Luca Corazzini Silvia D’Arrigo

Emanuele Millemaci Pietro Navarra

The Influence of Personality

Traits on University Performance:

Evidence from Italian Freshmen Students

ISSN: 1827-3580 No. 19/WP/2020

W o r k i n g P a p e r s

D e p a r t m e n t o f E c o n o m i c s

C a ’ F o s c a r i U n i v e r s i t y o f V e n i c e

N o . 1 9 / W P / 2 0 2 0

ISSN 1827-3580

The Working Paper Series

is available only on line

(http://www.unive.it/pag/16882/)

For editorial correspondence, please contact:

wp.dse@unive.it

Department of Economics

Ca’ Foscari University of Venice

Cannaregio 873, Fondamenta San Giobbe

30121 Venice Italy

Fax: ++39 041 2349210

The Influence of Personality Traits on University Performance: Evidence from Italian Freshmen Students

Luca Corazzini Depar tment o f Economic s and VERA, Ca’ Foscari University of Venice

Silvia D’Arrigo Depar tment o f Economic s and Bus ine s s , Univ e rs i t y o f Sas sar i

Emanuele Millemaci Depar tment o f Economic s , Univ e rs i t y o f Me s s ina

Pietro Navarra Depar tment o f Economic s , Univ e rs i t y o f Me s s ina

Abstract: Despite several at tempts to prov ide a def ini te pat te rn regarding the e f fect s of personal i t y t ra i t s on performance in higher educat ion, the debate over the nature of the re lat ionship is far f rom be ing conc lus ive . The use of d if fe rent subject pools and sampl e s izes , as we l l as the use of ident i f icat ion s t rateg ies that e i ther do not adequate ly account for se lec t ion bias o r are unable to es tabl ish causal i t y be tween measures of academic performance and noncognit ive sk i l l s , are poss ible sources of heterogene i ty . This paper inves t igates the impact of the B ig F ive t ra i t s , as measured before the beg inning of the academic year , on the grade point average achieved in the f i rs t year af te r the enrolment , t ak ing advantage of a unique and large datase t f rom a cohort of I ta l i an s tudents in a l l undergraduate programs containing de ta i led inf ormat ion on s tudent and parenta l charac ter is t ic s . Re ly ing on a robus t s t rategy to c redibly sat is fy the condit iona l independence assumpt ion, we f ind that higher leve ls of consc ient iousness and ope nness to experience pos i t ive ly af fec t s tudent score .

Keywords: Noncognitive skills; Personality traits; Educational attainment; Economic Psychology

JEL Codes: I21, J24, D90

PsycINFO Classifications: 3100, 3550

Address for correspondence: Silvia D’Arrigo

Department of Economics and Business University of Sassari

Via Muroni 25

07100 Sassari (SS), Italy

Email: sdarrigo@uniss.it

This Working Paper is published under the auspices of the Department of Economics of the Ca’ Foscari University of Venice. Opinions expressed herein are those of the authors and not those of the Department. The Working Paper series is designed to divulge preliminary or incomplete work, circulated to favour discussion and comments. Citation of this paper should consider its provisional character.

1

The Influence of Personality Traits on University Performance:

Evidence from Italian Freshmen Students*

Luca Corazzini,† Silvia D’Arrigo,‡ Emanuele Millemaci,§ Pietro Navarra**

Abstract. Despite several attempts to provide a definite pattern regarding the effects of

personality traits on performance in higher education, the debate over the nature of the

relationship is far from being conclusive. The use of different subject pools and sample

sizes, as well as the use of identification strategies that either do not adequately account

for selection bias or are unable to establish causality between measures of academic

performance and noncognitive skills, are possible sources of heterogeneity. This paper

investigates the impact of the Big Five traits, as measured before the beginning of the

academic year, on the grade point average achieved in the first year after the enrolment,

taking advantage of a unique and large dataset from a cohort of Italian students in all

undergraduate programs containing detailed information on student and parental

characteristics. Relying on a robust strategy to credibly satisfy the conditional

independence assumption, we find that higher levels of conscientiousness and openness

to experience positively affect student score.

Keywords: Noncognitive skills; Personality traits; Educational attainment; Economic

Psychology.

JEL Classifications: I21, J24, D90.

PsycINFO Classifications: 3100, 3550.

*Acknowledgements. We are grateful to Marco Bertoni, Danilo Cavapozzi, Giuseppe De Luca, Francesco

Drago, Ruben Durante and Giorgio Monti for useful suggestions on the empirical analysis. We also thank

Stellario Antonio Curcuruto, Giuseppe Mannino and Rosaria Nisi at the IT Center of the University of

Messina for providing us access to the data and Dario Maimone Ansaldo Patti for assistance with the

data. All errors remain ours. † Department of Economics and VERA, University of Venice “Ca’ Foscari,” Cannaregio, 821, 30121

Venezia (VE), Italy. Email: luca.corazzini@unive.it ‡ Corresponding Author. Department of Economics and Business, University of Sassari, Via Muroni, 25,

07100 Sassari (SS), Italy. Email: sdarrigo@uniss.it § Department of Economics, University of Messina, Piazza Pugliatti, 1, 98122 Messina (ME), Italy.

Email: emillemaci@unime.it ** Department of Economics, University of Messina, Piazza Pugliatti, 1, 98122 Messina (ME), Italy.

Email: navarrap@unime.it

2

1. Introduction

In a seminal study, Borghans et al. (2008) propose a conceptualization of personality traits as

skills that make an independent contribution to the map of individuals' preferences, choices

and behaviours compared with the conventional measures of mental or cognitive skills

adopted by economists. While foremost economic literature has thoroughly examined the role

of cognitive functioning to learn, process and store knowledge in explaining socioeconomic

success, more recently, economists have started focusing on metrics of personality

endowments using the theoretical frames defined by personality psychologists. While

remaining relatively stable in adulthood, personality traits observe a steady process of

maturation in childhood and adolescence, during which public policy interventions are

impressive and have long-term consequences.1 Educational investments in noncognitive skills

that the labour market rewards can contribute to explain the connection between schooling

and returns to earnings (Bowles et al., 2001). Empirical evidence indeed documents that

personality traits affect the employment status and individual earnings (Caliendo, 2014;

Cobb-Clark & Tan, 2011; Fletcher, 2013; Nyhus & Pons, 2005), possibly through their

impact on productivity (Cubel et al., 2016; Salgado, 1997). In addition, while cognitive

abilities show a critical impact on achievement test scores and educational attainment,

personality traits produce distinct incentives to knowledge acquisition (Cunha & Heckman,

2008; Poropat, 2009). Findings in this branch of research are helping policy makers in better

designing educational public policies for improving student achievement and adult outcomes,

especially through early intervention strategies that aim at influencing the noncognitive skill

formation (J. J. Heckman et al., 2006; J. Heckman et al., 2013; Kosse et al., 2020).

Improving quality of learning and educational attainment to meet the increasing demand

for high-level skills in the labour market are challenging objectives of public policies. As

strategic elements of its political agenda for economic growth, in 2010, the European Union

(EU) set the targets of improving tertiary education attainment and reducing early school

leavers by 2020. Albeit the proportion of 25-34 years old individuals with a tertiary level of

education increased by 9 percentage points on average across OECD countries in the decade

2008-2018 (OECD, 2019), structural discrepancies in performance persist. In this respect,

Italy appears to be one of the most problematic countries. For instance, the percentage of

Italian population aged 30-34 who had successfully completed tertiary studies was only 27.6

against the EU average of 40.3 in 2019, while school dropout rates amounted to 13.5 per cent

(EU 10.2) of the population in the range 18-24 (Eurostat, 2019a, 2019b). Moreover, the

number of young people aged 20-34 who discontinued university studies, according to the

most recent data, was more than half a million, far above the EU average (Eurostat, 2016).

Since earlier investments in human capital beget skills at later stages of life in the dynamic

process of learning (Heckman, 2000), educational programmes fostering noncognitive skills

can provide school-age children with valuable resources to pursue a successful student career.

The pivotal nature of the connection between noncognitive skills and increases in education

can provide some guidance to reduce the divergent trajectories of national education systems.

1 See Almlund et al. (2011) and Kautz et al. (2014) for a detailed discussion about the possible mechanisms of

development and change of personality traits.

3

The present study focuses on the personality-related determinants of performance in post-

secondary education. The empirical analysis takes advantage of a unique dataset from a

cohort of 3,242 freshmen students aged between 18-24 years that in 2016 entered the degree

programmes of the University of Messina, a large and public Italian university. The interest of

this paper is in estimating the impact of self-reported scales of personality traits on the grade

point average (GPA), a measure of the exams successfully passed at the end of the first year

of courses. Personality assessment relates to the five-factor model, an internationally well-

established framework of personality traits, based on the following broad dimensions or Big

Five: extraversion, agreeableness, conscientiousness, emotional stability (also referred to as

the opposite pole of neuroticism) and openness to experience. At the time of matriculation,

prospective students are required to provide answers to the Italian adaptation of the Ten-Item

Personality Inventory (Chiorri et al., 2015), a popular personality questionnaire to infer

measures of the Big Five traits (Gosling et al., 2003).

The results indicate that higher scores on conscientiousness and openness to experience

have a positive impact on GPA. The estimates suggest that these Big Five traits affect

academic achievement besides the human capital background acquired through schooling.

Parental education, occupational status and industry are not responsible for the estimated

effect of personality traits on academic performance. We find significant gender differences in

performance, meaning that women outperform men in GPA. Albeit the density distributions

of the Big Five trait scores — except for openness to experience — significantly diverge

between men and women, we do not find gender-dependent effects of the Big Five on GPA.

Previous studies focusing on the predictive value of the Big Five traits on performance in

higher education report mixed results. The use of different subject pools and sample sizes

(e.g. Farsides & Woodfield, 2003) or the implementation of identification strategies that

either do not properly account for selection bias or are unable to establish causality between

measures of academic performance and noncognitive skills are possible explanations for such

inconclusive results. The present paper tackles these empirical issues as follows. First, results

are not subject to the selection bias due to limitations in the sample size, given that all

freshmen participated in the survey. Second, by the fact that data refer to university students

from all fields, results can be considered representative and applicable to the entire category.

