Content uploaded by Benedikt Hell
Author content
All content in this area was uploaded by Benedikt Hell on Feb 10, 2014
Content may be subject to copyright.
Intellectual Curiosity Predicts Academic Performance
1
Running Head: INTELLECTUAL CURIOSITY PREDICTS ACADEMIC PERFORMANCE.
The hungry mind: Intellectual curiosity is the third pillar of academic performance
Sophie von Stumma*, Benedikt Hellb, & Tomas Chamorro-Premuzicc
a Department of Psychology, University of Chichester, Chichester, PO19 6PE, West
Sussex, UK.
b School of Applied Psychology, University of Applied Sciences Northwestern
Switzerland, 4600 Olten, Switzerland.
c Department of Psychology, Goldsmiths University of London, New Cross, SE14
6NW, London, UK.
Key words: Intelligence; academic performance; conscientiousness; intellectual
curiosity; meta-analysis.
*Corresponding author:
Sophie von Stumm
Department of Psychology
University of Chichester
College Lane
PO19 6PE
Chichester, West Sussex
E-mail address: s.vonstumm@chi.ac.uk (Sophie von Stumm).
Intellectual Curiosity Predicts Academic Performance
2
Abstract
Over the past century, academic performance has become the gatekeeper to
institutions of higher education, shaping career paths and individual life trajectories.
Accordingly, much psychological research has focused on identifying predictors of academic
performance, with intelligence and effort emerging as core determinants. In this paper, we
propose expanding on the traditional set of predictors by adding a third agency: intellectual
curiosity.
A series of path-models based on a meta-analytically derived correlation matrix
showed that a) intelligence is the single most powerful predictor of academic performance; b)
the effects of intelligence on academic performance are not mediated by personality traits; c)
intelligence, Conscientiousness (as marker of effort), and Typical Intellectual Engagement (as
marker of intellectual curiosity) are direct, correlated predictors of academic performance;
and d) the additive predictive effect of the personality traits of intellectual curiosity and effort
rival that the influence of intelligence. Our results highlight that a „hungry mind‟ is a core
determinant of individual differences in academic achievement.
Abstract word count: 160.
Intellectual Curiosity Predicts Academic Performance
3
Curiosity is, in great and generous
minds, the first passion and the last.
Samuel Johnson (1750)
The selection of candidates for higher education and professional recruitment is
traditionally based upon previous academic performance. Inter-personal variations in
academic performance, for example in school and university, have been explained in terms of
individual differences in intelligence and personality traits (e.g., Alexander, 1935; Poropat,
2009; Webb, 1915). In particular, students with higher cognitive ability (quicker learners), as
well as those who are more hard-working and well-organized (higher Conscientiousness),
tend to perform better in educational settings. That is, ability and effort are important
determinants of academic achievement; however, their application is driven by a third, to date
often overlooked factor: intellectual curiosity.
In this article, we first briefly review the societal function of academic performance in
the context of educational and occupational status attainment. Then, we discuss the research
literature that focuses on ability and non-ability factors as psychological predictors of
academic performance. Finally, based on meta-analytic evidence and theoretical
considerations, we demonstrate the importance of a curious mind for scholarly success in
addition and in relation to ability and effort.
Academic Performance: Why it Matters
In the second half of the 19th century, the industrial revolution led to an increasing
specialization and complexity of jobs. As a result, compulsory schooling was introduced in
the US and Europe to enable the general population to meet the latest job demands (Martin,
2008). Because of the new emphasis on educational qualifications, individual careers became
Intellectual Curiosity Predicts Academic Performance
4
less predefined by social class background or parental occupation, but depended more on
demonstrated ability and skill.
Prior to World War I, only a small fraction of the population extended their education
beyond elementary schooling, and even Ivy League universities, including Harvard and Yale,
could admit all applicants without reaching the limits of their capacity (Hubin, 1988;
Lehmann, 1999). However, as more and more people sought higher education to enhance
their employability, universities had to introduce selective student admissions1. Thus,
previous academic performance became the gatekeeper to higher education and a master key
to the labor market. Today, academic performance continues to be understood as an accurate
proxy for aptitude and is a core determinant of career paths and status attainment, even though
some question its value (Chamorro-Premuzic & Furnham, 2010). Academic performance is
also a key to understanding the development of one of psychology‟s most well-known „tools‟,
namely the intelligence test.
Academic Performance and Intelligence: Criterion par excellence?
Sir Francis Galton (1822 – 1911), the father of intelligence research (Fancher, 1985)2,
was the first to suggest that individual differences in intelligence were reflected in academic
performance outcomes:
There can hardly be a surer evidence of the enormous difference
between the intellectual capacity of men, than the prodigious
1 This is admittedly a simplified account of American education history; please see Lehmann (1999) for a more
detailed review.
2 In 1575, the Spanish physician Juan Huarte de San Juan published „Examen de ingenios para las ciencias‟,
which may be considered the earliest scientific writing on intelligence (Fernandez-Ballesteros & Colom, 2004).
Intellectual Curiosity Predicts Academic Performance
5
differences in the numbers of marks obtained by those who gain
mathematical honours at Cambridge (Galton, 1869, p. 16).
Because academic performance was thought to mirror individual differences in ability,
it became the “the criterion par excellence” for intelligence tests (Chamorro-Premuzic &
Furnham, 2006, p. 253). Indeed, Théodore Simon (1872 – 1961) and Alfred Binet (1857 –
1911) developed the first intelligence test to identify children who struggled with the school
curriculum and their academic performance. Likewise, subsequently developed ability tests
were (and continue to be) validated by educational achievement as accurate measures of
intelligence (e.g., Spearman, 1904; Terman, 1916). Indeed, if an intelligence test failed to
account for inter-individual differences in academic performance, it was not regarded as a
meaningful measure of intellectual capacity (e.g., Bolton, 1892; Sharp, 1899).
At present, an abundance of empirical research shows that mental ability test scores
are substantially correlated with academic performance, reaching values of up to r = .81
(Deary, Strand, Smith, & Fernandes, 2007), although cross-sectional correlations tend to be
lower than r = .50 (e.g., Johnson, McGue, & Iacono, 2005; see also Hell, Trapmann, &
Schuler, 2007; Kuncel, Hezlett & Ones, 2004; Poropat, 2009; Sacket, Kuncel, Arneson,
Cooper & Waters, 2009). The association between cognitive ability and academic
performance persists across educational levels, although it tends to decrease in more advanced
academic settings due to differential range restrictions. For instance, in graduate school
candidates have been selected already on the basis of their intellectual capacity, which
increases the relative variability and importance of non-ability factors (cf. Jensen, 1980). In
line with this, recent research has assessed the degree to which individual differences in
academic performance can be explained by personality factors.
Intellectual Curiosity Predicts Academic Performance
6
Academic Performance beyond Intelligence: Does Personality Matter?
Although intelligence is an important predictor of academic success, recent research
has shown that personality dispositions, notably traits assessing individuals‟ typical levels of
persistence and hard-work, account for substantial amounts of variance in academic
performance (e.g., O‟Connor & Paunonen, 2007; Poropat, 2009; Trapmann, Hell, Weigand &
Schuler, 2007). For example, Chamorro-Premuzic and Furnham (2003b) found that
personality traits accounted for quadruple the variance in exam results of elite university
students compared to intelligence. This result echoes the effect of range restriction in
intelligence on the predictive validity for non-ability factors (see above).
Maximum versus Typical Performance
Non-ability traits have traditionally been operationalized by typical performance
measures, reflecting the strength of a behavioral tendency for accomplishment, whereas
ability is ordinarily construed as measures of maximal performance (Cronbach, 1949; Fiske &
Butler, 1963). Ability test scores indicate what an individual can do whereas personality
scales provide a measure of what a person is most likely to do (Fiske & Butler, 1963, p. 258-
259). Klehe and Anderson (2007) demonstrated in a recent laboratory study that behavior
dispositions, including the direction and level of effort, as well as participants‟ perceived self-
efficacy, were more predictive of typical than of maximum performance outcomes.
Conversely, ability, which was conceptualized in terms of declarative knowledge and
procedural skills, was found to be of greater significance for maximum than typical
performance outcomes (Klehe & Anderson, 2007). The authors concluded that psychological
predictors of accomplishment vary in their predictive validity across maximum versus typical
performance settings depending on the nature of the measurement instrument in question.
Intellectual Curiosity Predicts Academic Performance
7
In education, assessments of academic achievement span both conditions of maximum
and typical performance. For example, students usually demonstrate their learning success by
addressing specific questions or topics in timed examinations. Although such examinations
constitute a maximum performance setting, the assessment outcome also reflects the students‟
typical performance in terms of their breadth and depth of preparation. Therefore, non-ability
factors are plausibly more meaningful variables when determining academic performance
outcomes than cognitive ability measures, which invariably capture maximum rather than
typical potential.
