthfulactivities;changesinconscientiousnessovertimepredictchangesinhealthbehaviorsandhealthstatus.</li><li>TypeA/hostility:hostilitylinkedtocardiovascularrisk;broaderTypeAconstructlacksrobustpsychometricsupport.</li><li>Relationships</li><li>Assortativematingforopennessandagreeableness;littleevidenceforconvergenceovertime;somegender−specificnuancesinspecificstudies.</li><li>Relationshipoutcomes(satisfactionanddissolution)linkedtoconscientiousnessandagreeableness(positive)andneuroticism(negative).</li><li>Enduringdynamicsvsemergentdistress:earlyevidencesupportsenduringdynamicsasastablelinkbetweentraitandrelationshipsatisfaction,withimplicationsfordissolutionrisk.</li></ul></li><li>Quicktakeawaysforexampreparation<ul><li>Conscientiousnessisthemostrobustunconditionalpredictorofachievementandhealthoutcomesacrossdomains;itseffectsareoftenmediatedbyhealth−promotingbehaviorsandwork−relatedeffort.</li><li>Cognitiveabilityisaverystrongpredictorofachievement−relatedoutcomes(especiallyjobperformance)butpersonalityaddsincrementalvalidity.</li><li>Opennessisastrongpredictorofeducationalattainment(highestdegree)andengagement;conscientiousnessaddsbeyondabilityinseveraleducationalcontexts.</li><li>Inhealth,conscientiousnessrelatestolongevitylargelyviahealthbehaviors;hostility(afacetoflowagreeableness)islinkedtocardiovascularrisk.</li><li>Inrelationships,conscientiousnessandagreeablenesspredictbetterrelationshipfunctioning;neuroticismtendstopredicthigherdissolutionrisk;assortativematingcontributestotraitsimilarityinsomedomains.</li><li>Replicationdatagenerallysupportrobustassociations,buteffectsizesareoftensmallerinreplicationthanininitialstudies;thereplicationcrisisemphasizesusingrobustmethods,preregistration,andlarger,diversesamples.</li></ul></li><li>Reflectionpromptsfortheexam<ul><li>Howdodirect,indirect,andinteractiveeffectshelpexplainwhyabroadpersonalitytraitpredictsaspecificoutcome?</li><li>Whymightconscientiousnesspredicthealthbehaviorsandlongevityacrossthelifespan?Whatmechanismscouldaccountforthis?</li><li>Compareandcontrastthepredictivevalidityofconscientiousnessvs.cognitiveabilityforeducationalandoccupationaloutcomes.</li><li>Whataretheimplicationsofthereplicationfindingsforapplyingpersonalityassessmentineducationalororganizationalsettings?</li></ul></li><li>Noteonterminologyusedinthereading<ul><li>“Assortativemating”referstosimilarity−basedpairing;“environmentalchanneling”referstonormativechangesinpersonalityduetorelationshipenvironments;“enduringdynamics”referstostabletrait–outcomeassociationsovertime;“emergentdistress”referstochangesinthestrengthorformoftheseassociationsovertime.</li></ul></li></ul><p></li> <li>Occupational outcomes and job type</li> <li>Conscientiousness predicts broad performance; effect stronger for “effort-related” criteria than “skill-related” criteria.</li> <li>Extraversion predicts performance in management and sales contexts; low neuroticism also predictive in some client-facing roles.</li> <li>Educational outcomes</li> <li>GPA: cognitive ability is the strongest predictor; conscientiousness adds beyond cognitive ability in U.S. data; Openness and Agreeableness show smaller or no additional predictive value beyond cognitive ability in some contexts.</li> <li>Highest education level: Openness strongest predictor; associated with education engagement, breadth/depth of reading, and choice of college major (e.g., psychology students often higher in agreeableness and openness).</li> <li>Health outcomes</li> <li>Longevity: conscientiousness robustly linked to lifespan across diverse samples; connection mediated by health-promoting behaviors.</li> <li>Health behaviors: conscientiousness predicts engagement in healthful activities; changes in conscientiousness over time predict changes in health behaviors and health status.</li> <li>Type A/hostility: hostility linked to cardiovascular risk; broader Type A construct lacks robust psychometric support.</li> <li>Relationships</li> <li>Assortative mating for openness and agreeableness; little evidence for convergence over time; some gender-specific nuances in specific studies.</li> <li>Relationship outcomes (satisfaction and dissolution) linked to conscientiousness and agreeableness (positive) and neuroticism (negative).</li> <li>Enduring dynamics vs emergent distress: early evidence supports enduring dynamics as a stable link between trait and relationship satisfaction, with implications for dissolution risk.</li></ul></li> <li>Quick takeaways for exam preparation<ul> <li>Conscientiousness is the most robust unconditional predictor of achievement and health outcomes across domains; its effects are often mediated by health-promoting behaviors and work-related effort.</li> <li>Cognitive ability is a very strong predictor of achievement-related outcomes (especially job performance) but personality adds incremental validity.</li> <li>Openness is a strong predictor of educational attainment (highest degree) and engagement; conscientiousness adds beyond ability in several educational contexts.</li> <li>In health, conscientiousness relates to longevity largely via health behaviors; hostility (a facet of low agreeableness) is linked to cardiovascular risk.</li> <li>In relationships, conscientiousness and agreeableness predict better relationship functioning; neuroticism tends to predict higher dissolution risk; assortative mating contributes to trait similarity in some domains.</li> <li>Replication data generally support robust associations, but effect sizes are often smaller in replication than in initial studies; the replication crisis emphasizes using robust methods, preregistration, and larger, diverse samples.</li></ul></li> <li>Reflection prompts for the exam<ul> <li>How do direct, indirect, and interactive effects help explain why a broad personality trait predicts a specific outcome?</li> <li>Why might conscientiousness predict health behaviors and longevity across the lifespan? What mechanisms could account for this?</li> <li>Compare and contrast the predictive validity of conscientiousness vs. cognitive ability for educational and occupational outcomes.</li> <li>What are the implications of the replication findings for applying personality assessment in educational or organizational settings?</li></ul></li> <li>Note on terminology used in the reading<ul> <li>“Assortative mating” refers to similarity-based pairing; “environmental channeling” refers to normative changes in personality due to relationship environments; “enduring dynamics” refers to stable trait–outcome associations over time; “emergent distress” refers to changes in the strength or form of these associations over time.</li></ul></li> </ul> <p> ext{Key correlations and formulas to remember:}<br/><br />r{ ext{cognition-pf}} \approx 0.51r{ ext{conscientiousness-pf}} \approx 0.31<br/><br />R^2{ ext{cognition-only}} \approx 0.51^2 \approx 0.26R^2{ ext{cognition+conscientiousness}} \approx 0.60^2 \approx 0.36<br/><br />R^2_{ ext{cognition+conscientiousness+integrity}} \approx 0.65^2 \approx 0.42<br/><br />c = c' + a b \, ext{(mediation)}$$

      • Reading cue: Soto et al. (2019) Life Outcomes and Personality meta-synthesis; replication rates around 87% for overall associations with similar effect sizes to the original findings; gender and ethnicity generalizability generally reasonable in larger replications.
      • Break period note from the instructor (context): reading updates and slight changes to which paper is designated as the primary reading for the week; always align with the LMS listing for exam preparation.