Download Understanding Personality: Traits, Response Functions, and Measured Personality and more Study Guides, Projects, Research Personality Psychology in PDF only on Docsity!
Personality Psychology and Economics^1
Mathilde Almlund University of Chicago Department of Economics
Angela Lee Duckworth University of Pennsylvania
James Heckman University of Chicago University College Dublin American Bar Foundation Cowles Foundation, Yale University
Tim Kautz University of Chicago Department of Economics
(^1) This research was supported by grants from NIH R01-HD054702, R01-HD065072, and K01-AG033182; the University of Chicago; The Institute for New Economic Thinking (INET); A New Science of Virtues: A Project ofthe University of Chicago; the American Bar Foundation; a conference series from the Spencer Foundation; the JB & MK Pritzker Family Foundation; the Buffett Early Childhood Fund; and the Geary Institute, University CollegeDublin, Ireland. The opinions expressed in this report are those of the authors and do not necessarily reflect the views of any of the funders. Amanda Agan and Pietro Biroli are major contributors to this essay through theirsurveys of the effect of personality on crime (presented in Web Appendix A7.B) and health (presented in Web Appendix A7.A), respectively. We are grateful to Pia Pinger for her analyses of the German Socio-Economic Panel(GSOEP) survey data. We have benefited from comments received from Amanda Agan, Dan Benjamin, Pietro Biroli, Dan Black, Daniel Cervone, Deborah Cobb-Clark, Flavio Cunha, Kathleen Danna, Thomas Dohmen, StevenDurlauf, Joel Han, Moshe Hoffman, John Eric Humphries, Miriam Gensowski, Bob Krueger, Jongwook Lee, Xiliang Lin, Dan McAdams, Terrance Oey, Lawrence Pervin, Pia Pinger, Armin Rick, Brent Roberts, MollySchnell, Bas ter Weel, and Willem van Vliet. We also benefited from a workshop at the University of Illinois, Department of Psychology, on an early draft of this paper and presentations of portions of this paper at theSpencer/INET workshop at the University of Chicago, December 10-11, 2010. Additional material that supplements the text is presented in a Web Appendix (http://jenni.uchicago.edu/personality_economics/). Parts of this paperbuild on an earlier study by Borghans, Duckworth, Heckman et al. [2008].
Abstract This paper explores the power of personality traits both as predictors and as causes of academic and economic success, health, and criminal activity. Measured personality is interpreted as a construct derived from an economic model of preferences, constraints, and information. Evidence is reviewed about the “situational specificity” of personality traits and preferences. An extreme version of the situationist view claims that there are no stable personality traits or preference parameters that persons carry across different situations. Those who hold this view claim that personality psychology has little relevance for economics. The biological and evolutionary origins of personality traits are explored. Personality measurement systems and relationships among the measures used by psychologists are examined. The predictive power of personality measures is compared with the predictive power of measures of cognition captured by IQ and achievement tests. For many outcomes, personality measures are just as predictive as cognitive measures, even after controlling for family background and cognition. Moreover, standard measures of cognition are heavily influenced by personality traits and incentives. Measured personality traits are positively correlated over the life cycle. However, they are not fixed and can be altered by experience and investment. Intervention studies, along with studies in biology and neuroscience, establish a causal basis for the observed effect of personality traits on economic and social outcomes. Personality traits are more malleable over the life cycle compared to cognition, which becomes highly rank stable around age 10. Interventions that change personality are promising avenues for addressing poverty and disadvantage.
- Introduction Contents
- Psychology 2.A. A Brief History of Personality Psychology 2. Personality and Personality Traits: Definitions and a Brief History of Personality
- Conceptualizing Personality and Personality Traits Within Economic Models
- 3.A. An Approach Based on Comparative Advantage
- 3.B. Allowing for Multiple Tasking
- 3.C. 3.D. Identifying Personality TraitsExtensions of the Roy Model
- 3.E. Adding Preferences and Goals
- 3.F. Adding Learning and Uncertainty
- 3.G. Definition of Personality Within an Economic Model...................................................
- 3.H. 3.I. Life Cycle DynamicsRelationship of the Model in This Section to Existing Models in Personality Psychology
- Measuring Personality
- 4.A. Linear Factor Models
- 4.B. 4.C. Discriminant and Convergent Validity 48Predictive Validity..........................................................................................................
- 4.D. Faking
- 4.E. The Causal Status of Latent Variables
- Implementing the Measurement Systems
- 5.A. 5.B. CognitionPersonality Traits............................................................................................................