The Role of Conscientiousness
Since the early 1990s, there has been a growing consensus on the Five Factor Model
as the preferred taxonomy for classifying individual differences in normal personality (e.g.,
Goldberg, 1990). This model comprises five major dimensions of personality: Neuroticism,
Extraversion, Openness to Experience, Agreeableness and Conscientiousness (Costa &
McCrae, 1992).
Of these, Conscientiousness has been repeatedly shown to be positively related to the
academic performance of university students (e.g., Chamorro-Premuzic & Furnham, 2003a, b;
2008; Poropat, 2009) as well as to several job performance criteria across a broad range of
occupations (Barrick & Mount, 1991; Chamorro-Premuzic & Furnham, 2010; Salgado, 1997;
Tett, Jackson, & Rothstein, 1991). Conscientiousness spans six facets – Competence
(efficacy), Order (planning ahead), Dutifulness (following rules), Achievement striving
(effort), Self-discipline, and Deliberation (Costa & McCrae, 1992) – that indicate individual
differences in persistence, responsibility, and effort, all of which are associated with better
academic and occupational performance. Several recent meta-analyses estimated associations
between indicators of academic performance and Conscientiousness from r = .23 to r = .27
(O‟Connor & Paunonen, 2007; Poropat, 2009; Trapmann et al., 2007). Even though the
Intellectual Curiosity Predicts Academic Performance
8
magnitude of these associations confirms the importance of Conscientiousness in academic
settings, the construct was not initially conceptualized for the purpose of predicting school or
university performance.
Intelligence and Conscientiousness have been found to be largely independent,
although some studies reported modest negative correlations between Conscientiousness and
ability measures (e.g., Ackerman & Heggestad, 1997; Moutafi, Furnham, & Crump, 2006).
To explain this negative association, it has been argued that „less‟ able individuals may
become increasingly more conscientious in order to compensate for their lower levels of
cognitive ability, whereas more intelligent people rely to a greater extent on their intelligence
and can „afford‟ to be less dutiful and organized and nevertheless excel (Chamorro-Premuzic
& Furnham, 2005). According to this theory, the effects of intelligence on academic
performance would be mediated by Conscientiousness in an inconsistent mediation model
(MacKinnon & Fairchild, 2009). That is, intelligence would have a direct positive effect on
academic performance, as well as an indirect negative effect, mediated by Conscientiousness.
Therefore, direct and indirect effects would be of opposite signs or inconsistent. It is not clear
yet if intelligence and Conscientiousness are independent predictors of academic performance
or if one mediates the effects of the other.
Effort, Intelligence… and what else?
It has been argued that crystallized intelligence, which “consists of discriminatory
habits long established in a particular field” (Cattell, 1943, p. 178) results from the application
of fluid intelligence, which is the “ability to discriminate and perceive relations between any
fundaments, new or old” (Cattell, 1943, p. 178). In simple words, knowledge and expertise
result from applying one‟s reasoning ability. The direction and strength of such application, in
turn, is directed by so-called „investment traits‟ (Cattell, 1943; 1971); that is, personality
Intellectual Curiosity Predicts Academic Performance
9
characteristics that determine where, when and how people apply their mental capacity.
Accordingly, investment traits explain inter-individual differences in the pursuit of learning
opportunities, such as visiting museums and galleries; solving riddles and puzzles; and
reading the newspapers. Hayes (1962) suggested that all variation in intelligence resulted
from individual differences in the drive or motivation to pursue learning opportunities. He
claimed that “differences commonly referred to as intellectual [are] nothing more than
differences in acquired abilities” (p. 303), and rejected the existence of a general intelligence
factor. Even though Hayes‟ (1962) motivational-experiential theory takes an extreme stand
(cf. McDougall, 1933), it is plausible that the motivation to learn is reflected in differences in
acquired skills.
In the psychological literature, numerous theoretical and psychometric concepts have
been proposed to capture individual differences in the desire to comprehend and engage in
cognitively demanding tasks and hence, to invest in one‟s intellectual competence (von
Stumm, 2010). However, these so-called investment traits have to date not been explicitly
associated with research on curiosity and exploration (Ackerman & Heggestad, 1997;
Berlyne, 1954; 1960; Litman & Spielberger, 2003), despite their striking resemblance.
Investment and Curiosity
Historically, different types of curiosity have been identified: Hume (1777/ 1888, p.
453) theoretically differentiated the curiosity of “love of knowledge” from the “passion
derived from a quite different principle [that is] an insatisfiable desire for knowing the actions
and circumstances of neighbours”. Berlyne (1954) proceeded to introduce the conceptual
distinction between epistemic and perceptual curiosity. Epistemic curiosity refers to
individual differences in seeking out opportunities for intellectual engagement, acquiring facts
and knowledge, or simply the “drive to know” (Berlyne, 1954, p. 187), whereas perceptual
curiosity is evoked by visual, auditory, and tactile stimulation and refers to a “drive to
Intellectual Curiosity Predicts Academic Performance
10
experience and feel” (Berlyne, 1954). Later, Litman and colleagues developed corresponding
psychometric scales to assess epistemic and perceptual curiosity (cf. Litman & Spielberger,
2003; Collins, Litman, & Spielberger, 2004). Epistemic curiosity is conceptually very similar
to other intellectual investment traits, all of which refer to a desire or hunger for knowledge.
For example, Cacioppo and Petty (1982) sought “to identify differences among individuals in
their tendency to engage in and enjoy thinking” (p. 116), and thus developed the concept and
scale of „Need for Cognition‟, which stretches from “cognitive misers to cognizers”
(Cacioppo, Petty, Feinstein, & Jarvis, 1996; p. 197). Later, Goff and Ackerman (1992)
proposed Typical Intellectual Engagement (TIE) as “a dispositional construct that […] is
associated with intelligence as typical performance” (p. 539). The TIE scale captures people‟s
typical expression of engaging with and understanding their environment, and their desire to
solve and be absorbed by complex, intellectual problems (Goff & Ackerman, 1992). To that
effect, TIE specifically refers to settings of advanced stages of education, in which the
predictive validity of maximal intelligence is diminished (Goff & Ackerman, 1994).
Need for Cognition, epistemic curiosity, and TIE are exemplary representatives of a
group of investment trait constructs that describe tendencies to seek out, engage in, enjoy, and
pursue opportunities for effortful cognitive activity; in short, intellectual curiosity. In addition
to their conceptual similarities, trait scales of intellectual curiosity also share a number of
semantically identical items (von Stumm, 2010). Not surprisingly, epistemic curiosity, need
for cognition and other investment traits have been found to lack discriminant validity3 (e.g.,
Mussel, 2010; Roecklin, 1994; Woo, Harms, & Kuncel, 2007). Furthermore, investment traits
are uniformly positively associated with academic performance with medium effect sizes
(Cacioppo et al., 1996; von Stumm, 2010), and also with intelligence but to a notably lesser
extent (e.g., Cacioppo, Petty, & Morris, 1983; Furnham, Swami, Arteche, & Chamorro-
3 Discriminant validity describes the degree to which one measurement instrument diverges from others, which
are theoretically different (Campbell & Fiske, 1959).
Intellectual Curiosity Predicts Academic Performance
11
Premuzic, 2008; Goff & Ackerman, 1992). That is, measures of intellectual investment and
curiosity have matching conceptual roots, include semantically identical items, and share
criteria validity for academic performance and intelligence; therefore, they appear to assess
the same trait dimension, and corresponding scales might be interchangeably used.
Investment and Openness to Experience
In the Five Factor Model, Openness to Experience comprises six facets, namely
Fantasy (vivid imagination); Aesthetic sensitivity; attentiveness to inner Feelings; Actions
(engagement in unfamiliar and novel activities); Ideas (intellectual curiosity); and Values,
which refers to the readiness to re-examine traditional social, religious, and political concepts
(Costa & McCrae, 1992; McCrae, 1994). Openness to Experience is conceptually very similar
to intellectual investment trait scales (Ackerman, 1996; Chamorro-Premuzic & Furnham,
2006). Furthermore, Openness is associated with general intelligence and domain-specific
knowledge (e.g., Ackerman & Heggestad, 1997; Ackerman & Rolfhus, 1999). It has been
argued that more intelligent individuals are better capable of understanding difficult
information and of processing new experiences, which in turn facilitates open-minded
attitudes and expands knowledge (e.g., Moutafi et al., 2006). Conversely, individuals with
low levels of intelligence are more challenged by intellectually demanding tasks, and prefer
routine and to some degree closed-mindedness (that is not to say, smart individuals could not
also be closed-minded and dogmatic). However, three recent meta-analyses on Openness and
academic performance estimated correlations between .06 and .13 (O‟Connor & Paunonen,
2007; Poropat, 2009; Trapmann et al., 2007), suggesting that Openness may have negligible
effects on academic performance outcomes.