Third, information on personality traits were collected before the beginning of the academic

year, a fact that allows us to establish the causal effect of students’ noncognitive skills on

their academic performance. Fourth, to the extent that personality traits can affect students’

preferences over degree programmes, we examine the independent distribution of personality

profiles across fields of study once controlling for an ample set of student and parental

characteristics. Finally, we use two recent empirical approaches (Bellows & Miguel, 2009;

Oster, 2019) to confirm that the estimated effects of personality traits on GPA are not driven

by selection on unobservables, therefore supporting the causal interpretation of our results.

All the above mentioned robustness checks suggest the validity of our empirical strategy and

our main conclusions.

The remainder of the paper is structured as follows. The next section reviews the existing

literature on the relationship between the Big Five personality traits and performance in

higher education. Sections 3 and 4 present data and empirical strategy, respectively. Section 5

4

discusses the results, while Section 6 presents the robustness checks. The last section

concludes.

2. Overview of the existing literature

Personality traits are the “relatively enduring patterns of thoughts, feelings, and behaviors”

(Roberts, 2009, p. 140) that, interactively with cognitive processes, motivational aspects and

beliefs, reflect adaptive responses to external conditions. Leading psychometric theories that

aim at conceptualizing latent personality traits in a standing and descriptive framework rely

upon a set of broad psychological constructs that hierarchically encompass narrower qualities

(or facets). The Big Five Factor structure has gained increasing popularity in the last decades

and is largely accepted as the most influential taxonomy of personality traits (Borghans et al.,

2008). Its structure rests on the lexical hypothesis developed from the semantic studies of

Allport and Odbert (1936) that the most important individual differences are encoded as

single terms in the natural language (Goldberg, 1990).

A number of studies in the literature focus on the relationship between the Big Five traits

and performance in higher education. Table 1 summarizes the main characteristics and

findings of relevant and influential studies that relate the Big Five traits to academic

examination grades.

Table 1. Previous findings on the effect of Big Five personality traits on post-secondary

examination grades.

Study Sample Method Controls E A C ES O

Burks et al.

(2015)

Undergraduate

students

(n=100), US

Tobit model Age, sex, family income,

time preference, risk

aversion, cognitive

abilities (non-verbal IQ,

numeracy, planning

ability)

0 0 + 0 0

Chamorro-

Premuzic

& Furnham

(2003a)

Undergraduate

students in

psychology

(n=247), GBR

Correlation analysis/

Multiple regression

None 0/- 0 + +/0 0

Chamorro-

Premuzic &

Furnham

(2003b) -

sample 1

Undergraduate

students in

psychology

(n=70), GBR

Correlation analysis/

Multiple regression

None 0 0 + + 0

Conard (2006) Undergraduate

students in

psychology

(n=289), US

Correlation analysis None 0 0 + 0 0

Undergraduate

students in

psychology

(n=186), US

Hierarchical

regression

analysis

Class attendance, SAT 0 0 + 0 0

5

Diseth (2003)

sample 1 Undergraduate

students in

psychology

(n= 127), NO

Correlation analysis None 0 0 0 0 0

sample 2 Undergraduate

students in

psychology

(n=101), NO

Correlation analysis None 0 - 0 0 +

Farsides &

Woodfield

(2003)

Undergraduate

students

(n=432), GBR

Correlation analysis/

Hierarchical

regression analysis

None/Cognitive abilities

(verbal IQ, spatial IQ),

seminar attendance, non-

assessed work

submission indicators

0 +/0 0 0 +

Furhnam et al.

(2003)

Undergraduate

students in

psychology

(n=93), GBR

Correlation analysis/

Hierarchical

regression analysis

None/Gender, general

cognitive ability (WPT),

beliefs about intelligence

(BAI)

- 0 + 0 0

Gray &

Watson

(2002)

Undergraduate

students

(n=334), US

Correlation analysis None 0 + + 0 +

Lounsbury et al.

(2003)

Undergraduate students in

psychology

(n=175), US

Correlation analysis None 0 0 + 0 +

Kappe &

van der Flier

(2012)

Undergraduate

students in

professional school of

human resource

management

(n=137), NL

Multiple regression

analysis

Intelligence, intrinsic

motivation, anxiety,

need for pressure, need for status, study

motivation

0 0 + 0 0

Komarraju

et al. (2009)

Undergraduate

students (n=308), US

Correlation analysis/

Multiple regression analysis

None/ Academic

motivation (AMS)

0 + + 0/- +

McCredie &

Kurtz (2020)

Undergraduate

students

(n=143),US

Raw

correlation/Partial

correlation

None 0 0 + 0 0

Noftle &

Robins (2007)

sample 1 Undergraduate

students in

psychology

(n=10,497), US

Correlation

analysis/Regression

analysis

None/Gender, SAT

verbal, SAT math

- +/0 + - +/0

sample 2 Undergraduate students

(n=475), US

Correlation analysis/Regression

analysis

None/Gender, SAT verbal, SAT math

0 0 + 0 +/0

Smidt (2015)

sample 1 College

students

(n=465), DE

Correlation

analysis/Multiple

linear regression analysis

None/Gender, age,

immigration

background, socio-economic status, school-

leaving GPA

0 0 + + 0

6

sample 2 University

students

(n=238), DE

Correlation

analysis/Multiple

linear regression

analysis

None/Gender, age,

immigration

background, socio-

economic status, school-

leaving GPA

0/- 0/- + 0 0/+

Vedel,

Thomsen

& Larsen

(2015)

University

students

(n=1,067), DK

Correlation

analysis/Multiple

regression analysis

None 0 +/0 + 0 +

Studies reported in Table 1 show that conscientiousness is the most meaningful dimension

of personality for academic performance. This evidence is in line with other results reported

in the literature on adult outcomes such as productivity, earnings, work attendance and health-

related behaviours (Bogg & Roberts, 2004; Cubel et al., 2016; Heineck & Anger, 2010;

Störmer & Fahr, 2013). Borghans et al. (2008) define conscientiousness “the degree to which

a person is willing to comply with conventional rules, norms, and standards”, where the

proactive side of the need for achievement and commitment to work combines with the

inhibitive side of moral scrupulousness and cautiousness (Costa et al., 1991). Burks et al.

(2015) observe that the proactive (but not the inhibitive) aspects of conscientiousness are

positive predictors of final college GPA. Psychological research suggests that achievement-

oriented and self-disciplined students with a persistent, hard-working aptitude despite any

sources of distraction obtain higher college grades (Costa et al., 1991; Gray & Watson, 2002).

Additional empirical evidence indicates that the association between conscientiousness and

students' grades may derive from the mediation role of having a stronger achievement

motivation, a dispositional propensity toward the academic effort, a positive self-perception

of the own academic ability and more productive study habits (Delaney et al., 2013; Noftle &

Robins, 2007; Richardson & Abraham, 2009).

The relationship between the Big Five personality traits other than conscientiousness and

grades appears more heterogeneous and smaller in magnitude, with openness to experience

the most sizeable factor (Almlund et al., 2011).

Openness to experience is “the degree to which a person needs intellectual stimulation,

change, and variety” (Borghans et al., 2008). Some studies have failed to identify significant

associations while others have provided supporting evidence that openness to experience

positively affects grades and GPA (Farsides & Woodfield, 2003; Gray & Watson, 2002;

Lounsbury et al., 2003). Since openness to experience relates positively to a deep interest in

learning and intellectual curiosity, students who score higher on this factor may more easily

study for a desire of knowledge, producing in this way positive effects on their academic

performance (Diseth, 2003; Von Stumm et al., 2011).

The extant literature has produced little evidence that extraversion, agreeableness and

emotional stability significantly correlate to grades. Extraversion describes the interpersonal

qualities of being outgoing, sociable and talkative, and a general propensity to experience

positive emotions (McCrae & John, 1992). Some studies have observed a negative

relationship between extraversion and performance at college (Noftle & Robins, 2007;

Chamorro-Premuzic & Furnham, 2003a; Furnham et al., 2003). A possible explanation for

7

these findings is that the need for social interactions of extroverted students negatively affects

their ability to accomplish hard tasks. On the other hand, Komarraju et al. (2009) find a

positive correlation between extraversion and extrinsic motivation among US undergraduate

students and they attribute it to the fact that, in pursuing academic goals, extroverted students

aim to satisfy their need for achievement more than being interested in the intrinsic rewards of

learning.

Meta-analyses of agreeableness and post-secondary performance indicate weak evidence

of a positive relationship (O’Connor & Paunonen, 2007; Richardson et al., 2012; Vedel,

2014). For instance, Farsides and Woodfield (2003) argue that lower rates of absences from

seminars (Delaney et al., 2013) may mediate the positive correlation between agreeableness

and final grades among US college students. Since, by definition, agreeable individuals have

an altruistic, compliant and tender-minded nature (McCrae & John, 1992), Komarraju et al.

(2009) attribute the positive performance of the more agreeable students to a higher degree of

motivation and the consequent propensity towards engagement and cooperative behaviour in

the classroom.

The empirical literature that informs the relationship between emotional stability and post-

secondary performance has produced contradictory results. In most cases, the evidence does

not show statistically significant correlations. While some studies report a positive association

with academic performance (Chamorro-Premuzic & Furnham, 2003a, 2003b; Lounsbury et

al., 2003), other studies present findings that move to the opposite direction (De Feyter et al.,

2012; Komarraju et al., 2009; Noftle & Robins, 2007), and meta-analytic reviews of the

psychological literature suggest that emotional stability is not a strong predictor of

performance (O’Connor & Paunonen, 2007; Vedel, 2014). Emotionally stable individuals are

less likely to experience feelings of anxiety, internal distress and self-consciousness than

neurotic individuals (McCrae & John, 1992). With respect to the academic performance, on

the one hand, emotionally stable students are expected to be more productive than neurotic

students because of their better ability to manage academic stress (Kaiser & Ozer, 1997). On

the other hand, emotionally stable students may make less efforts to prepare their exams

because are conscious to be able to adapt more readily to adverse events or, equivalently, face

a lower expected cost from a bad outcome. In support of this explanation, Delaney et al.

(2013) observe that students who score higher on neuroticism are those more likely to do

extra hours of study.