- 5.C. Operationalizing the Concepts
- 5.D. Personality Constructs
- 5.D.1. Self-Esteem and Locus of Control Are Related to Big Five Emotional Stability
- 5.E. 5.D.2. IQ and Achievement Test Scores Reflect Incentives and Capture Both Cognitive and Relating the Big Five to Measures of Psychopathology
- Personality Traits
- 5.F. The Evidence on the Situational Specificity Hypothesis
- Personality and Preference Parameters
- 6.A. 6.B. Evidence on Preference Parameters and Corresponding Personality Measures 94Mapping Preferences into Personality..........................................................................
- 6.C. Do Measured Parameters Predict Real World Behavior?
- 6.D. Integrating Traits into Economic Models.....................................................................
- 6.D.1. Traits as Constraints
- 6.D.2.The Predictive Power of Personality Traits 125 Traits as Preferences
- 7.A. Educational Attainment and Achievement
- 7.B. Labor Market Outcomes...............................................................................................
- 7.C. Personality and Health
- 7.D. Crime
- 8.A.Stability and Change in Personality Traits and Preferences 170 Broad Evidence on Changes in Traits over the Life Cycle
- 8.B. Evidence on Ontogenic and Sociogenic Change..........................................................
- 8.C. External Changes to Biology........................................................................................
- Interventions 185 8.D.1. Evidence of Change in Traits from Other Studies of Parental Investment 8.D. The Evidence on the Causal Effects of Parental Investment, Education, and
- 8.D.2. The Effects of Schooling on Cognitive and Personality Traits.............................
- 8.D.3. Evidence from Interventions
- 8.E. Stability of Economic Preference Parameters
- Summary and Conclusions
Figure 1 shows that, by age ten, treatment group mean IQs were the same as control group mean IQs. Many critics of early childhood programs seize on this and related evidence to dismiss the value of early intervention studies.^7 Yet on a variety of measures of socioeconomic achievement, the treatment group was far more successful than the control group.^8 The annual rate of return to the Perry Program was in the range 6-10% for boys and girls separately.^9 These rates of return are statistically significant and above the returns to the US stock market over the post-war period.^10 The intervention changed something other than IQ, and that something produced strong treatment effects. Heckman, Malofeeva, Pinto et al. [2010] show that the personality traits of the participants were beneficially improved in a lasting way.^11 This chapter is about those traits.
(^78) See the Westinghouse study of Head Start (Project Head Start [1969]). 9 See Heckman, Malofeeva, Pinto et al. [2010], and Heckman, Moon, Pinto et al. [2010a].See Heckman, Moon, Pinto et al. [2010b]. (^1011) See DeLong and Magin [2009] for estimates of the return on equity. Participants of both genders had better “externalizing behavior” while girls also had improved “internalizingWe discuss this evidence in Section 8. The traits changed were related to self-control and social behavior. behavior.” See Heckman, Malofeeva, Pinto et al. [2010]. Duncan and Magnuson [2010b] offer a differentinterpretation of the traits changed by the Perry experiment, but both analyses agree that it was not a boost in IQ that improved the life outcomes of Perry treatment group members.
Figure 1. Perry Preschool Program: IQ, by Age and Treatment Group
Notes: IQ measured on the Stanford-Binet Intelligence Scale (Terman and Merrill [1960]). Test wasadministered at program entry and each of the ages indicated. Source: Cunha, Heckman, Lochner et al. [2006] and Heckman and Masterov [2007] based on dataprovided by the High Scope Foundation.
Personality psychologists mainly focus on empirical associations between their measures of personality traits and a variety of life outcomes. Yet for policy purposes, it is important to know mechanisms of causation to explore the viability of alternative policies.^12 We use economic theory to formalize the insights of personality psychology and to craft models that are useful for exploring the causal mechanisms that are needed for policy analysis. We interpret personality as a strategy function for responding to life situations. Personality traits, along with other influences, produce measured personality as the output of personality strategy functions. We discuss how psychologists use measurements of the performance of persons on tasks or in taking actions to identify personality traits and cognitive
(^12) See Heckman [2008a].
What is the relationship between standard measures of personality and measures of psychopathology and child temperament? (3) What is the relationship between economic preference parameters and psychological measurements? (4) How stable across situations and over the life cycle are preference parameters and personality traits? (5) What is the evidence on the predictive power of cognitive and personality traits? (6) What is the evidence on the causal power of personality on behavioral outcomes? (7) Can personality be altered across the life cycle? Are interventions that change personality traits likely fruitful avenues for policy? (8) Do the findings from psychology suggest that conventional economic theory should be enriched? The plan of the paper is as follows. Section 2 presents a definition of personality that captures central ideas in the literature on personality psychology. It also presents a brief history of personality psychology and the person-situation debate that paralyzed the field for 20 years and that still influences behavioral economics. Section 3 defines measured personality as a response function using an economic model of preferences, expectations, and constraints. Our model distinguishes measured personality from personality traits. We interpret personality as a response function mapping variables that characterize traits and situations to manifest (measured) personality. Our definition formalizes various definitions of personality used in the literature on personality psychology and facilitates the analysis of personality using the tools of economics. We also sketch a dynamic model of trait formation.