In a recent series of studies, DeYoung and colleagues (2005; 2007; 2009) empirically
substantiated previous notions of Openness incorporating two related, but distinct factors
(e.g., Saucier, 1992): Intellect, reflecting intellectual engagement with the facet Ideas as main
Intellectual Curiosity Predicts Academic Performance
12
marker, and Openness, comprising artistic and contemplative qualities related to engagement
in sensation and perception including facets of Fantasy, Aesthetics, Feelings and Actions.
Note that the facet scale of Values was not found to be a distinct marker of either Openness
aspect (DeYoung, Petterson, & Higgins, 2005). Using fMRI in a sample of 104 community
members from the Washington area, DeYoung, Shamosh, Green, Braver, and Gray (2009)
showed that Intellect was associated with brain activity in neural systems of working memory
but Openness was not. The authors concluded that Openness to Experience comprised two
separable, neurally distinctive aspects of one larger personality domain (Figure 1).
---------------------------------
Insert Figure 1 here
---------------------------------
Further evidence for the two-dimensionality of Openness comes from behavior
genetics. Wainwright and colleagues (2008) analyzed data from 754 families on intelligence,
academic achievement, and the six Openness facets. The results showed a general genetic
factor that explained variance in intelligence, academic performance and several Openness
facets. Most notably, the general factor was associated with the facets Ideas and Values.
Conversely, a specific genetic factor was related to the facets Fantasy, Aesthetics, Feelings
and Action (Wainwright et al., 2008). Overall, these results suggest that Intellect, marked by
Ideas and Values, shares more genetic variance with intelligence and academic performance,
than Openness, marked by Fantasy, Aesthetics, Feelings and Actions. Studies reporting low
phenotypic associations of Openness and intellectual accomplishments typically measure
Openness as higher order factor, rather than sampling its facets. Therefore, the apparent lack
of empirical evidence for associations of Openness and academic performance may be due to
a methodological problem. That is, the investment theory is not invalidated because of
negligible associations between Openness and academic performance but an alternative, more
precise conceptualization of intellectual curiosity should be put to test.
Intellectual Curiosity Predicts Academic Performance
13
The Current Study
To date, the role of intellectual curiosity has not been studied within the complex
nexus of academic achievement predictors. In this study, we to (1) empirically evaluate our
proposal of intellectual curiosity as core determinant of academic performance; (2) compare
associations of Openness to Experience and intellectual curiosity with academic performance;
and (3) disentangle curiosity‟s associations with intelligence and Conscientiousness.
Following Viswesvaran and Ones‟ (1995) approach, we compose a correlation matrix of
meta-analytic coefficients to fit a series of path models. In part, correlation coefficients were
extracted from previously published meta-analyses on associations between academic
performance, intelligence, Openness and Conscientiousness. Because no meta-analysis to
date reported corresponding associations with intellectual curiosity, four new meta-analyses
were conducted focusing on TIE as representative construct for intellectual curiosity. We
chose TIE as representative scale because compared to other investment trait scales, such as
Need for Cognition and epistemic curiosity, it has been most frequently employed in research
on intelligence, personality and academic performance. Below, we briefly outline the
employed methods; a detailed account can be found in the Appendix.
Methods
Database Searches
The psychological database PsychInfo was searched for large-scale meta-analytic
reviews that investigated associations among two or more variables, including academic
performance, Conscientiousness and Openness (measured within the Five Factor Model), and
intelligence. We identified three excellent studies: Kuncel, Hezlett and Ones (2004); Judge,
Jackson, Shaw, Scott and Rich (2007); and Poropat (2009). From each of those we borrowed
one or more meta-analytic coefficients to compose a correlation matrix for our analysis;
Intellectual Curiosity Predicts Academic Performance
14
details on these studies, their methods, and the choice of coefficients are outlined in the
Appendix.
For TIE, no suitable meta-analytic study was previously published4 and subsequently,
four new, independent meta-analyses were conducted. To this end, a literature search was
conducted on PsychInfo and ERIC with the key term “typical intellectual engagement”.
Identified studies were excluded from the analysis if they a) did not include empirical data, b)
did not include zero-order correlations, and c) reported previously published data (e.g.,
Rocklin, 1994). References of all studies were screened for additional manuscripts. Overall,
11 studies were identified (Table 1), all of which employed the same measure of TIE (Goff &
Ackerman, 1992) and comprised predominantly student samples. Without exception, the
identified studies operationalized Conscientiousness and Openness to Experience with
measures from the Five Factor Model. Similarly, academic performance was consistently
assessed as grade point average (GPA) or as an academic performance composite. For
intelligence, only tests measuring general intelligence and omnibus IQ tests were included.
The obtained coefficients were corrected for sampling and measurement error, and meta-
analyzed following the validation generalization approach in random effect models (see
Appendix).
-----------------------------------
Insert Table 1 here
-----------------------------------
Results
Results of the TIE meta-analyses
As shown in Table 2, TIE was most strongly associated with Openness to Experience
at
ˆ
= .64 (N = 1,998), followed by academic performance with
ˆ
= .33 (N = 608). The
4 Ackerman and Heggestad (1997) computed meta-analytic associations of TIE and factors of intelligence;
however, these coefficients were mostly based on an insufficient number of studies.
Intellectual Curiosity Predicts Academic Performance
15
association between TIE and Conscientiousness was at
ˆ
= .28 (N = 1,662) and between TIE
and intelligence at
ˆ
=.22 (N = 1,230). Table 3 shows the correlation matrix that was used
for the subsequent path models.
-----------------------------------
Insert Table 2 and 3 here
-----------------------------------
Models of Academic Performance
In a stepwise process, five path models were fitted; Table 4 shows the model fit index
results across tested models. Model fit was assessed using the model χ² (Jöreskog, 1969), the
incremental goodness-of-fit indices including Comparative Fit Index (CFI), the Tucker-Lewis
Index (TLI), as well as the Root-mean-square error of approximation (RMSEA). CFI and TLI
indicate an adequate model fit at values of .90 and .95 or above (Hu & Bentler, 1999), while
RMSEA values of .08 and below are considered acceptable (Browne & Cudeck, 1993).
--------------------------------------
Insert Figure 2 & 3 here
--------------------------------------
Model 0 was a full inter-correlation model whereby all predictor variables were
allowed to freely correlate and to directly affect academic performance. It was a just-
identified, saturated model with 0 degrees of freedom and a χ² of 0, which makes the
computation of informative fit indices impossible (Table 4). The model showed a negative
association between Openness and academic performance of -.26 after controlling for the
predictor‟s high inter-correlations with the other variables; subsequently, Openness was
excluded from all models (Figure 2 & 3). Model 1 constituted a full mediation model,
whereby TIE and Conscientiousness independently, fully mediated the effects of intelligence
on academic performance (Figure 2). The model fit indices suggested a poor fit. Model 2
replicated the previous model but allowed additionally for the two mediators – TIE and
Intellectual Curiosity Predicts Academic Performance
16
Conscientiousness – to freely correlate. Model fit indices suggested a slight improvement
over the previous model but by no means an adequate fit. Model 3 tested if intelligence had a
direct effect on academic performance and also indirectly influenced the outcome through
TIE and Conscientiousness. That is, TIE and Conscientiousness were specified to mediate the
impact of intelligence on academic performance (Figure 2). Model 3 showed a poor fit to the
data. Overall, the series of models suggested that personality traits were inadequate mediators
of intelligence effects on academic performance and thus, a final Model 4 conceptualized
Conscientiousness, intelligence and TIE as direct predictors of academic performance,
controlling for the predictor‟s inter-correlations (Figure 2). Note that there was no correlation
between Conscientiousness and intelligence. This model accounted overall for 25.7% of the
variance in academic performance and proved a superior and acceptable fit (Figure 3b). All
paths were significant at p < .01. TIE correlated with intelligence at .23 and with
Conscientiousness at .29. After controlling for these positive associations among the predictor
variables, intelligence sustained the strongest effect on academic performance with a path
weight of .35. TIE and Conscientiousness had slightly lower, identical path parameters of .20.