In general, studies in the literature contain limitations that may bias results. First, the

adopted datasets are often of limited size and not representative of the global student

population. Second, some studies consider datasets aggregating students at different stages of

their university career (freshmen, sophomores or more advanced), without taking explicitly

into account differences in cumulated academic experience that, possibly interacting with

personality traits, influence skill accumulation process. Third, using contemporaneous

measures of personality traits and educational attainment as well as not controlling for

potentially important factors such as student schooling information, parental education and

employment sector arise endogeneity issues, which in turn limit the validity and

interpretability of the statistical results. In the next pages, we accurately describe the empirical

strategy that we use to solve the above mentioned empirical issues and provide more robust

evidence on the relationship between personality traits and GPA.

8

3. Data

We use data from a cohort of freshmen at the University of Messina, a large public university

located in southern Italy, which enrolled over 23,000 students in the academic year 2016-

2017. As part of the European Higher Education Area (in application of the so-called

“Bologna Process”2), the Italian university system currently distinguishes between the first-

level degree (Laurea), a three-year study programme corresponding to a Bachelor’s degree in

the European university system, and a second-level of academic qualification that is obtained

after two years of postgraduate studies and requires an undergraduate academic degree for the

admission to the programmes (called Laurea Magistrale, corresponding to a Master’s degree).

In addition, a limited number of study programmes for the access to regulated professions, i.e.

Architecture, Dentistry, Human Medicine, Law, Pharmacy and Veterinary, is structured in a

unique cycle lasting either five or six years (Laurea magistrale a ciclo unico).

We conduct our research on undergraduate students of either first-cycle or single-cycle

degree courses aged between 18-24 years at the final deadline for matriculation by using the

90th percentile of the age distribution as reference threshold to fix the range. There are several

reasons to examine the academic performance of freshmen students. First of all, the first year

of study is a critical step in the educational pathway as suggested by the fact that the majority

of university dropouts occur during this time.3 Moreover, the ability with which a student

copes with a different and more independent learning system in the first year enhances the

ability to manage the workload in the subsequent years and, therefore, the likelihood that the

student will persist in education. Finally, the first-cycle and single-cycle degree programmes

usually provide more common and basic courses in the first year and more differentiated and

specialized courses in the subsequent years.

Data sources are a web-based survey and administrative data from the University

Information System Office. In the academic year 2016-17, the participation in the web-based

survey constituted an obligatory step in the enrolment procedure when the student applied to

the university. While administrative data provides us with a wide range of student’s personal

and schooling information, the questionnaire allows us to collect detailed information on

parental characteristics – educational attainment, occupational status and industry – aside

from on the student’s personality profile.

The personality questionnaire administered to the students is the Ten-Item Personality

Inventory (TIPI) with the adaptation for Italian subjects (Chiorri et al., 2015) of the

commonly adopted questionnaire of Gosling et al. (2003). The TIPI is designed to optimize

content validity — with two extensive descriptors per item that capture the spectrum of the

personality constructs — and is a short-form questionnaire to obtain measures of the broad

2 https://ec.europa.eu/education/policies/higher-education/bologna-process-and-european-higher-education-

area_en 3 The percentage of freshmen students of the cohort 2016-2017 in Italy who abandoned university after the first

year was around 12 per cent for first-cycle degree courses and 7.5 per cent for single-cycle degree courses

(Anvur, 2018).

9

domains of personality in the context of ample sample surveys4. More precisely, the Big Five

personality traits are inferred from a list of two items per scale that proxy for the positive and

the negative poles of the corresponding latent factor. Respondents rate to what extent each

statement describes them on a seven-point Likert-type scale. We obtain the measures of the

Big Five traits by first recoding the reverse-scored items and then averaging the scores of the

two items for each scale.

We consider only students for whom we can observe exam performance, that is, students

who have passed at least one of the scheduled classes for the academic year. This strategy

limits the following two sources of measurement error. First, since in the Italian university

system grades are assigned on a 30-points scale and failing grades – below the minimum

score of 18 – are not registered, we cannot detect a dimension of student achievement in the

absence of academic results. Second, and related to the first point, we cannot distinguish

between those students who fail the exam and those who never try.

The outcome variable in the empirical analysis is a weighted measure of the grade point

average or GPA of the exams the student passed in his first year of study. In particular, the

GPA is calculated as the average of the marks – ranging from 18 to 30 – weighted on the

credit point values assigned for each completed course. Since in the European Credit Transfer

System one academic credit is conventionally equivalent to 25 hours of work, the GPA

incorporates the expected individual effort that is associated with the knowledge requirements

for passing courses. The GPA is more a measure of the quality of learning rather than of the

overall performance, given that students who are more effective in credit accumulation but

receive lower marks obtain inferior scores than better-performing students with a lesser

amount of academic credits. Notice that we cannot include this information in the set of

independent variables because, otherwise, depending the total number of academic credits

earned on personality traits, we would not be able to identify treatment effects on GPA any

more (Angrist & Pischke, 2008). However, we do not find a trade-off between quality (GPA)

and quantity (credits) for these students as the Spearman's rank correlation coefficient

between GPA and total amount of credits earned is positive and statistically significant

(rho=.27, p<.01). To account for possible sources of heterogeneity in the characteristics of

the study programmes, such as class scheduling, teaching methods and performance

evaluation, affecting the distribution of student scores and hence their comparability, we

standardize the GPA within each course of study5.

4

Gosling et al. (2003) used existing Big-Five instruments such as the well-established, 44-item Big Five Inventory (BFI, John et al., 2008) to select labels for a frame of ten items. The authors report remarkable

convergent correlations between the TIPI and the BFI, far exceeding discriminant correlations. Positive findings

for these tests derive from analogous comparisons with the 240-item NEO Personality Inventory, Revised (NEO-

PI-R, Costa & McCrae, 1992). The TIPI showed also high test-retest reliability and good performance results in

terms of external correlations compared to the BFI scales. Similarly, Chiorri et al. (2015) document that their

Italian version of the TIPI has an acceptable factor structure and performs well for convergent and discriminant

correlations with the BFI, external correlations and test-retest correlations. 5 To address further distributional issues, we also consider an alternative measure of university performance

obtained from the percentile ranks of the GPA scores by course of study. Since the estimated effects are not

qualitatively different from the analyses using the standardized values of GPA, we employ for simplicity this

latter measure in the empirical analysis. The results are available from the authors upon request.

10

Excluding 54 students for whom information about school background is not available or

incomplete, the dataset consists of 3,242 units of observation. Table 2 reports the descriptive

statistics referring to the main variables considered in the econometric analysis.

Table 2. Descriptive statistics.

(1) (2) (3) (4) (5)

Variables Number of

observations

Mean Standard

deviation

Min. Max.

Grade point average (GPA) 3,242 25.070 2.614 18 30

Extraversion 3,242 4.074 1.354 1 7

Agreeableness 3,242 5.406 1.055 1 7

Conscientiousness 3,242 5.534 1.116 1 7

Emotional stability 3,242 4.633 1.252 1 7

Openness to experience 3,242 4.766 .940 1 7

Female 3,242 .590 .492 0 1

Age at 30 Dec 2016 3,242 19.975 1.171 17.966 24

Type of upper secondary school

Liceo for classical studies 581 .179 .384 0 1

Liceo for scientific studies 1,290 .398 .490 0 1

Liceo for other studies 594 .183 .387 0 1 Technical/vocational school 777 .240 .427 0 1

Educational attainment

Father

up to lower secondary school degree 1,099 .339 .473 0 1

upper secondary school degree 1,463 .451 .498 0 1

graduate in matched field 174 .054 .225 0 1

graduate not in matched field 506 .156 .363 0 1

Mother

up to lower secondary school degree 969 .299 .458 0 1

upper secondary school degree 1,559 .481 .500 0 1

graduate in matched field 144 .044 .206 0 1

graduate not in matched field 570 .176 .381 0 1

Occupation

Father

unemployed, in education 306 .094 .292 0 1

employee 1,925 .594 .491 0 1

self employed 787 .243 .429 0 1

other 45 .014 .117 0 1

retired 179 .055 .228 0 1

Mother

unemployed, in education 959 .296 .456 0 1 employee 1,695 .523 .500 0 1

self employed 273 .084 .278 0 1

other 280 .086 .281 0 1

retired 35 .011 .103 0 1

Industry

Father

business/personal services,

public administration

1,058 .326 .469 0 1

professional, scientific, technical activities 383 .118 .323 0 1

manufacturing, construction 209 .065 .246 0 1

other 1,592 .491 .500 0 1

Mother

business/personal services, 772 .238 .426 0 1

11

public administration

professional, scientific, technical activities 131 .040 .197 0 1

manufacturing, construction 35 .011 .103 0 1

other 2,304 .711 .454 0 1

Note. The grade point average (GPA) variable is measured as the average of the grades obtained in the first year of the

study programme weighted by the credits associated with those grades. The variable “graduate in matched field” is a

dummy variable on whether father/mother graduated in a field of study analogous to the child.

Table 3 presents descriptive statistics of the Big Five trait measures and the results of a set

of Mann-Whitney tests of equality in distribution, broken down by gender. The Big Five

density distribution by gender is also graphically presented in Figure 1. The null hypothesis is

rejected for all the domains with the exception of openness to experience (p-value= 0.682).

With respect to men, women score higher on agreeableness and conscientiousness but lower

on extraversion and emotional stability. These results overlap with a large literature that

studies the correlation between personality traits and gender, which especially posits that

women are, on average, more agreeable and less emotionally stable than men (Chapman et al.,

2007; Costa et al., 2001; McCrae et al., 2005). Consistent with the findings from

psychological literature that women are more inclined to experience anxiety and internal

distress than men since the early stages of adolescence (De Bolle et al., 2015), the strongest

gender differences emerge for emotional stability, where female students score on average

0.73 points less than male students.

Table 3. Descriptive statistics of the Big Five personality traits by gender.

Note. The last column reports p-values from a Wilcoxon rank-sum test of the null hypothesis that the two

independent samples are from populations with equal distribution.