Section 4 discusses alternative criteria that psychologists use to define traits. It examines the strengths and limitations of each approach. We link our abstract definition to linear factor models that are commonly used to identify personality and cognitive traits. Section 5 presents the main systems used to measure personality and cognition and discusses the relationship among the systems. We illustrate a nonidentification result developed in Section 3 by showing how scores on IQ tests are greatly affected by incentives and context. We present additional evidence showing that the scores on achievement tests depend on cognitive and personality measurements, with a substantial predictive role for personality measures. Measures of “IQ” commonly used in economics and social science conflate measures of cognition and personality. Section 6 discusses economic preferences and examines the evidence relating economic preference parameters to psychological parameters. Section 7 surveys the evidence on the predictive validity of personality measures for education, crime, health, and labor market outcomes. The material presented in the main text summarizes a large and growing empirical literature. A Web Appendix presents additional detail on the literature relating cognition and personality in each of these areas of economic and social life.^16 Section 8 presents evidence on the causal impact of personality on outcomes, as well as evidence on the stability and malleability of personality traits and preferences. We extend the theoretical framework for trait formation introduced in Section 3 and discuss a corresponding measurement system. We discuss the evidence from intervention studies. Section 9 concludes with provisional answers to the eight questions.
(^16) The Web Appendix can be found online at http://jenni.uchicago.edu/personality_economics/. Amanda Agan and Pietro Biroli are authors of some of these surveys as noted in the appendix.
There are many different models of personality.^19 A prototypical model that captures many features of a wide class of models in personality psychology is one due to Roberts [2006]. He presents the schematic displayed in Figure 2 to relate personality traits to behavior.^20 He distinguishes mental abilities from personality traits (the items in the boxes will be discussed in later sections of this chapter). These, along with preferences (motives, interests, and values) and narratives (the stories people tell themselves in organizing their lives and making meanings of them), shape one’s identity and reputation, including the views of the person by others and the person’s perception of how others perceive him. Identity and reputation in turn shape the roles of individuals in the economy and the society and the larger culture to which they belong. Personality is the system of relationships that map traits and other determinants of behavior into measured actions. In Roberts' vision of personality, feedback processes operate among all components of Figure 2. Thus his broad conception of personality includes the possibility that identity shapes traits and abilities, perhaps through a mechanism such as epigenetics, in which environment affects gene expression.^21 Measured personality results from interactions among components of the system. Personality traits are one determinant of personality and need to be carefully distinguished from the full expression of personality, which is generated by the traits interacting with other factors. Personality is seen as a system of behaviors, thoughts, and feelings that emerge from the interacting components.
(^1920) See the models in John, Robins and Pervin [2008]. presents a formal model.Graphical models like Figure 2 are the rule in personality psychology. Explicit formal models are rare. Section 3 (^21) See, e.g., Rutter [2006a].
Figure 2. Roberts’s Model of Personality as the Output of a System
Source: Roberts [2006].
In Section 3, we formalize aspects of Roberts’ framework for personality within an economic model of production, choice, and information. Figure 2 presages our discussion of a basic identification problem discussed in Section 3. Measurements and behaviors that arise from responses to incentives and interactions with culture are used to infer personality traits and abilities. Personality traits and cognitive abilities, along with the other “units of analysis” in Figure 2, produce the observed behaviors that are used to infer the generating traits. To infer traits from behaviors requires “parsing out” or standardizing for all of the other contributing factors that produce the observed behavior—a challenging task. The inability to parse and localize behaviors that depend on a single trait or ability gives rise to a fundamental
Arthur Jensen, an intellectual heir of Spearman who is widely regarded as a proponent of g as an explanatory factor of success and failure in many domains of life, writes: “What are the chief personality traits which, interacting with g , relate to individual differences in achievement and vocational success? The most universal personality trait is conscientiousness, that is, being responsible, dependable, caring, organized and persistent” Jensen [1998, p. 575].
The Pioneers of Personality Psychology Over the past century, interest in personality among psychologists has fluctuated dramatically. During the first half of the twentieth century, many of the most prominent psychologists (e.g., Gordon Allport, Raymond Cattell, Hans Eysenck, Charles Spearman, Lewis Terman) were vigorously engaged in the study of individual differences in behaviors and traits. Psychologists studied personality traits along with intelligence, interests, and motivation and measured differences and similarities within and across individuals. A systematic approach to the study of personality was conceived by early psychologists who believed that the most important dimensions on which human beings differed would be captured in natural language. These personality pioneers extracted words from the (English) dictionary that characterized individual differences between people (e.g., irritable, proud), after eliminating synonyms and words not associated with traits. They designed and administered studies of trait inventories to large samples of individuals and applied the same factor analytic methods developed by Galton, Spearman, Binet, Pearson, Cattell, and Thorndike to these assessments in order to isolate “ g^ ” to identify the structure of cognitive abilities.