-----------------------------------
Table 4 here
-----------------------------------
Models of Academic Performance:
The Interplay of Intelligence, Effort and Investment
Previous research has identified intelligence and effort as „core pillars‟ of academic
performance but other variables have traditionally received less attention in the prediction of
scholarly success. The current study evaluated intellectual curiosity as potentially meaningful
Intellectual Curiosity Predicts Academic Performance
17
third „pillar‟ of academic achievement. To this end, four independent meta-analyses of
Typical Intellectual Engagement (TIE) estimated its associations with general intelligence,
Conscientiousness, Openness to Experience, and academic performance. Furthermore,
psychological predictors of academic performance were investigated within meta-analytic
path models.
Consistent with previous research (e.g., Goff & Ackerman, 1992; Mussel, 2010;
Rocklin, 1994), Openness and TIE overlapped considerably, sharing 41% of variance. Also
consistent with previous research (e.g., Ackerman, 1996; Ackerman, Kanfer & Goff, 1995;
Goff & Ackerman, 1992), both constructs differed in their associations with general
intelligence and academic performance. TIE was more strongly related to academic
performance than to intelligence, even though the difference was small. In contrast, Openness
shared a substantial amount of variance with intelligence, and almost none with academic
performance (cf. Judge et al., 2007; Poropat, 2009). TIE and Openness also differed in their
associations with Conscientiousness, with TIE being more strongly linked to this omnibus
measure of persistence and diligence than Openness.
The second set of analyses evaluated a series of path models (Figure 2 and 3). Here,
TIE and Openness differed substantially in the direction of association with academic
performance, after controlling for their associations with the remaining predictor variables. To
that effect, Openness was shown to negatively affect academic performance, whereas TIE was
a strong positive predictor, despite its considerable inter-correlations with the remaining
predictors.
The observed differences in associations of Openness and TIE with academic
performance, intelligence and Conscientiousness are in our opinion best explained in terms of
the theoretical and psychometric designs of these two investment traits. Openness was
originally conceptualized as a multifarious trait construct, which entails not only intellectual
curiosity but also aesthetic awareness, heightened imagination or fantasy life, and receptivity
Intellectual Curiosity Predicts Academic Performance
18
to one‟s own inner feelings (McCrae, 1994). Griffin and Hesketh (2004) reported differential
validities of the facets of Openness for the prediction of job performance and suggested
distinguishing two factors of Openness: internal experience, including aesthetics, fantasy,
feelings, and external experience, spanning actions, ideas and values. It seems plausible that
internal experience is unrelated to effort and knowledge acquisition, whereas external
experience may capture conscientious behaviors that are elementary to transform actions and
ideas into reality. To that effect, the inclusion of an undifferentiated Openness construct (i.e.
at factor rather than facet level) in the current study may have blurred the association between
external experience and Conscientiousness, as indicated by negligible correlation coefficient.
This perspective is also consistent with findings from behavior genetics and brain imaging
studies (DeYoung et al., 2005; 2009; Wainwright et al., 2008), which suggested two distinct
factors of Openness (cf. Figure 1).
Conversely, TIE was designed to assess intelligence as typical behavior and
constitutes a precise measure of intellectual engagement in the pursuit of knowledge (e.g.,
Ackerman & Rolfhus, 1999). In the current study, TIE was used as a representative for
intellectual investment traits that a) are scattered across the literature, b) share conceptual
roots and even scale items, c) are alike in criterion validity, d) lack discriminant validity3, and
e) therefore might be used interchangeably (Mussel, 2010; von Stumm, 2010; Woo et al.,
2007). TIE was initially defined as “desire to engage and understand [the] world” and as
“need to know” (Goff & Ackerman, p. 539). As such, it refers to a consistent and purposeful
process of learning, which is without doubt also effortful. Accordingly, individuals, who seek
intellectual stimulation, present an increased level of persistence and zeal, which is reflected
in TIE‟s positive association with Conscientiousness (cf. Arteche, Chamorro-Premuzic,
Ackerman, & Furnham, 2009).
Our results ran counter the idea that effects of intelligence on academic performance
are in any way mediated by personality traits (Chamorro-Premuzic & Furnham, 2005; 2006;
Intellectual Curiosity Predicts Academic Performance
19
Moutafi et al., 2006), as all mediation models failed to achieve adequate model fit. Instead,
the data was best represented by a path model, in which intelligence, TIE and
Conscientiousness were direct, inter-correlated predictors of academic performance (Figure
3b). In this model, intelligence accounted for the greatest amount of variance; however, the
combined effects of curiosity and effort equaled the impact of intelligence on academic
performance. This model confirmed intelligence and effort as antecedents of academic
performance but added incremental validity by including intellectual curiosity. Therefore, the
current results supported that intellectual investment is a key determinant of academic
performance (Ackerman, 1996; Chamorro-Premuzic et al., 2006a, b; Goff & Ackerman,
1992).
The Hungry Mind: Vindicating Intellectual Curiosity
Pre-modern writers, including Aristotle (384 BC – 322 BC) and Cicero (106 BC – 43
BC), understood curiosity as “an intense, intrinsically motivated appetite for information”
(Loewenstein, 1994, p. 77). In a similar vein, the American psychologist and philosopher
John Dewey (1859 – 1952) stated:
The curious mind [is] constantly alert and exploring [and] seeking
material for thought, as a vigorous and healthy body is on the qui
vive for nutriment.[...] Such curiosity is the only sure guarantee of
acquisition of primary facts [...]. (Dewey, 1910, p. 31).
Dewey (1910) proposed a developmental perspective of curiosity, beginning with “an
abundant organic energy” (p. 31) that is associated with children‟s hunger to explore and
probe their surroundings. This basic experimentation is hardly intellectual but essential to
Intellectual Curiosity Predicts Academic Performance
20
later develop reflective reasoning (Dewey, 1910). In the second developmental stage, social
stimuli affect curiosity resulting in children‟s endless series of „why?‟ questions. Dewey
(1910) noted that this „why‟ is not aimed at a precise, scientific explanation but illustrates the
mastery of gathering and processing information, both of which constitute “the germ of
intellectual curiosity” (Dewey, 1910, p. 32). Finally, “curiosity raises above organic and
social planes [and] is transformed into interest in problems provoked by the observation of
things and the accumulation of material” and hence, becomes a “positive intellectual force”
(Dewey, 1910, p.32). Therefore, curiosity may start as a hungry and exploratory mind but
ultimately transforms into intellectual maturity.
Practical Implications
The association of intellectual curiosity with academic performance, has two
important practical implications for higher education. For one, academic performance may be
further enhanced if students‟ intellectual curiosity is continuously stimulated and nurtured.
Dewey (1910, p. 33) observed:
In a few people, intellectual curiosity is so insatiable that nothing
will discourage it, but in most its edge is easily dulled and blunted.
[...] Some lose it in indifference or carelessness; others in a frivolous
flippancy; many escape these evils only to become incased in a hard
dogmatism which is equally fatal to the spirit of wonder.
Schools and universities must early on encourage intellectual hunger and not
exclusively reward the acquiescent application of intelligence and effort (Charlton, 2009). It
is not only the diligent class winner who writes an excellent term paper but also the one who
asks annoyingly challenging questions during the seminar (a habit that is unfortunately not
Intellectual Curiosity Predicts Academic Performance
21
appreciated by all teachers). Also, intellectually stimulated students are likely to be more
satisfied with their university experience and to enjoy their studies to a greater extent than
students, who fell victim to Dewey‟s hard dogmatism. It is worth noting here that curiosity
may be as much a trait as a state (Berlyne, 1960; Loewenstein, 1994), suggesting that
educational settings should fully exploit their plentiful opportunities to induce and inspire
curiosity.
For the other, selection methods for university admissions and professional
recruitment should pay greater attention to intellectual curiosity as important indicator of
potential and ability. In fact, intellectual curiosity incorporates intelligence, zeal and the
hunger for information and novelty in one. To this effect, it seems imperative to expand
current research efforts in this field, and to investigate effects of intellectual curiosity on job
performance and cognitive development throughout lifespan. That said, this study, like most
personality research, relied on self - report measures, which only capture the explicit
(accessible by introspection) personality, but not implicit (inaccessible by introspection)
personality (James, 1998). Furthermore, self-report measures of personality are susceptible to
fakability (e.g., Furnham, 1986; Visweswaran & Ones, 1999); however, such distortions do
not affect criterion related validity (e.g., Hough, Eaton, Dunnette, Kamp, & McCloy, 1990;
Martin, Bowen, & Hunt, 2005). It seems unlikely that the current findings are merely a
consequence of faking-good or social desirability. However, future research on intellectual
investment must employ psychometric tests that are less susceptible to reporting bias, such as
observer ratings or conditional reasoning tests (cf. James, 1998).