Men

Women

(N=1,330) (N=1,912)

Variable Mean Standard

deviation

Min. Max. Mean Standard

deviation

Min. Max. Difference

(p-value)

Extraversion 4.202 1.323 1 7 3.985 1.369 1 7 0.000

Agreeableness 5.298 1.037 1 7 5.480 1.060 1.5 7 0.000

Conscientiousness 5.384 1.119 1 7 5.639 1.103 1 7 0.000

Emotional stability 5.061 1.170 1 7 4.334 1.222 1 7 0.000

Openness to experience 4.775 0.939 2 7 4.759 0.941 1 7 0.682

12

Figure 1. Density distribution of the Big Five trait scores by gender.

4. Empirical strategy

Consider the following equation describing the main determinants of student performance,

measured with the grade point average (GPA), for student i=1, 2, ..., n at time t+1 (first year

of course):

𝐺𝑃𝐴𝑖 (𝑡+1)

= 𝛼 + ∑ 𝛽𝑗𝑓𝑎𝑐𝑡𝑜𝑟𝑖,𝑗(𝑡)5𝑗=1 + 𝛾 𝑋𝑖 + 𝛿 𝑍𝑖 + 휀𝑖, (1)

where 𝑓𝑎𝑐𝑡𝑜𝑟 represents the Big Five personality trait j=1,2,…,5 at time t (matriculation

time); 𝑋 is a vector of predetermined characteristics, 𝑍 is a vector of additional controls and 휀

is an unobserved error term. The coefficient of interest, 𝛽𝑗, estimates the average effect of a

one-point increase in the personality trait j under the conditional independent assumption that

the personality type is as good as randomly distributed across students once conditioning on

the vector of observed characteristics, 𝑋.

The personality trait j is a function of the vector of observables, 𝑋, and unobservable

factors. As long as the latter is uncorrelated with 휀- the error term included in equation (1) -

or, equivalently, with the unexplained variance of GPA, 𝛽𝑗 is an unbiased estimator of the

effect of the Big Five on GPA.

0.1

.2.3

den

sity

1 2 3 4 5 6 7factor scores

men women

Extraversion

0.1

.2.3

den

sity

1 2 3 4 5 6 7factor scores

men women

Agreeableness

0.1

.2.3

den

sity

1 2 3 4 5 6 7factor scores

men women

Conscientiousness0

.1.2

.3

den

sity

1 2 3 4 5 6 7factor scores

men women

Emotional stability

0.1

.2.3

den

sity

1 2 3 4 5 6 7factor scores

men women

Openness to experience

13

The set of the predetermined characteristics (𝑋) includes a gender indicator, age, school

background and parents’ information. Table 2 provides a list of the variables considered in the

empirical analysis to check the conditional independence assumption.

The school background variables consist of indicators for categories of upper secondary

school. The Italian upper secondary school education typically lasts five years (from age 14 to

19) and concludes with the State Examination - a set of national exams whose subjects

depend on the type of school attended. The school leaving qualification or diploma di

maturità enables the student to access higher education and provides a final score reflecting

the general background knowledge acquired through schooling years. Since noncognitive

skills play a role in the human capital accumulation process and affect the scholastic success

(Borghans et al., 2008; Borghans et al., 2011; Di Fabio & Busoni, 2007), the educational

achievement captured by the measure of diploma score also depends on personality factors.

Accordingly, including this variable in the model would not allow to correctly identify the

effects of the Big Five traits on university performance (Angrist & Pischke, 2008). The

indicators for the type of upper secondary school allow us to distinguish between the licei that

provide more theoretical and university-oriented skills, from technical and vocational schools

that provide professional training for direct access to the job market.6 These indicators are

particularly useful to gather information on human capital endowment in the context of the

Italian education system where required effort and study subjects may vary markedly across

school programmes.

The parental variables are indicators of educational level, occupational status and industry.

As measures of the socioeconomic status of the student at the time of matriculation, the

parental variables are potential vectors of academic performance (Delaney et al., 2011;

Rodríguez-Hernández et al., 2020). Moreover, the relationship between personality traits and

student scores may be influenced by parental characteristics as educational level and

occupational industry. For instance, personality traits may help predicting student’s autonomy

in the choice of the field of study with respect to parental pressures. We check for this

hypothesis by including in some specifications a dummy on whether the parent holds a

university degree or equivalent qualification related to the field of study of the child.

The vector of controls, 𝑍 , includes firstly the student’s field of study. Following the

classification used by the European Research Council (ERC) to organize disciplines in

coherent research domains, we group students in three fields of study or ERC sectors: Social

Sciences and Humanities (N=1,450), Physical Sciences and Engineering (N=276) and Life

Sciences (N=1,516). Moreover, vector 𝑍 accounts for individual-specific

municipality/province of residence fixed effects that may help explaining heterogeneity in the

outcome variable for the following reasons. First, time for travelling to and from university

may differ significantly across students of different areas and affect their performance.

Second, the short-term opportunity-cost of studying versus leisure time may vary significantly

6 Notice that, regardless the type of school attended, each student can freely choose one of the courses of study

that are provided by Italian universities.

14

across different (although close) areas. Third, different expectations on the job opportunities

in the local labour market may be another source of variation in academic performance.7

Since students of the same course of study are exposed to common unobservable

components that influence their university track and involving, for instance, interactions with

the peers, class size, course content and teaching style, we account for within-student

correlation by adjusting standard errors for clustering.

An important issue is the influence on estimates of measurement error possibly contained

in self-reported factor scores. For instance, even unconsciously respondents may distort self-

reports of personality traits to respond in a socially desirable manner (Almlund et al., 2011).

However, several facts reassure us on this respect. First, the applicants who participated in the

web-survey were aware that the data provided by filling in the personality inventory served

research purposes only. Second, if the measurement error is classical (random), the only effect

on estimates is on standard errors that, becoming larger, determine a reduction in the

statistical significance of estimated coefficients. Third, the literature suggests how induced

faking behaviour in a hypothetical scenario at the time of university admission does not affect

the criterion validity of personality domain-scores relative to examination performance

(Ziegler et al., 2010) and the impact of social desirability on the relationship between the Big

Five personality traits and cumulative college GPA is minimal (Peterson et al., 2006).

Another issue that deserves attention is the stability of personality traits from the time of

the survey participation to the exams time, occurring months later during the first year of

course, in one or multiple different dates. In particular, if personality traits are endogenous

with respect to the outcome of the exams, results will be biased. However, as the literature

suggests that dramatic changes in personality are unlikely to occur in young adults, it seems

unlikely that the outcome of an exam is able to determine a significant change in personality

traits at this stage of life and, especially, over a short-length period of one year. Individuals

experience a gradual maturation in their personality traits after the developmental stages of

childhood but reach a substantially stable path at the stage of life roughly corresponding to the

university matriculation (Delaney et al., 2013).8 Absolute changes in the Big Five traits are

small to moderate in magnitude, with the exception of conscientiousness, and do not appear to

be statistically reliable for the vast majority of individuals during the transitional life-path

between ages 15-24; the effect sizes are also negligible among working-age adults over a time

length of four years (Cobb-Clark & Schurer, 2012). In addition, rank-order consistency –

ordinal positioning of the trait scores within a group – increases as a function of age (Caspi et

al., 2005; Pullmann et al., 2006; Roberts & DelVecchio, 2000).

7 The groupings of residence include: city of Messina, province of Messina (except the city of Messina), any

other place of residence in the Sicilian region, city of Reggio Calabria, province of Reggio Calabria (except the

city of Reggio Calabria), any other place of residence in the Calabria region and area of provenance other than

the ones listed above, with the excluded category being the city of Messina. 8 Mean-level changes occur at lower degrees later in life (McCrae et al., 2002; Pullmann et al., 2006; Roberts et

al., 2006; Srivastava et al., 2003).

15

5. Results

Table 4 reports the estimation results of the effects of the Big Five personality traits on GPA.

To facilitate the interpretation of the results, we standardize the personality trait scores and the

continuous independent variables to have zero mean and standard deviation of one. The

stability of the estimated parameters on the Big Five personality traits is evaluated

considering a number of specifications with different sets of controls. Column 1 of Table 4

reports the estimated effects from a baseline regression without controls. The subsequent

columns present the results from the estimation of specifications in which we cumulatively

add the following controls: (2) demographics, including gender, age and age squared, school

background, including indicators for type of upper secondary school and indicators for ERC

sectors; (3) parental controls, including measures of educational attainment, occupational

status and industry; (4) indicators for student’s municipality or province of origin.

The findings suggest that the coefficients on the Big Five factor scores are substantially

robust to the inclusion of the control variables. Conscientiousness and openness to experience

exhibit positive and statistically significant coefficients, while the other Big Five traits do not

appear to have a significant impact on GPA. Conscientiousness has the more intensive effect

on GPA, which is consistent with the literature emphasizing the role played by motivational

aspects and self-discipline on student achievement. Specifically, one standard deviation

increase in conscientiousness is associated with an increase of 9.3 percent of a standard

deviation in GPA. One standard deviation increase in openness to experience raises 3.8

percent of a standard deviation GPA. This result is in line with the view that the positive

disposition toward intellectual engagement and novel ideas of the open to experience

individuals determines a more intense interest in knowledge acquisition and, predictably,

more positive achievement scores (Diseth, 2003; Von Stumm et al., 2011).

The choice between liceo and technical or vocational school is associated with path

differences in university performance. As predictable, students who received a more

theoretical secondary education appear outperforming those with a technical or vocational

background. In addition, students who attended other categories of licei that are more focused

on visual arts, architecture and design, foreign languages, music or human sciences with

respect to humanities or scientific disciplines perform less well than students who attended

the liceo for classical studies.

According to a literature highlighting the existence of gender differences in academic

performance (Almås et al., 2016; Conger & Long, 2010; Pomerantz et al., 2002) the results

suggest that women obtain higher GPA scores than men. As discussed in Section 3, the

density distribution of the Big Five trait scores significantly differs between men and women,

the only exception being openness to experience. Is then the effect of personality traits on

university performance gender-dependent? We examine this question by additionally

including interaction terms between the Big Five traits and the gender dummy in the

specifications listed in Table 4. The results - available upon request - suggest that the effects

of personality traits on GPA are not heterogeneous between men and women.

16

Table 4. Estimated effects of the Big Five personality traits on GPA.