The fruits of several decades of research in this tradition beginning in the 1970s have produced a widely (but not universally) shared consensus taxonomy of traits, known as the Big Five, that is arrived at through factor analysis of observer and self-reports of behaviors.^25 The Big Five posits a hierarchical organization for personality traits, with five factors at the highest level and progressively more narrowly defined traits (or facets) at lower and lower levels.
Trait Table 1. The Big Five TraitsDefinition of Trait* I. Openness to Experience (Intellect) The tendency to be open to newaesthetic, cultural, or intellectual experiences. II. Conscientiousness The tendency to be organized,responsible, and hardworking. III. Extraversion An orientation of one’s interests andenergies toward the outer world of people and things rather than the innerworld of subjective experience; characterized by positive affect andsociability. IV. Agreeableness The tendency to act in a cooperative,unselfish manner. V. Neuroticism (Emotional Stability) Neuroticism is a chronic level ofemotional instability and proneness to psychological distress.Emotional stability is predictability and consistency in emotional reactions,with absence of rapid mood changes.
- From the American Psychological Association Dictionary [2007]. Table 1 presents the Big Five traits. They are Openness to Experience (also called Intellect or Culture), Conscientiousness, Extraversion, Agreeableness, and Neuroticism (also
(^25) Goldberg [1993], Barenbaum and Winter [2008], John and Srivastava [1999], Krueger and Johnson [2008] discuss the Big Five.
focused on the importance of the situation compared to the individual traits featured in personality psychology. Mischel noted that correlations between behavioral task measures of personality and questionnaire measures seldom, if ever, exceeded 0.3.27,28^ The implication of such within- individual behavioral heterogeneity suggested to Mischel that “the behaviors which are often construed as stable personality trait indicators are highly specific and depend on the details of the evoking situations and the response mode employed to measure them” (p. 37). Mischel wrote “… with the possible exception of intelligence, highly generalized behavioral consistencies have not been demonstrated, and the concept of personality traits as broad dispositions is thus untenable ” – Mischel [1968, p. 146] Mischel went on to write that global (i.e., domain-general) traits (e.g., “impulsive”, “confident”) measured in one situation did not predict future behavior and outcomes in other situations. His view was that global traits, in attempting to summarize behavioral dispositions without regard to situational contingencies, were “excessively crude, gross units to encompass adequately the extraordinary complexity and subtlety of the discriminations that people constantly make” (p. 301). Mischel now suggests [2004] that behaviors can be consistent across time, but that the locus of consistency is to be found in highly contextualized if-situation/then-behavior contingencies (e.g., “ If I feel threatened, then I am aggressive”). Variance across situations was,
(^27) There is great irony in the fact that none of the correlations of cognitive measures with outcomes that are reported in Table A1 in the Web Appendix are as high as 0.3, but no one questions the power of cognition in predictingoutcomes in social life. Few studies in social psychology show correlations as high as 0.2 (see Richard, Bond and Stokes-Zoota [2003]). (^28) Psychologists often work with standardized variables (variables normalized by standard deviations). They report correlations between standardized variables as “effect sizes.”
in Mischel’s view, improperly treated by most personality psychologists as “error.”^29 Indeed, in his view, the systematic variation of behavior across situations points to underlying motivations, beliefs, schemas, strategies, and other factors that collectively and interactively give rise to coherence in any individual’s measured personality. His revised view of personality is broadly consistent with Robert’s Figure 2. In Section 3, we formalize the “if-then” relationship using an economic model. We show that the person-situation debate boils down to an empirical question about the relative importance of person, situation, and their interaction in explaining behaviors. Although Mischel may have intended otherwise, proponents of the situationist view have used his monograph as ammunition in the battle against accepting evidence from personality psychology into economics. Like most heated debates in social science, this one occurred in the absence of much data. In Section 5, we discuss the body of evidence that has emerged over the past four decades on the existence of stable personality traits. The debate over the relative importance of person and situation in the 1960s and 1970s reflected deeper currents in psychology and social science more generally, that still run strong. Behaviorism, associated with B. F. Skinner was influential. It posited that experience explains all aspects of behavior. There was the widely held notion that situation and experience were all powerful—that people were born as blank slates.^30 This captured the interventionist spirit of the times. Inter-individual heterogeneity in traits was ignored. Ross and Nisbett [1991] summarize the position of many social psychologists:
(^2930) I.e., unobserved heterogeneity. Pinker [2002].