Strengths and Limitations
The greatest strength of the current study is perhaps also its greatest weakness, namely
the fact that results are based on meta-analytic correlation coefficients. An extensive body of
Intellectual Curiosity Predicts Academic Performance
22
research was synthesized across a large number of studies and participants; however, the
current study design was equally constrained by the quality of the re-analyzed studies and
datasets. Furthermore, the current methodological approach limited the number of variables
that could be included; that is, other trait determinants of academic performance, such as self-
estimates of ability, fluid and crystallized intelligence and sex, were presently not included
because we failed to identify suitable meta-analyses that would have summarized the effects
of these factors across studies. Also, only Conscientiousness was included as a measure of
effort but not others, such as academic motivation, self-efficacy, achievement striving or
ambition. In a similar vein, the personality factor V from the Five Factor Model was
predominantly conceptualized in terms of Openness to Experience albeit other, related trait
designs - for example Goldberg‟s (1990) Intellect – may constitute more reliable constructs
(e.g., De Raad, 1994).
Applying path modeling to meta-analytic data is commonly associated with three
statistical challenges: determining the appropriate sample size to fit the model, recognizing
the sampling variation across studies, and analyzing a correlation rather than a covariance
matrix (Cheung & Chan, 2005). These factors all potentially bias model fit indices and
standard errors of parameters; thus, the current results are to be cautiously interpreted. Finally,
most studies included in the previous and present meta-analyses were single-wave and not
longitudinal, which makes causal inferences somewhat speculative. Specifically, academic
performance and personality are likely to be not only associated in a one-way direction but to
have reciprocal effects on one another; that is, achieving a high grade may increase the
probability of future conscientious and curious behaviors, as well as vice versa.
Despite these limitations, this study crucially advances the understanding of academic
performance. For one, it shows that variances in academic performance are best accounted for
by a combination of predictor variables (Chamorro-Premuzic & Furnham, 2006). For the
other, intellectual curiosity was demonstrated to constitute a meaningful addition to the
Intellectual Curiosity Predicts Academic Performance
23
traditional set of predictors of academic performance. In fact, Conscientiousness and
intellectual curiosity influenced academic performance to the same extent as intelligent.
However, our final model accounted only for a quarter of the variance in academic
performance; therefore, other variables, as for example age, sex, choice of subject, social class
of origin, learning style and self-confidence, are likely to be influential, too.
Conclusions
The current study suggests that traditional sets of predictors of academic performance,
notably general intelligence and Conscientiousness, should be accompanied by a third factor:
intellectual curiosity. Jensen (1998) stated that “[general intelligence] g acts only as a
threshold variable that specifies the essential minimum ability required for different kinds of
achievement. Other, non-g special abilities and talents, along with certain personality factors
[…], are also critical determinants of educational and vocational success” (p. 544-545). A
remarkable number of studies on determinants of academic achievement have focused
exclusively on ability and effort; the present findings, however, recommend further expanding
the „g-nexus‟ for a better understanding of individual differences in academic performance.
The latter requires – beyond intelligence and effort – a hungry mind.
Acknowledgements. We would like to thank Phillip Ackerman, Barbara Spellman,
Robert Sternberg, as well as three anonymous reviewers for their helpful comments on earlier
drafts of this manuscript.
Intellectual Curiosity Predicts Academic Performance
24
References
* marks studies included in the TIE meta-analysis
Ackerman, P. L. (1996). A theory of adult intellectual development: Process, personality,
interests, and knowledge. Intelligence, 22, 227-257.
Ackerman, P. L., & Heggestad, E. D. (1997). Intelligence, personality, and interests: Evidence
for overlapping traits. Psychological Bulletin, 121, 219-245.
*Ackerman, P. L., Kanfer, R., & Goff, M. (1995). Cognitive and noncognitive determinants
and consequences of complex skill acquisition. Journal of Experimental Psychology:
Applied, 1, 270-304.
Ackerman, P. L., & Rolfhus, E. L. (1999). The locus of adult intelligence: Knowledge,
abilities, and non-ability traits. Psychology and Aging, 14, 314-330.
Alexander, W. P. (1935). Intelligence, concrete and abstract: A study in differential traits.
British Journal of Psychology Monograph, 6 (19).
Arteche, A., Chamorro-Premuzic, T., Furnham, A., & Ackerman, P. (2009). Typical
intellectual engagement as a byproduct of Openness, learning approaches, and self-
assessed intelligence. Educational Psychology, 29, 357-367.
Barrick, M. R., & Mount, M. K. (1991). The Big Five personality dimensions and job
performance: A meta-analysis. Personnel Psychology, 44, 1-26.
Beretvas, S. N., & Furlow, C. F. (2006). Evaluation of an approximate method for
synthesizing covariance matrices for use in meta-analytic SEM. Structural Equation
Modeling: A Multidisciplinary Journal, 13, 153-185.
Berlyne, D. E. (1954). A theory of human curiosity. British Journal of Psychology, 45, 180–
191.
Berlyne, D. E. (1960). Conflict, arousal, and curiosity. New York: McGraw-Hill.
Intellectual Curiosity Predicts Academic Performance
25
Binet, A., & Simon, T. (1905). Methodes nouvelles pour le diagnostique du niveau
intellectual des anormaux [New methods for the diagnosis of the intellectual levels of
subnormals] (E. S. Kite, Trans.). In J. J. Jenkins & D. G. Paterson (Reprint Eds.),
Studies in individual differences: The search for intelligence (pp. 90-96). New York:
Appleton-Century-Crofts (Reprinted in 1961).
Bolton, T. L. (1892). The growth of memory in school children. American Journal of
Psychology, 4, 362-380.
Browne, M. W. & Cudeck, R. (1993). Alternative ways of assessing model fit. In Bollen, K.
A. & Long, J. S. (Eds.), Testing structural equation models, Newbury Park, CA: Sage,
136-162.
Cacioppo J. T., & Petty, R. E. (1982). The need for cognition Journal of Personality and
Social Psychology, 42, 116–131.
Cacioppo, J. T., Petty, R. E., Feinstein J. A., & Jarvis, B. G. (1996). Dispositional differences
in cognitive motivation: The life and times of individuals varying in need for
cognition. Psychological Bulletin, 119, 197–253.
Cacioppo, J. T., Petty, R. E., & Morris, K. (1983). Effects of need for cognition on message
evaluation, recall, and persuasion. Journal of Personality and Social Psychology, 45,
805-818.
Campell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the
multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105.
Carroll, J. B. (1993). Human cognitive abilities. A survey of factor-analytic studies.
Cambridge: Cambridge University Press.
Cattell, R. B. (1943). The measurement of adult intelligence. Psychological Bulletin, 40, 153-
193.
Cattell, R. B. (1971). Abilities: Their structure, growth, and action. Boston: Houghton
Mifflin.
Intellectual Curiosity Predicts Academic Performance
26
Chamorro-Premuzic, T., & Furnham, A. (2003a). Personality predicts academic performance:
evidence from two longitudinal studies on British University students. Journal of
Research in Personality, 37, 319-338.
Chamorro-Premuzic, T., & Furnham, A. (2003b). Personality traits and academic exam
performance. European Journal of Personality, 17, 237-250.
Chamorro-Premuzic, T., & Furnham, A. (2005). Personality and intellectual competence.
Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
Chamorro-Premuzic, T., & Furnham, A. (2006). Intellectual Competence and the intelligent
personality: A third way in differential psychology. Review of General Psychology,
10, 251-267.
Chamorro-Premuzic, T., & Furnham, A. (2010). The psychology of personnel selection. New
York: Cambridge University Press.
*Chamorro-Premuzic, T., & Furnham, A., Ackerman, P. (2006a). Ability and non-ability
correlates of general knowledge. Personality and Individual Differences, 41, 419-429.
*Chamorro-Premuzic, T., Furnham, A., Ackerman, P. (2006b). Incremental validity of typical
intellectual engagement as predictor of different academic performance measures.
Journal of Personality Assessment, 87, 261-268.
Charlton, B. G. (2009). Why are modern scientists so dull? How science selects for
perseverance and sociability at the expense of intelligence and creativity. Medical
Hypotheses, 72, 237-243.
Cheung, M. W. L., & Chan, W. (2005). Meta-analytic structural equation modeling: a two-
stages approach. Psychological Methods, 10, 40-64.
Collins, R. P., Litman, J. A., & Spielberger, C. D. (2004). The measurement of perceptual
curiosity. Personality and Individual Differences, 36, 1127–1141.
Intellectual Curiosity Predicts Academic Performance
27
Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO-PI-R)
and NEO Five-Factor Inventory (NEO-FFI): Professional manual. Odessa, FL:
Psychological Assessment Resources.