(1) (2) (3) (4)

Extraversion -0.025 -0.025 -0.024 -0.022

(0.019) (0.018) (0.019) (0.018)

Agreeableness -0.037 -0.039 -0.041* -0.037

(0.023) (0.023) (0.023) (0.024)

Conscientiousness 0.110*** 0.094*** 0.093*** 0.092***

(0.019) (0.017) (0.017) (0.017)

Emotional stability -0.057** -0.035 -0.031 -0.030

(0.027) (0.024) (0.023) (0.023)

Openness to experience 0.036** 0.039** 0.038** 0.038**

(0.018) (0.018) (0.017) (0.017)

Female 0.140*** 0.138*** 0.148***

(0.044) (0.046) (0.043)

Age -1.049** -1.114** -0.978* (0.494) (0.493) (0.500)

Age squared 1.021** 1.089** 0.957*

(0.499) (0.497) (0.505)

Type of upper secondary school

Liceo for scientific studies -0.094 -0.101* -0.103*

(0.056) (0.055) (0.053)

Liceo for other studies -0.168*** -0.173*** -0.164*** (0.039) (0.036) (0.039)

Technical/vocational school -0.347*** -0.345*** -0.353***

(0.064) (0.059) (0.057)

ERC sectors

Physical Sciences and Engineering 0.093* 0.077 0.060

(0.047) (0.049) (0.049)

Life Sciences 0.055 0.055 0.065 (0.050) (0.054) (0.054)

Parental controls YES YES

Municipality/ province of provenience YES

Constant -0.000 0.035 0.198* 0.232**

(0.005) (0.056) (0.101) (0.098)

Observations 3242 3242 3242 3242

F 7.251 14.940 81.575 105.332 p-value 0.000 0.000 0.000 0.000

Note. Robust standard errors clustered by course of study indicators are in parentheses. The

omitted category of upper secondary school is the liceo for classical studies. The omitted category of ERC sector is Social Sciences and Humanities. Parental controls include educational attainment,

occupational status and industry. * p < 0.10, ** p < 0.05, *** p < 0.01

6. Robustness checks

In addition to the observed controls included in the analysis presented in Section 5, further

information on the socioeconomic conditions of the household, as well as the number of

members, is available for a subsample (72 per cent) of students who submitted the Equivalent

Economic Situation Indicator (ISEE) for university, which is a certificate that gives access to

17

scholarship programmes, reductions in the tuition fees and other student benefits.9 Thus, to

verify the stability of the results, we test the following alternative specifications: (i) we

include the ISEE variable (measured as the standardized logarithms of the ISEE values) and

the number of household members as regressors, but excluding the parental controls; (ii) we

include the ISEE variable, the number of household members and the parental controls

simultaneously. The estimated effects – available upon request – remain qualitatively

unchanged with respect to those presented in Table 4.

Since the student’s choice of the course of study to attend reflects individual aspirations,

aptitudes and preferences relating to personality traits, a non-random distribution of

personality profiles across degree programmes would not be surprising. To check whether the

Big Five traits are conditional-balanced across fields of study, we run a series of regressions -

with standard errors clustered at the course of study level - of each trait on the ERC sectors,

including student and parental characteristics and dummies for the area of residence as

controls. The only sources of unbalance are: (i) students in Physical Sciences and Engineering

are less extraverted than students in the other two academic disciplines (p<.05 with respect to

students in Social Sciences and Humanities; p<.01 with respect to students in Life Sciences);

(ii) students in Social Sciences and Humanities are less emotionally stable than students in

Life Sciences (p<.05). To address this issue, we estimate an equation including, apart from

the Big Five trait scores, the full set of individual and parental controls, the ERC sectors and

the interaction terms between the Big Five trait not uniformly distributed across fields of

study and the corresponding ERC sector. Specifically, the interaction terms that we consider

are the product of (i) extraversion by Physical Sciences and Engineering, (ii) emotional

stability by Social Sciences and Humanities and (iii) emotional stability by Life Sciences. As

neither the effect of extraversion on GPA for students in Physical Sciences and Engineering

(or other students) nor the interaction effect between the variables are statistically significant,

we conclude that this group does not drive the estimates in main analysis of Section 5. On the

other hand, as emotional stability is negatively and significantly related to student

performance in the Life Science sector (p<.01), we note that for this trait the effect is group-

specific. Contrary to the intuition that emotional stability supports better performance

outcomes, this evidence suggests that neurotic students in Life Sciences are more averse to a

bad exam experience and for this reason are prone to work harder to make this an unlikely

event. The results of this robustness analysis are available from the authors upon request.

As explained in the sections above, the estimation strategy of this study relies on the

fulfilment of the conditional independence assumption and, therefore, on the limitation of the

bias due to omitted confounders. In this view, one evaluates the stability of the coefficients

after the inclusion of controls in the equation. In theory, this procedure may not be able to

warrant consistent estimates against the omitted variables problem and this risk is higher

9 Tuition fees paid by the students at Italian public universities in each academic year depend on the ISEE

certificate for university use based on information about income, asset situation and composition of the

household. Some students with high socioeconomic status (ISEE > Eur 60,000) did not provide the ISEE

certificate since they were required to pay the maximum amount of tuition fees in any case. Other students did

not request their ISEE because were exempted from paying tuition fees (e.g., students with maximum school

leaving qualification or students eligible for disability benefits).

18

when the observed variables explain only a small fraction of the variance of the outcome

variable. To discard further concerns about self-selection of students and to account for the

influence of other potential unobservable characteristics, we implement the methodology

introduced by Altonji et al. (2005) to estimate the ratio of selection on unobservables relative

to selection on observables that would be required for the treatment effect to be entirely due to

the omitted variable bias. The rationale of this test is that the sensitivity of the estimated

coefficients to the inclusion of controls is not a sufficient diagnostic of the omitted variable

bias. Coefficient changes are proportional to the omitted variable bias only if they are scaled

by changes in the fraction of explained variance when the control variables are included

(Oster, 2019).

This paper presents two tests for selection on unobservables. The first test, which follows

the empirical approach introduced by Bellows and Miguel (2009)10 and its application in

Nunn and Wantchekon (2011), presents an estimator of the coefficient of proportionality

under the null hypothesis that the treatment effect is not significantly different from zero. In

turn, each personality trait represents the treatment variable. We distinguish between �̂�R, the

estimated coefficient from a restricted regression equation which controls only for the 5-j

factor scores (specification I of Table 4), and the estimated coefficient �̂�F from a regression

that includes the full set of control variables (specification IV of Table 4). Hence, the

coefficient of proportionality 𝛿 is calculated as follows: �̂�F / (�̂�R - �̂�F).11 The greater is the

magnitude of �̂� F, the larger is the effect that needs to be explained by selection on

unobservables. The smaller are the coefficient movements in the denominator, the lower is the

selection on observables and the greater should be the selection on unobservables to explain

away the observed effect.

The second test follows the approach recently employed by Oster (2019), i.e. to set a

value for R-squared in the case that the full set of controls (observable and unobservable

variables) were included in the model, Rmax, and then to calculate the value of the coefficient

of proportionality 𝛿 assuming the treatment effect equal to zero. Oster posits that the approach

of Altonji et al. (2005) is restrictive in assuming that, if the unobservable factors were

included in the regression model, the R-squared coefficient would be equal to 1, since it does

not take into account possible sources of exogenous variation in the outcome such as

unobserved factors which are not predetermined characteristics or measurement error. Thus,

the bounding value of Rmax is set to 1.3 times the R-squared coefficient from a regression

which includes the complete set of observed controls, �̃�. Oster calculates the multiplier as the

cut-off value that would allow at least 90 per cent of the randomized controlled trials

published in the five top journals between 2008 and 2013 to survive the selection adjustment

criteria (for nonrandomized data the survive rate drops to 45 per cent). Again, each

personality trait represents the treatment variable. The results are summarized in Table 5. The

10 Bellows and Miguel (2009) build up an estimator of the coefficient of proportionality in the case of a linear

regression model with a continuous treatment variable. Their approach diverges from the methodology of Altonji

et al. (2005) in that they implicitly assume that observed and unobserved controls are equal in variance

(Gonzáles & Miguel, 2015). 11

A formal description of the ratio is provided in the Appendix of Bellows and Miguel (2009). Oster (2019)

formalises the assumptions of the model and develops a consistent estimator of the omitted variable bias.

19

Table 5. Amount of selection on unobservables relative to selection on observables for β=0.

δ ratio for β=0

Treatment variable Bellows and Miguel

(2009)

Oster

(2019)

Extraversion 7.543 16.777

Agreeableness -421.058 95.467

Conscientiousness 4.908 7.568

Emotional stability 1.132 2.185

Openness to experience -17.106 -51.000

Note. The table shows the amount of selection on unobservables relative to selection on

observables to explain away the entire treatment effect. In column 1, we use the Bellows and

Miguel method, which satisfies the condition that observed and unobserved controls equally explain the variance of the outcome. In column 2, we use the Oster formula and set the upper

bound on 𝑅𝑚𝑎𝑥 equal to 1.3 times the R-squared from a regression equation that includes the

entire set of observed controls.

𝛿 values reported in column 1 are computed using the Bellows and Miguel formula, while in

column 2 follow the Oster procedure. All the ratios are greater than one in absolute value and

generally robust to validation analysis. The coefficient of proportionality for emotional

stability presents a smaller effect of 1.13 in the method proposed by Bellows and Miguel

compared to the value obtained using the Oster formula, signalling possible confounding

effects due to omitted bias under their selection assumptions. The estimated effects of

conscientiousness and openness to experience appear robust to the selection adjustment.

Notice that the delta values for openness to experience are negative in sign. To the extent that

selection on observables and selection on unobservables are positively correlated, a negative

ratio means that, if anything, the estimated effect is downward biased by selection on

unobservables. If we double the value of the multiplier in the Oster formula, still selection on

unobservables should be 1.85 times more important than selection on observables to capture

the entire effect on performance (the delta ratio is -10.16 for openness to experience).