Cronbach, L. J. (1949). Essentials of psychological testing. New York, NY: Harper.
Deary, I. J., Strand, S., Smith, P., & Fernandes, C. (2007). Intelligence and Educational
Achievement. Intelligence, 35, 13-21.
*Dellenbach, M., & Zimprich, D. (2008). Typical intellectual engagement and cognition in
old age. Aging, Neuropsychology, and Cognition, 15, 208-231.
De Raad, B. (1994). An expedition in search of a fifth universal factor: key issues in the
lexical approach. European Journal of Personality, 8, 229-250.
DeYoung, C. G., Peterson, J. B., & Higgins, D. M. (2005). Sources of Openness/Intellect:
Cognitive and neuropsychological correlates of the fifth factor of personality. Journal
of Personality, 73, 825-858.
DeYoung, C. G., Quilty, L. C., & Peterson, J. B. (2007). Between facets and domains: 10
Aspects of the Big Five. Journal of Personality and Social Psychology, 93, 880-896.
DeYoung, C. G., Shamosh, N. A., Green, A. E., Braver, T. S., & Gray, J. R. (2009). Intellect
as distinct from Openness: Differences revealed by fMRI of working memory. Journal
of Personality and Social Psychology, 97, 883-892.
Dewey, J. (1910). How we think. Lexington, US: Heath.
Erez, A., Bloom, M. C., & Wells, M. T. (1996). Using random rather than fixed effects
models in meta-analysis: Implications for situational specifity and validity
generalization. Personnel Psychology, 49, 275-306.
Fancher, R. E. (1985). The intelligence men: Makers of the IQ controversy. New York, US:
W.W. Norton & Company.
Intellectual Curiosity Predicts Academic Performance
28
*Ferguson, E. (1999). A facet and factor analysis of typical intellectual engagement (TIE):
Associations with locus of control and the five-factor model of personality. Social
Behavior and Personality, 27, 545-561.
Fernández-Ballesteros, R., & Colom, R. (2004). The psychology of human intelligence in
Spain. In R. Sternberg (Ed.), International Handbook if Intelligence (pp. 79-103). New
York, US: Cambridge University Press.
Fiske, D. W., & Butler, J. M. (1963). The experimental conditions for measuring individual
differences. Educational and Psychological Measurement, 23, 249-266.
Fried, Y., Shirom, A., Gilboa, S., & Cooper, C. (2008). The mediating effects of job
satisfaction and propensity to leave on role stress-job performance relationships:
Combining meta-analysis and structural equation modeling. International Journal of
Stress Management, 15, 305-328.
Furnham, A. (1986). Response bias, social desirability and dissimulation. Personality and
Individual Differences, 7, 385-400.
*Furnham, A., Swami, V., Arteche, A., & Chamorro-Premuzic, T. (in press). Cognitive
ability, learning approaches and personality correlates of general knowledge.
Educational Psychology.
Galton, F. (1869). Hereditary genius. London, UK: Macmillan.
*Goff, M., & Ackerman, P. (1992). Personality-intelligence relations: Assessment of typical
intellectual engagement. Journal of Educational Psychology, 84, 537-552.
Goldberg, L. R. (1990). An alternative “Description of personality”: The Big-Five factor
structure. Journal of Personality and Social Psychology, 59, 1216-1229.
*Gow, A. J., Whiteman, M. C., Pattie, A. & Deary, I. J. (2005). The personality-intelligence
interface: insights from an ageing cohort. Personality and Individual Differences, 39,
751-761.
Intellectual Curiosity Predicts Academic Performance
29
Griffin, B., & Hesketh, B. (2004). Why Openness to Experience is not a Good Predictor of
Job Performance. International Journal of Selection and Assessment, 12, 243-251.
Hayes, K. J. (1962). Genes, drives, and intellect. Psychological Reports, 10, 299-342.
Hell, B., Trapmann, S., & Schuler, H. (2007). Eine Metaanalyse der Validität von
fachspezifischen Studierfähigkeitstests im deutschsprachigen Raum [A meta-analytic
investigation of subject-specific admission tests in German-speaking countries].
Empirische Pädagogik, 21, 251-270.
Heller, D., Watson, D., & Ilies, R. (2004). The role of person versus situation in life
satisfaction: A critical examination. Psychological Bulletin, 130, 574-600.
Hough, L. M., Eaton, N. K., Dunnette, M. D., Kamp, J. D., & McCloy, R. A. (1990).
Criterion-related validities of personality constructs and the effects of response
distortion on those validities [Monograph]. Journal of Applied Psychology, 75, 581-
585.
Hu, L., & Bentler, P. M. (1999). Cut-off criteria for fit indexes in covariance structure
analysis: Conventional criteria versus new alternatives. Structural Equation Modeling,
6, 1-55.
Hubin, D. R. (1988). The Scholastic Aptitude Test: Its development and introduction, 1900-
1947. Retrieved 10th of July, 2009, from the University of Oregon Web site:
http://darkwing.uoregon.edu/hubin/
Hunter, J. E., & Schmidt, F. L. (2000). Fixed effects vs. random effects meta-analysis models:
Implications for cumulative research knowledge. International Journal of Selection
and Assessment, 8, 275-292.
Hunter, J. E., & Schmidt, F. L. (2004). Methods of meta-analysis: Correcting error and bias
in research findings. New York: Academic Press.
Hunter, J. E, Schmidt, F. L., Jackson, G. B. (1982). Meta-analysis: cumulating research
findings across studies. Beverly Hills, CA: Sage Publications.
Intellectual Curiosity Predicts Academic Performance
30
James, L. R. (1998). Measurement of personality via conditional reasoning. Organizational
Research Methods, 1, 131-163.
Jensen, A. R. (1980). Uses of sibling data in educational and psychological research.
American Educational Research Journal, 17, 153-170.
Jensen, A. R. (1998). The g factor: The science of mental ability. Westport, CT: Praeger
Publishers.
Johnson, W., McGue, M., & Iacono, W. G. (2006). Genetic and environmental influences on
academic achievement trajectories during adolescence. Developmental Psychology,
42, 514-532.
Jöreskog, K. G. (1969). A general approach to confirmatory maximum likelihood factor
analysis. Psychometrika, 34, 183-202.
Judge, T., Jackson, C., Shaw, J., Scott, B., & Rich, B. (2007). Self-efficacy and work-related
performance: The integral role of individual differences. Journal of Applied
Psychology, 92, 107-127.
Klehe, U., & Anderson, N. (2007). Working hard and working smart: motivation and ability
during typical and maximum performance. Journal of Applied Psychology, 92, 978-
992.
Koenig, K. A., Frey, M. C., & Detterman, D. K. (2008). ACT and general cognitive ability.
Intelligence, 36, 153-160.
Kuncel, N. R., Hezlett, S. A., & Ones, D. S. (2004). Academic Performance, Career Potential,
Creativity, and Job Performance: Can One Construct Predict Them All? Journal of
Personality and Social Psychology, 86, 148-161.
Kyllonen, P. (1997). Smart testing. In R. Dillon (Ed.), Handbook on testing (pp. 347-368).
Westport: Greenwood Press/Greenwood Publishing Group, Inc.
Lehmann, N. (1999). The big test: The secret history of the American meritocracy. New
York: Farrar, Straus, and Giroux.
Intellectual Curiosity Predicts Academic Performance
31
Litman, J. A., & Spielberger, C. D. (2003). Measuring epistemic curiosity and its diversive
and specific components. Journal of Personality Assessment, 80, 75-86.
Loewenstein, G. (1994). The psychology of curiosity: a review and reinterpretation.
Psychological Bulletin, 116, 75–98.
MacKinnon, D. P., & Fairchild, A. J. (2009). Mediation analysis. Current Directions in
Psychological Science, 18, 16-20.
Martin, G. A. (2008). A brief history of state intervention in British schooling. In D.
Matheson (Ed.), An introduction to the study of education (pp. 206-239). Oxon, UK:
Routledge.
Martin, B. A., Bowen, C. C., & Hunt, S. T. (2002) How effective are people at faking on
occupational personality questionnaires? Personality and Individual Differences, 32,
247-256.
McCrae, R. R. (1994). Openness to experience: Expanding the boundaries of Factor V.
European Journal of Personality, 8, 251-272.
McDougall, W. (1933). The energies of man: A study of the fundamentals of dynamic
psychology. New York: Charles Scribner‟s Sons.
Mount, M. K., Barrick, M. R., Scullen, S. M., & Rounds, J. (2005). Higher-order dimensions
of the big five personality traits and the big six vocational interest types. Personnel
Psychology, 58, 447-478.