7. Concluding comments and policy implications

This paper provides insights that the Big Five dimensions of personality play a role in

explaining the documented heterogeneity in student achievement and adds a piece of evidence

to a strand of literature that attempts to trace the fundamentals of the relationship between

personality traits and post-secondary performance. To address this research issue, we have

taken advantage of an ample set of information about a cohort of freshmen students within the

degree programs of a large public university in Italy. Additional specification checks verify

that the estimated effects of personality traits are robust to endogenous factors informing the

selection rule of the degree programmes. Our findings suggest that conscientiousness and

openness to experience are positive personality traits in determining the grade point average

of the exams that students passed throughout their first university year. In particular, the

observational estimates reinforce the explanation that the aptitudes for dutifulness and self-

20

discipline help conscientious individuals to sustain the enduring commitment to study

required to achieve better performance outcomes. In addition, openness to experience appears

a strategic trait for knowledge acquisition, since the aspect of intellectual engagement and the

ability to develop an independent approach to study are critical qualities to build up

professional expertise that is valuable in the labour market.

To investigate to what extent the estimated results can be interpreted as evidence of a

causal relationship between the Big Five personality traits and academic performance, we

have implemented the empirical techniques proposed by Bellows and Miguel (2009) and

Oster (2019). We have found that, while for conscientiousness selection on unobservables

should be strictly greater than the unit relative to selection on the observed variables to

attribute the entire treatment effect to selection bias, in the case of openness to experience the

estimated effect is likely to be downward biased by selection on unobservables.

Although policymakers devote increasing attention to inefficiencies in the education

system, what interventions are effective in reducing university dropout rates is an open

empirical question. Advances in the comprehension of the relationship between noncognitive

skills and student achievement can indicate a variety of strategies to improve the

competitiveness of education systems, especially for those countries that, as Italy, show

higher indices of university dropouts among developed economies. The results of this

research support, in the long term, public investment in intervention programs aimed at

enhancing the Big Five personality traits that are positively associated with productivity in

adulthood. Since noncognitive skills are more malleable later in life than cognitive abilities,

efforts in this direction focus on educational strategies designed to foster learning skills and

intellectual engagement by shaping students' noncognitive skills between childhood and

adolescence (Heckman et al., 2019). A clear understanding of the mechanisms that drive

noncognitive skill formation in schooling would sustain later investments in higher education.

Several studies have investigated the impact of intervention programmes that targeted

disadvantaged children or low-performing students, showing that adequate mentoring and

training practices, either in the classroom or with extracurricular and social activities, can

produce persistent improvements in intellectual engagement and behavioural changes

(Heckman et al., 2019). These programmes aimed to stimulate emotional awareness and

prosocial behaviour (Bierman et al., 2010; Kosse et al., 2020), to encourage internal

motivation, perseverance and individual effort in achievement goals (Alan et al., 2019) and to

provide problem-solving techniques and study methods that foster self-discipline (Martins,

2017).

Furthermore, the results of our research suggest that university institutions may more

effectively emphasize the resources of talented students who are highly motivated and

hardworking (conscientiousness), intellectually curious and creative (openness to experience)

by offering them degree programmes that embrace different academic disciplines. In this

spirit, top universities such as Harvard University, Massachusetts Institute of Technology and

University of Oxford have introduced multidisciplinary courses that integrate Humanities

with Science and Engineering.

In addition, university institutions may consider that differences in individuals’

personality traits influence their approach to learning and, consequently, they may seek to

support students through organizational changes and targeted services fostering better

21

productivity levels. For instance, universities may consider introducing compulsory

attendance rules and the opportunity for students to take multiple exams, each one in partial

fulfilment of the knowledge requirements for the completion of the course, instead of taking a

single overall exam. On the one hand, these settings may serve as commitment strategies for

students who are less diligent and self-disciplined to prevent inefficient procrastination. On

the other hand, students who lack intellectual curiosity may benefit from engaging in a series

of short-term deadlines to maintain the study effort. Understanding which strategies can have

a positive impact on the performance of undergraduate students with a personality profile that

is less rewarding in the university system is an interesting pathway for future research.

22

References

Alan, S., Boneva, T., & Ertac, S. (2019). Ever Failed, Try Again, Succeed Better: Results

from a Randomized Educational Intervention on Grit. The Quarterly Journal of

Economics, 134(3), 1121–1162. https://doi.org/10.1093/qje/qjz006

Allport, G. W., & Odbert, H. S. (1936). Trait-names: A psycho-lexical study. Psychological

Monographs, 47(171). https://doi.org/10.1037/h0093360

Almås, I., Cappelen, A. W., Salvanes, K. G., Sørensen, E., & Tungodden, B. (2016). What

explains the gender gap in college track dropout? Experimental and administrative

evidence. American Economic Review, 106(5), 296–302.

https://doi.org/10.1257/aer.p20161075

Almlund, M., Duckworth, A. L., Heckman, J., & Kautz, T. (2011). Personality Psychology

and Economics. In Handbook of the Economics of Education (Vol. 4, Issue 2008).

Elsevier B.V. https://doi.org/10.1016/B978-0-444-53444-6.00001-8

Anvur (2018). Rapporto biennale sullo stato del sistema universitario e della ricerca.

Altonji, J. G., Elder, T. E., & Taber, C. R. (2005). Selection on observed and unobserved

variables: Assessing the effectiveness of Catholic schools. Journal of Political Economy,

113(1), 151–184. https://doi.org/10.1086/426036

Angrist, J. D., & Pischke, J. S. (2008). Mostly harmless econometrics: An empiricist’s

companion. Princeton: Princeton University Press, (Chapter 3).

Bellows, J., & Miguel, E. (2009). War and local collective action in Sierra Leone. Journal of

Public Economics, 93(11–12), 1144–1157.

https://doi.org/10.1016/j.jpubeco.2009.07.012

Bierman, K. L., Coie, J. D., Dodge, K. A., Greenberg, M. T., Lochman, J. E., McMahon, R.

J., & Pinderhughes, E. (2010). The Effects of a Multiyear Universal Social-Emotional

Learning Program: The Role of Student and School Characteristics. Journal of

Consulting and Clinical Psychology, 78(2), 156. https://doi.org/10.1037/a0018607

Bogg, T., & Roberts, B. W. (2004). Conscientiousness and health-related behaviors: A meta-

analysis of the leading behavioral contributors to mortality. Psychological Bulletin,

130(6), 887. https://doi.org/10.1037/0033-2909.130.6.887

Borghans, L., Duckworth, A. L., Heckman, J. J., & Weel, B. (2008). The Economics and

Psychology of Personality Traits. The Journal of Human Resources, 43(4), 972–1059.

Borghans, L., Golsteyn, B. H. H., Heckman, J., & Humphries, J. E. (2011). Identification

problems in personality psychology. Personality and Individual Differences, 51(3), 315–

320. https://doi.org/10.1016/j.paid.2011.03.029

Bowles, S., Gintis, H., & Osborne, M. (2001). The determinants of earnings: A behavioral

approach. Journal of Economic Literature, 39(4), 1137–1176.

https://doi.org/10.1257/jel.39.4.1137

Burks, S. V., Lewis, C., Kivi, P. A., Wiener, A., Anderson, J. E., Götte, L., DeYoung, C. G.,

& Rustichini, A. (2015). Cognitive skills, personality, and economic preferences in

collegiate success. Journal of Economic Behavior and Organization, 115, 30–44.

https://doi.org/10.1016/j.jebo.2015.01.007

23

Caliendo, M., Fossen, F., & Kritikos, A. S. (2014). Personality characteristics and the

decisions to become and stay self-employed. Small Business Economics.

https://doi.org/10.1007/s11187-013-9514-8

Caspi, A., Roberts, B. W., & Shiner, R. L. (2005). Personality Development: Stability and

Change. Annual Review of Psychology, 53(453–84).

https://doi.org/10.1146/annurev.psych.55.090902.141913

Chamorro-Premuzic, T., & Furnham, A. (2003a). Personality predicts academic performance:

Evidence from two longitudinal university samples. Journal of Research in Personality,

37(4), 319–338. https://doi.org/10.1016/S0092-6566(02)00578-0

Chamorro-Premuzic, T., & Furnham, A. (2003b). Personality Traits and Academic

Examination Performance. European Journal of Personality, 17(3), 237–250.

https://doi.org/10.1002/per.473

Chapman, B. P., Duberstein, P. R., Sörensen, S., & Lyness, J. M. (2007). Gender differences

in Five Factor Model personality traits in an elderly cohort. Personality and Individual

Differences, 43(6), 1594–1603. https://doi.org/10.1016/j.paid.2007.04.028

Chiorri, C., Bracco, F., Piccinno, T., Modafferi, C., & Battini, V. (2015). Psychometric

properties of a revised version of the ten item personality inventory. European Journal of

Psychological Assessment, 31(2), 109–119. https://doi.org/10.1027/1015-5759/a000215

Cobb-Clark, D. A., & Schurer, S. (2013). Two economists’ musings on the stability of locus

of control. Economic Journal, 123(570), F358–F400. https://doi.org/10.1111/ecoj.12069

Cobb-Clark, D. A., & Tan, M. (2011). Noncognitive skills, occupational attainment, and

relative wages. Labour Economics, 18(1), 1–13.

https://doi.org/10.1016/j.labeco.2010.07.003

Conard, M. A. (2006). Aptitude is not enough: How personality and behavior predict

academic performance. Journal of Research in Personality, 40(3), 339–346.

https://doi.org/10.1016/j.jrp.2004.10.003

Conger, D., & Long, M. C. (2010). Why are men falling behind? gender gaps in college

performance and persistence. Annals of the American Academy of Political and Social

Science, 627(1), 184–214. https://doi.org/10.1177/0002716209348751

Costa, P T, & McCrae, R. R. (1992). Revised NEO personality inventory (NEO-PI-R) and

NEO five-factor inventory (NEO-FFI). In Odessa FL Psychological Assessment

Resources.

Costa, Paul T., McCrae, R. R., & Dye, D. A. (1991). Facet scales for agreeableness and

conscientiousness: A revision of tshe NEO personality inventory. Personality and

Individual Differences, 12(9), 887–898. https://doi.org/10.1016/0191-8869(91)90177-D

Costa, Paul T., Terracciano, A., & McCrae, R. R. (2001). Gender differences in personality

traits across cultures: Robust and surprising findings. Journal of Personality and Social

Psychology, 81(2), 322–331. https://doi.org/10.1037/0022-3514.81.2.322

Cubel, M., Nuevo-Chiquero, A., Sanchez-Pages, S., & Vidal-Fernandez, M. (2016). Do

Personality Traits Affect Productivity? Evidence from the Laboratory. Economic

Journal, 126(592), 654–681. https://doi.org/10.1111/ecoj.12373

24

Cunha, F., & Heckman, J. J. (2008). Formulating, Identifying and Estimating the Technology

of Cognitive and Noncognitive Skill Formation. Journal of Human Resources, 43(4),

738–782. https://doi.org/10.1353/jhr.2008.0019

De Bolle, M., De Fruyt, F., McCrae, R. R., Löckenhoff, C. E., Costa, P. T., Aguilar-Vafaie,

M. E., Ahn, C. K., Ahn, H. N., Alcalay, L., Simonetti, F., Allik, J., Avdeyeva, T. V.,

Bratko, D., Brunner-Sciarra, M., Cain, T. R., Chan, W., Chittcharat, N., Vanno, V.,

Crawford, J. T., … Terracciano, A. (2015). The emergence of sex differences in

personality traits in early adolescence: A cross-sectional, cross-cultural study. Journal of

Personality and Social Psychology, 108(1), 171–185. https://doi.org/10.1037/a0038497

De Feyter, T., Caers, R., Vigna, C., & Berings, D. (2012). Unraveling the impact of the Big

Five personality traits on academic performance: The moderating and mediating effects

of self-efficacy and academic motivation. Learning and Individual Differences, 22(4),

439–448. https://doi.org/10.1016/j.lindif.2012.03.013

Delaney, L., Harmon, C., & Redmond, C. (2011). Parental education, grade attainment and

earnings expectations among university students. Economics of Education Review, 30(6),

1136–1152. https://doi.org/10.1016/j.econedurev.2011.04.004

Delaney, L., Harmon, C., & Ryan, M. (2013). The role of noncognitive traits in undergraduate

study behaviours. Economics of Education Review, 32(1), 181–195.

https://doi.org/10.1016/j.econedurev.2012.07.009

Di Fabio, A., & Busoni, L. (2007). Fluid intelligence, personality traits and scholastic success:

Empirical evidence in a sample of Italian high school students. Personality and

Individual Differences, 43(8), 2095–2104. https://doi.org/10.1016/j.paid.2007.06.025

Diseth, Å. (2003). Approaches to Learning, Cognitive Style, and Motives as Predictors of

Academic Achievement. Educational Psychology, 23(2), 195–207.

https://doi.org/10.1080/01443410303225

Eurostat (2016). European Labour Force Survey ad hoc module on young people on the

labour market. https://ec.europa.eu/eurostat/web/lfs/data/database. Accessed 16.06.2020

Eurostat (2019a). Headline indicators database: Tertiary educational attainment by sex, age

group 30-34. https://ec.europa.eu/eurostat/data/database. Accessed 16.06.2020

Eurostat (2019b). Headline indicators database: Early leavers from education and training by

sex. https://ec.europa.eu/eurostat/data/database. Accessed 16.06.2020

Farsides, T., & Woodfield, R. (2003). Individual differences and undergraduate academic

success: The roles of personality, intelligence, and application. Personality and

Individual Differences, 34(7), 1225–1243. https://doi.org/10.1016/S0191-

8869(02)00111-3

Fletcher, J. M. (2013). The effects of personality traits on adult labor market outcomes:

Evidence from siblings. Journal of Economic Behavior and Organization, 89, 122–135.

https://doi.org/10.1016/j.jebo.2013.02.004

Furnham, A., Chamorro-Premuzic, T., & McDougall, F. (2003). Personality, cognitive ability,

and beliefs about intelligence as predictors of academic performance. Learning and

Individual Differences, 14(1), 47–64. https://doi.org/10.1016/j.lindif.2003.08.002

25

Goldberg, L. R. (1990). An Alternative “Description of Personality”: The Big-Five Factor

Structure. Journal of Personality and Social Psychology, 59(6), 126.

https://doi.org/10.1037/0022-3514.59.6.1216

González, F., & Miguel, E. (2015). War and local collective action in sierra leone: A

comment on the use of coefficient stability approaches. Journal of Public Economics,

128, 30–33. https://doi.org/10.1016/j.jpubeco.2015.05.004

Gosling, S. D., Rentfrow, P. J., & Swann, W. B. (2003). A very brief measure of the Big-Five

personality domains. Journal of Research in Personality, 37(6), 504–528.

https://doi.org/10.1016/S0092-6566(03)00046-1

Gray, E. K., & Watson, D. (2002). General and specific traits of personality and their relation

to sleep and academic performance. Journal of Personality, 70(2), 177–206.

https://doi.org/10.1111/1467-6494.05002

Heckman, J. J. (2000). Policies to foster human capital. Research in Economics, 54(1), 3–56.

https://doi.org/10.1006/reec.1999.0225

Heckman, J. J., Stixrud, J., & Urzua, S. (2006). The effects of cognitive and noncognitive

abilities on labor market outcomes and social behavior. Journal of Labor Economics,

24(3), 411–482. https://doi.org/10.1086/504455

Heckman, J., Jagelka, T., & Kautz, T. (2019). Some Contributions of Economics to the Study

of Personality (No. w26459). National Bureau of Economic Research.

https://doi.org/10.3386/w26459

Heckman, J., Pinto, R., & Savelyev, P. (2013). Understanding the mechanisms through which

an influential early childhood program boosted adult outcomes. American Economic

Review, 103(6), 2052–2086. https://doi.org/10.1257/aer.103.6.2052

Heineck, G., & Anger, S. (2010). The returns to cognitive abilities and personality traits in

Germany. Labour Economics, 17(3), 535–546.

https://doi.org/10.1016/j.labeco.2009.06.001

John, O. P., Naumann, L. P., & Soto, C. J. (2008). Paradigm shift to the integrative Big Five

Trait taxonomy: History, measurement, and conceptual issues. In Handbook of

personality: Theory and research. https://doi.org/10.1016/S0191-8869(97)81000-8

Kaiser, R. T., & Ozer, D. J. (1997). Emotional stability and goal-related stress. Personality

and Individual Differences, 22(3), 371–379. https://doi.org/10.1016/S0191-

8869(96)00208-5

Kappe, R., & Van Der Flier, H. (2012). Predicting academic success in higher education:

What’s more important than being smart? European Journal of Psychology of Education,

27(4), 605–619. https://doi.org/10.1007/s10212-011-0099-9

Kautz, T., Heckman, J. J., Diris, R., Weel, B. T., Borghans, L., ter Weel, B., & Borghans, L.

(2014). Fostering and measuring skills: Improving cognitive and non-cognitive skills to

promote lifetime success (NBER Working Paper No. 20749). OECD Education Working

Papers. https://doi.org/10.1787/19939019

Komarraju, M., Karau, S. J., & Schmeck, R. R. (2009). Role of the Big Five personality traits

in predicting college students’ academic motivation and achievement. Learning and

Individual Differences, 19(1), 47–52. https://doi.org/10.1016/j.lindif.2008.07.001

26

Kosse, F., Deckers, T., Pinger, P., Schildberg-Hörisch, H., & Falk, A. (2020). The formation

of prosociality: Causal evidence on the role of social environment. Journal of Political

Economy, 128(2), 000–000. https://doi.org/10.1086/704386

Lounsbury, J. W., Sundstrom, E., Loveland, J. M., & Gibson, L. W. (2003). Intelligence, “Big

Five” personality traits, and work drive as predictors of course grade. Personality and

Individual Differences. https://doi.org/10.1016/S0191-8869(02)00330-6

Martins, P. S. (2017). (How) Do Non-Cognitive Skills Programs Improve Adolescent School

Achievement? Experimental Evidence. CGR Working Paper 81.

McCrae, R. R., Costa, P. T., Terracciano, A., Parker, W. D., Mills, C. J., De Fruyt, F., &

Mervielde, I. (2002). Personality trait development from age 12 to age 18: Longitudinal,

cross-sectional, and cross-cultural analyses. Journal of Personality and Social

Psychology, 83(6), 1456–1468. https://doi.org/10.1037/0022-3514.83.6.1456

McCrae, R. R., & John, O. P. (1992). An Introduction to the Five‐Factor Model and Its

Applications. Journal of Personality, 60(2), 175–215. https://doi.org/10.1111/j.1467-

6494.1992.tb00970.x

McCrae, R. R., Terracciano, A., Khoury, B., Nansubuga, F., Kneževič, G., Djuric Jocic, D.,

Ahn, H. nie, Ahn, C. kyu, De Fruyt, F., Gülgöz, S., Ruch, W., Arif Ghayur, M., Avia, M.

D., Sánchez-Bernardos, M. L., Rossier, J., Dahourou, D., Fischer, R., Shakespeare-

Finch, J., Yik, M., … Camart, N. (2005). Universal features of personality traits from the

observer’s perspective: Data from 50 cultures. Journal of Personality and Social

Psychology, 88(3), 547–561. https://doi.org/10.1037/0022-3514.88.3.547

McCredie, M. N., & Kurtz, J. E. (2020). Prospective prediction of academic performance in

college using self- and informant-rated personality traits. Journal of Research in

Personality, 85(103911). https://doi.org/10.1016/j.jrp.2019.103911

Noftle, E. E., & Robins, R. W. (2007). Personality Predictors of Academic Outcomes: Big

Five Correlates of GPA and SAT Scores. Journal of Personality and Social Psychology,

93(1), 116–130. https://doi.org/10.1037/0022-3514.93.1.116

Nunn, N., & Wantchekon, L. (2011). The slave trade and the origins of Mistrust in Africa.

American Economic Review, 101(7), 3221–3252. https://doi.org/10.1257/aer.101.7.3221

Nyhus, E. K., & Pons, E. (2005). The effects of personality on earnings. Journal of Economic

Psychology, 26(3), 363–384. https://doi.org/10.1016/j.joep.2004.07.001

O’Connor, M. C., & Paunonen, S. V. (2007). Big Five personality predictors of post-

secondary academic performance. Personality and Individual Differences, 43(5), 971–

990. https://doi.org/10.1016/j.paid.2007.03.017

OECD (2019). Education at a glance. Paris: OECD Publications Service.

Oster, E. (2019). Unobservable Selection and Coefficient Stability: Theory and Evidence.