Moutafi, J., Furnham, A., & Crump, J. (2006). What facets of openness and conscientiousness
predict fluid intelligence score? Learning and Individual Differences, 16, 31-42.
Mussel, P. (2010). Epistemic curiosity and related constructs: Lacking evidence of
discriminant validity. Personality and Individual Differences, 49, 506-510.
O'Connor, M. C. & Paunonen, S. V. (2007). Big Five personality predictors of post-secondary
academic performance. Personality and Individual Differences, 43, 971-990.
Intellectual Curiosity Predicts Academic Performance
32
Poropat, A. (2009). A meta-analysis of the five-factor model of personality and academic
performance. Psychological Bulletin, 135, 322-338.
Raju, N. S., & Fleer, P. F. (2003). VG2M: A computer program for conducting validity
generalization analysis. Chicago: Illinois Institute of Technology.
Raju, N. S., Burke, M. J., Normand, J., & Langlois, G. M. (1991). A new meta-analytic
approach. Journal of Applied Psychology, 76, 432-446.
Rocklin, T. (1994). Relation between typical intellectual engagement and openness: Comment
on Goff & Ackerman. Journal of Educational Psychology, 86, 145-149.
*Rolfhus, E. L., & Ackerman, P. L. (1996). Self-report knowledge: At the crossroads of
ability, interest, and personality. Journal of Educational Psychology, 88, 174-188.
Sackett, P. R., Kuncel, N. R., Arneson, J., Cooper, S. R., & Waters, S. (2009). Socio-
economic status and the relationship between admissions tests and post-secondary
academic performance. Psychological Bulletin, 135, 1-22.
Salgado, J. F. (1997). The 5-Factor Model of Personality and Job-Performance in the
European-Community. Journal of Applied Psychology, 82, 30-43.
Saucier, G. (1992). Openness versus intellect: Much ado about nothing? European Journal of
Personality, 6, 381–386.
Schmidt, F.L., Oh, I.-S, & Hayes, T. (2009). Fixed vs. random models in meta-analysis:
Model properties and comparison of differences in results. British Journal of
Mathematical and Statistical Psychology, 62, 97 – 128.
Schulze, R. (2004). Meta-analysis: a comparison of approaches. Goettingen, GER: Hogrefe
& Huber.
Sharp, S. E. (1899). Individual psychology: A study in psychological method, American
Journal of Psychology, 10, 329-391.
Spearman, C. E. (1904). “General intelligence” objectively determined and measured.
American Journal of Psychology, 15, 201- 293.
Intellectual Curiosity Predicts Academic Performance
33
Spearman, C. E. (1927). The abilities of man. London: McMillan.
von Stumm, S. (2010). Intelligence, investment and intellect: Re-examining intelligence-
personality associations. Unpublished Doctoral Thesis, Goldsmiths University of
London.
Terman, L. M. (1916). The Measurement of Intelligence: An Explanation of and a Complete
Guide for the use of the Stanford Revision and Extension of the Binet-Simon
Intelligence Scale. Boston. Houghton Mifflin Co.
Tett, R. P., Jackson, D. N., & Rothstein, M. (1991). Personality Measures as Predictors of Job
Performance: A Meta Analytic Review. Personnel Psychology, 44, 703-742.
Trapmann, S., Hell, B., Hirn, J. O., & Schuler, H. (2007). Meta-analysis of the relationship
between the Big Five and academic success at university. Zeitschrift für
Psychologie/Journal of Psychology, 215, 132-151.
Verhaeghen, P., & Salthouse, T. A. (1997). Meta-analyses of age-cognition relations in
adulthood: Estimates of linear and nonlinear age effects and structural models.
Psychological Bulletin, 122, 231-249.
Viswesvaran, C., & Ones, D. S. (1995). Theory testing: Combining psychometric meta-
analysis and structural equations modeling. Personnel Psychology, 48, 865-885.
Viswesvaran, C., & Ones, D. S. (1999). Meta-analyses of fakability estimates: Implications
for personality measurement. Educational and Psychological Measurement, 59, 197-
210.
Viswesvaran, C., Ones, D. S., & Schmidt, F. L. (1996). Comparative analysis of the reliability
of job performance ratings. Journal of Applied Psychology, 81, 557-574.
Wainwright, M. A., Wright, M. J., Luciano, M., Geffen, G. M., & Martin, N. G. (2008).
Genetic covariation among facets of Openness to Experience and general cognitive
ability. Twin Research and Human Genetics, 11, 275-86.
Intellectual Curiosity Predicts Academic Performance
34
Webb, E. (1915). Character and Intelligence. Unpublished Doctoral Thesis, Goldsmiths
University of London.
*Wilhelm, O., Schulze, R., Schmiedek, F., & Süß, H. M. (2003). Interindividuelle
Unterschiede im typischen intellektuellen Engagement [Individual differences in
typical intellectual engagement]. Diagnostica, 49, 49-60.
*Woo, E. S., Harms, P. D., & Kuncel, N. R. (2007). Integrating personality and intelligence:
Typical intellectual engagement and need for cognition. Personality and Individual
differences, 43, 1635-1639.
Intellectual Curiosity Predicts Academic Performance
35
Table 1
Overview of studies included in the meta-analysis of TIE‟s associations with Openness,
intelligence, Conscientiousness, and academic performance.
Study
Criterion
Measure
N
r
Ackerman et al., 1995
C
NEO-PIR
93
.08
O
NEO-PIR
93
.70
Beier & Ackerman, 2001
O
NEO-FFI
153
.55
Chamorro-Premuzic et al. 2006a
g
Wonderlic
201
.36
C
NEO-PIR
201
.25
O
NEO-PIR
201
.26
Chamorro-Premuzic et al. 2006b
AP
Composite
104
.36
g
Composite
104
.27
C
NEO-FFI
104
.36
O
NEO-FFI
104
.24
Dellenbach & Zimprich, 2008
g
Raven‟s Matrices
364
.18
Ferguson, 1999
C
BAM
281
.20
O
BAM
281
.46
Furnham et al., 2009
g
Wonderlic
101
-.06
C
NEO-FFI
101
.28
O
NEO-FFI
101
.66
Goff & Ackerman, 1992
C
NEO-PI
138
.27
O
NEO-PI
138
.65
AP
GPA
138
.04
Gow et al., 2005
C
IPIP
460
.20
I
IPIP
460
.53
Intellectual Curiosity Predicts Academic Performance
36
g
Moray House Test
460
.13
Rolfhus, & Ackerman, 1996
C
NEO-FFI
203
.24
O
NEO-FFI
203
.67
Wilhelm et al., 2003
O
NEO-PIR
183
.50
AP
Humanities GPA
183
-.26a
AP
Science GPA
183
-.37a
Woo et al., 2007
C
BFI
81
.29
O
BFI
81
.65
a Wilhelm et al. (2003) tested German students; in the German marking system,
lower numbers indicate better grades. The coefficients were reversed for the analysis.
Note. The reported coefficient (r) values are not corrected for scale reliability. Criterion refers
to the variable that was investigated in a respective study in relation to TIE, while
measurement refers to the psychometric instrument that was used to assess the study‟s
variable of interest. Key: N = Sample size; r = Uncorrected study correlation; C =
Conscientiousness; O = Openness; g = general intelligence; AP = Academic performance;
GPA = Grade Point Average; NEO-PIR = NEO-Personality-Inventory-Revised; NEO-PI =
NEO-Personality Inventory; NEO-FFI= NEO Five Factor Inventory; BFI = Big Five
Inventory; BAM = Bipolar Adjective Markers; IPIP = International Personality Item Pool.
Intellectual Curiosity Predicts Academic Performance
37
Table 2
Meta-analytic coefficients of TIE with Conscientiousness, Openness to Experience, academic
performance and intelligence
Notes: N = Over-all sample size; k = Number of independent samples; M = Mean
observed correlation;
ˆ
= Sample size weighted and corrected validity;
2
ˆ
= Estimated
variance of ; SERE = Standard error of , random effects model; 95 % CIRE = Confidence
interval with p = .95, random effects model; 90 %; TIE = Typical Intellectual Engagement; C
= Conscientiousness; O = Openness; AP = Academic performance; ; g = general intelligence.
N
k
M
Mean est.
Reliability
ˆ
2
ˆ
SERE
95 % CIRE
TIE - C
1,662
9
.229
TIE: .870
C: .784
.277
.0
.022
[.233, .321]
TIE - O
1,998
11
.519
TIE: .870
O: .758
.639
.024
.050
[.542, .737]
TIE - AP
608
4
.260
TIE: .868
AP: .720
.328
.017
.080
[.171, .486]
TIE - g
1,230
5
.179
TIE: .864
g: .768
.224
.013
.061
[.104, .343]
Intellectual Curiosity Predicts Academic Performance
38
Table 3
Correlation matrix of meta-analytic coefficients from currently and previously conducted
studies
1
2
3
4
1
g
-
2
AP
.39
-
3
C
-.04
.24
-
4
O
.22
.07
.09
-
5
TIE
.22
.33
.28
.64
Note. Sample sizes range from N = 608 to N = 28,471.
Key: g = general intelligence; AP = Academic performance; C = Conscientiousness; O =
Openness; TIE = Typical Intellectual Engagement.
Intellectual Curiosity Predicts Academic Performance
39
Table 4
Model fit indices
Model
χ²
df
TLI
CFI
RMSEA
90%-CIRMSEA
0
Full inter-correlation
0
0
-
1.00
-
-
-
1
Full mediation model
561.96
2
-.661
.446
.345
.321
.369
2
Model 1 with correlation
345.58
1
-1.044
.659
.383
.349
.417
3
Partial mediation
216.38
1
-.278
.787
.302
.269
.337
4
Final Model
3.77
1
.984
.997
.034
.000
.074
Intellectual Curiosity Predicts Academic Performance
40
Figure 1
Typology of psychometric investment trait scales
Investment
Traits
Openness to
Experience
Intellect
Ideas
Values
Openness
Fantasy
Aesthetics
Feelings
Actions
Epistemic
Curiosity
Need for
Cognition
Typical
Intellectual
Engagement
Note. Only a small selection of investment trait scales is shown. For a full review of
existing investment trait scales and constructs see von Stumm (2010).
Intellectual Curiosity Predicts Academic Performance
41
Figure 2
Five different path models predicting academic performance
TIE
g
O
C
AP
TIE
g
C
AP
g
TIE
C
AP
g
TIE
C
AP
g
TIE
C
AP
Direct Predictor Models Mediation Models
Model 0
Model 1
Model 2
Model 3
Model 4
Note. Double-headed arrows represent correlations; single headed arrows imply direct causal
effects.
Key: O = Openness; TIE = Typical Intellectual Engagement; g = general intelligence; C =
Conscientiousness; AP = Academic performance.
Intellectual Curiosity Predicts Academic Performance
42
Figure 3
Results model of predictors of academic performance and their inter-relations.
O
C
g
TIE
AP
-.26
.37
.37
.18 ε
.23
.10
-.04
.64
.29
.22 TIE
g
C
AP
.20
.29 .20
.35
.23
(a) (b)
Note. (a) refers to the overall model including all variables as inter-correlated, direct
predictors, while (b) shows the final, best-fitting model.
Key: O = Openness; TIE = Typical Intellectual Engagement; g = general intelligence; C =
Conscientiousness; AP = Academic performance.
Intellectual Curiosity Predicts Academic Performance
43
APPENDIX
Herein, we will report the database procedures for identifying previous meta-analytic
reviews, our statistical approach to the TIE meta-analyses, and the prediction model of
academic performance.
1. Identifying meta-analytic coefficients from the previous literature
For associations of academic performance with Conscientiousness and Openness,
three meta-analyses were identified (i.e., O‟Connor & Paunonen, 2007; Poropat, 2009;
Trapmann, et al., 2007), all of which conceptualized Conscientiousness and Openness within
the framework of the Five Factor Model. Moreover, each used exam grades, essay marks and
grade point average (GPA) as indicators of academic performance but excluded academic
aptitude tests as outcome variable, such as the SAT (formerly Scholastic Aptitude Test) or the
American College entrance Test (ACT). However, the three meta-analyses were not
independent and differed considerably in their methodological approach. For the current
study, correlation coefficients between academic performance and Conscientiousness and
Openness were borrowed from Poropat (2009), who conducted the most comprehensive,
accurate meta-analysis on associations of personality and academic performance to date.
For the coefficient between intelligence and academic performance, Kuncel, Hezlett
and Ones (2004) summarized research of the Millers Analogies Test (MAT) and academic
performance and reported an estimated „true‟ score correlation of the MAT with reasoning
measures of .75 in a sample of N = 1,753 from 15 studies. In addition, Kuncel et al. (2004)
found the MAT to be closely related to verbal ability (.88; N = 3,614) and to math ability (.68;
N = 2,874). The MAT is composed of 100 analogies, which are considered to be excellent
markers of general intelligence (Carroll, 1993, p. 212; Spearman, 1927).
Intellectual Curiosity Predicts Academic Performance
44
Judge, Jackson, Shaw, Scott and Rich (2007)5 recently evaluated studies on
personality associations with general mental ability; their meta-analytic study included “valid
indicators of ability” (p.111), as well as measures of Conscientiousness and Openness within
the Five Factor taxonomy. From their study, we borrowed the intelligence-personality
coefficients of -.04 (N = 15,429) for Conscientiousness and .22 (N = 13,182) for Openness
with general intelligence, respectively. Finally, Mount and colleagues (2005) computed a full
inter-correlation matrix of the Five Factors re-evaluating scores from four standardization
samples with overall N = 4,000 and estimated the inter-correlation of Conscientiousness and
Openness at .09.
-----------------------------------
Insert Table I here
-----------------------------------
2. Methodological approach to TIE meta-analyses
Data were analyzed in line with the validation generalization approach (Raju, Burke,
Normand, & Langlois, 1991), which roots in the meta-analytic method of Hunter, Schmidt
and Jackson (1982). Raju and Fleer (2003) developed a software program for this purpose,
which was used in the present study to calculate meta-analytic coefficients under random-
effects (RE) conditions, which is suitable for the current research purpose (Erez, Bloom &
Wells, 1996; Hunter & Schmidt, 2000; Schmidt, Oh, & Hayes, 2009). Coefficients were
corrected for sampling error and attenuation by error of measurement in both predictors and
criteria. We used reliability coefficients from the primary studies, mostly the alpha coefficient
of internal consistency. In cases of missing reliability data, a weighted reliability estimate was
calculated based on the reliability-information given in the other studies.
5 Judge et al.‟s (2007) meta-analysis partially replicates Ackerman and Heggestad‟s (1997) earlier meta-analytic
results on intelligence-personality associations and is more comprehensive.
Intellectual Curiosity Predicts Academic Performance
45
3. Predictor models of Academic Performance
Following previous models of meta-analytically derived correlation matrix, estimates
of the true-score correlations were used for all matrix entries (e. g. Fried, Shirom, Gilboa, &
Cooper, 2008; Heller, Watson, & Ilies, 2004; Verhaeghen & Salthouse, 1997; Viswesvaran,
Ones, & Schmidt, 1996). Recently, Beretvas and Furlow (2006) inspected 26 studies that
applied SEM analyses to pooled correlation matrices not to covariance matrices. They
concluded (2006, p. 158) that “when the model of interest assesses relations between
psychological constructs (e. g., mathematical self-concept, motivation) that are measured
using different scales across studies, then correlations should be used instead of covariances
because the variances of different measures (of a single construct) will vary and thus so will
the associated covariances. The differences in the resulting covariances might not originate
only from the relation (i.e., correlation) between constructs but also from the scale of the
measures used to assess the construct”.
As cells or coefficients of the matrix differ in sample sizes, researchers have used a
variety of ad-hoc solutions to achieve an appropriate sample size, including the harmonic or
arithmetic mean, the median or the total of sample sizes (Cheung & Chan, 2005). Here, we
have opted for the harmonic mean (Nhamonic = 2,356) instead of the arithmetic mean (Narithmetic
= 8,383), which is recommended in the literature on unweigthed analysis of variance
(Viswesvaran & Ones, 1995).
Intellectual Curiosity Predicts Academic Performance
46
Table I (Appendix)
Coefficients borrowed from previous meta-analyses
Correlation
Source
N
k
M
Rho
C – AP
Poropat, 2009
32 887
92
-
.23 ab
O – AP
Poropat, 2009
28 471
77
-
.07 ab
O – C
Mount et al., 2005
4 000
4
-
.09 ab
g – AP
Kuncel et al., 2004
11 368
70
.27
.39 abc
g – C
Judge et al., 2007
15 429
56
-
-.04 ab
g – O
Judge et al., 2007
13 182
46
-
.22 ab
a Corrected for scale reliability of first variable.
b Corrected for reliability of second variable.
c Corrected for range restriction.
Key: C = Conscientiousness; AP = Academic performance; g = general intelligence; O =
Openness; TIE = Typical Intellectual Engagement; N = Overall sample size; k = number of
independent samples; M = Mean observed correlation; Rho = sample size weighted and
corrected validity.