Journal of Business and Economic Statistics, 37(2), 187–204.

https://doi.org/10.1080/07350015.2016.1227711

Peterson, C. H., Casillas, A., & Robbins, S. B. (2006). The Student Readiness Inventory and

the Big Five: Examining social desirability and college academic performance.

Personality and Individual Differences, 41(4), 663–673.

https://doi.org/10.1016/j.paid.2006.03.006

27

Pomerantz, E. M., Altermatt, E. R., & Saxon, J. L. (2002). Making the grade but feeling

distressed: Gender differences in academic performance and internal distress. Journal of

Educational Psychology, 94(2), 396–404. https://doi.org/10.1037/0022-0663.94.2.396

Poropat, A. E. (2009). A Meta-Analysis of the Five-Factor Model of Personality and

Academic Performance. Psychological Bulletin, 135(2), 322–338.

https://doi.org/10.1037/a0014996

Pullmann, H., Raudsepp, L., & Allik, J. (2006). Stability and change in adolescents’

personality: A longitudinal study. European Journal of Personality.

https://doi.org/10.1002/per.611

Richardson, M., & Abraham, C. (2009). Conscientiousness and achievement motivation

predict performance. European Journal of Personality. https://doi.org/10.1002/per.732

Richardson, M., Abraham, C., & Bond, R. (2012). Psychological correlates of university

students’ academic performance: A systematic review and meta-analysis. Psychological

Bulletin, 138(2), 353–387. https://doi.org/10.1037/a0026838

Roberts, B., & DelVecchio, W. F. (2000). The rank-order consistency of personality traits

from childhood to old age: A quantitative review of longitudinal studies. Psychological

Bulletin, 126(1), 3. https://doi.org/10.1037/0033-2909.126.1.3

Roberts, B. W. (2009). Back to the future: Personality and Assessment and personality

development. Journal of Research in Personality, 43, 137–145.

https://doi.org/10.1016/j.jrp.2008.12.015

Roberts, B. W., Walton, K. E., & Viechtbauer, W. (2006). Patterns of mean-level change in

personality traits across the life course: A meta-analysis of longitudinal studies.

Psychological Bulletin, 132(1), 1–25. https://doi.org/10.1037/0033-2909.132.1.1

Rodríguez-Hernández, C. F., Cascallar, E., & Kyndt, E. (2020). Socio-economic status and

academic performance in higher education: A systematic review. Educational Research

Review, 20. https://doi.org/10.1016/j.edurev.2019.100305

Salgado, J. F. (1997). The five factor model of personality and job performance in the

European Community. Journal of Applied Psychology, 82(1), 30.

https://doi.org/10.1037/0021-9010.82.1.30

Smidt, W. (2015). Big Five personality traits as predictors of the academic success of

university and college students in early childhood education. Journal of Education for

Teaching, 41(4), 385–403. https://doi.org/10.1080/02607476.2015.1080419

Srivastava, S., John, O. P., Gosling, S. D., & Potter, J. (2003). Development of Personality in

Early and Middle Adulthood: Set Like Plaster or Persistent Change? Journal of

Personality and Social Psychology, 84(5), 1041–1053. https://doi.org/10.1037/0022-

3514.84.5.1041

Störmer, S., & Fahr, R. (2013). Individual determinants of work attendance: Evidence on the

role of personality. Applied Economics, 45(19), 2863–2875.

https://doi.org/10.1080/00036846.2012.684789

Vedel, A. (2014). The Big Five and tertiary academic performance: A systematic review and

meta-analysis. Personality and Individual Differences, 71, 66–76.

https://doi.org/10.1016/j.paid.2014.07.011

28

Vedel, A., Thomsen, D. K., & Larsen, L. (2015). Personality, academic majors and

performance: Revealing complex patterns. Personality and Individual Differences, 85,

69–76. https://doi.org/10.1016/j.paid.2015.04.030

Von Stumm, S., Hell, B., & Chamorro-Premuzic, T. (2011). The hungry mind: Intellectual

curiosity is the third pillar of academic performance. Perspectives on Psychological

Science, 6(6), 574–588. https://doi.org/10.1177/1745691611421204

Ziegler, M., Danay, E., Schölmerich, F., & Bühner, M. (2010). Predicting academic success

with the big 5 rated from different points of view: Self-rated, other rated and faked.

European Journal of Personality: Published for the European Association of Personality

Psychology, 24(4), 341-355. https://doi.org/10.1002/per.753

29

Appendix (not intended for publication)

This Appendix includes further results discussed in Section 6 (“Robustness checks”) and will

be made available online by the authors.

30

Table A1. Estimated effects of the Big Five personality traits on GPA: percentile ranks of

GPA scores.

(1) (2) (3) (4)

Extraversion -0.008 -0.008 -0.008 -0.007

(0.006) (0.006) (0.006) (0.006)

Agreeableness -0.012* -0.013** -0.014** -0.012*

(0.006) (0.006) (0.006) (0.006)

Conscientiousness 0.032*** 0.028*** 0.027*** 0.027***

(0.006) (0.005) (0.005) (0.005)

Emotional Stability -0.017** -0.010 -0.009 -0.009

(0.008) (0.007) (0.006) (0.006)

Openness to experience 0.011** 0.012** 0.011** 0.011** (0.005) (0.005) (0.005) (0.005)

Female 0.039*** 0.038*** 0.042***

(0.013) (0.014) (0.013)

Age -0.321** -0.344** -0.300**

(0.144) (0.142) (0.146)

Age squared 0.313** 0.337** 0.294*

(0.146) (0.144) (0.148) Type of upper secondary school

Liceo for scientific studies -0.036** -0.038** -0.039**

(0.017) (0.017) (0.016)

Liceo for other studies -0.056*** -0.057*** -0.055***

(0.013) (0.012) (0.013)

Technical/vocational school -0.109*** -0.109*** -0.111***

(0.019) (0.017) (0.017) ERC sectors

Physical Sciences and Engineering 0.025* 0.020 0.015

(0.013) (0.014) (0.014)

Life Sciences 0.018 0.018 0.022

(0.015) (0.016) (0.017)

Parental controls YES YES

Municipality/ province of provenience YES

Constant 0.496*** 0.514*** 0.561*** 0.574***

(0.002) (0.016) (0.029) (0.028)

Observations 3242 3242 3242 3242

F 7.629 18.090 100.890 160.148

p-value 0.000 0.000 0.000 0.000

Note. The table shows the estimated effects of the Big Five traits on university performance, which

is obtained from the percentile ranks of GPA scores by course of study. Robust standard errors

clustered by course of study indicators are in parentheses. The omitted category of upper

secondary school is the liceo for classical studies. The omitted category of ERC sector is Social

Sciences and Humanities. Parental controls include educational attainment, occupational status and

industry. * p < 0.10, ** p < 0.05, *** p < 0.01

31

Table A2. Estimated effects of the Big Five personality traits on GPA: Equivalent Economic

Situation Indicator (ISEE).

(1) (2)

Extraversion -0.023 -0.021

(0.017) (0.018)

Agreeableness -0.037 -0.037

(0.026) (0.026)

Conscientiousness 0.110*** 0.106***

(0.022) (0.022)

Emotional stability -0.028 -0.027 (0.023) (0.022)

Openness to experience 0.040** 0.037**

(0.018) (0.018)

Female 0.108** 0.101**

(0.045) (0.046)

Age -0.500 -0.499

(0.646) (0.621)

Age squared 0.486 0.485

(0.654) (0.629)

Type of upper secondary school

Liceo for scientific studies -0.073 -0.080

(0.058) (0.059)

Liceo for other studies -0.165*** -0.176***

(0.052) (0.050)

Technical/vocational school -0.372*** -0.383***

(0.063) (0.057)

ERC sectors

Physical Sciences and Engineering 0.136** 0.130**

(0.058) (0.059)

Life Sciences 0.095* 0.090*

(0.050) (0.052)

Logarithm of ISEE -0.024 -0.016

(0.015) (0.016)

Size of the household 0.029 0.024

(0.030) (0.030)

Parental controls YES

Municipality/ province of provenience YES YES

Constant -0.023 0.063

(0.122) (0.146)

Observations 2341 2341

F 12.096 287.377

p-value 0.000 0.000

Note. The table shows the estimated effects of the Big Five traits on GPA

once including the ISEE variable (measured as the standardized logarithms of

the ISEE values) and the number of household members as regressors.

Robust standard errors clustered by course of study indicators are in

parentheses. The omitted category of upper secondary school is the liceo for

classical studies. The omitted category of ERC sector is Social Sciences and

Humanities. Parental controls include educational attainment, occupational

status and industry. * p < 0.10, ** p < 0.05, *** p < 0.01

32

Table A3. Estimated effects of the Big Five personality traits on GPA: heterogeneity analysis

on ERC sectors.

(1)

Extraversion -0.019 (0.019)

Agreeableness -0.039* (0.022)

Conscientiousness 0.093*** (0.017)

Emotional stability -0.015

(0.051)

Openness to experience 0.039**

(0.017)

ERC sectors

Physical Sciences and Engineering 0.064

(0.047)

Life Sciences 0.053

(0.051)

Extraversion × Physical Sciences and Engineering -0.044

(0.054)

Emotional stability × Social Sciences and Humanities 0.021

(0.058)

Emotional stability × Life Sciences -0.057

(0.054)

Female 0.135***

(0.046)

Age -1.079**

(0.501)

Age squared 1.055**

(0.505)

Type of upper secondary school

Liceo for scientific studies -0.102*

(0.055) Liceo for other studies -0.174***

(0.036) Technical/vocational school -0.345***

(0.059) Parental controls YES

Constant 0.200*

(0.101)

Observations 3242

F 124.843

p-value 0.000

Note. The table shows a heterogeneity analysis of the effects of extraversion and

emotional stability on GPA when interactions between each Big Five trait not

uniformly distributed across fields of study and the corresponding ERC sector are

included in the regression equation. Robust standard errors clustered by course of

study indicators are in parentheses. The omitted category of upper secondary school is the liceo for classical studies. The omitted category of ERC sector is Social

Sciences and Humanities. Parental controls include educational attainment,

occupational status and industry. The estimated effect of emotional stability for

students in Life Sciences is negative and statistically significant (β= -.072, p<.01). * p < 0.10, ** p < 0.05, *** p < 0.01