Journal of Economic Psychology 33 (2012) 1115–1128

Contents lists available at SciVerse ScienceDirect

Journal of Economic Psychology

journal homepage: www.elsevier.com/locate/joep

Review

Does expertise influence the impact of overconfidence on judgment,

valuation and investment decision?
Jérôme Lambert a,⇑, Véronique Bessière a,1, Gilles N’Goala b
a
MRM – Montpellier University, IAE de Montpellier, Place Eugène Bataillon, 34095 Montpellier Cedex 05, France
b
MRM – Montpellier University, ISEM Montpellier, Espace Richter – Bât. B – Rue Vendémiaire, CS 19519 34960 Montpellier Cedex 2, France

a r t i c l e i n f o a b s t r a c t

Article history: Empirical research documents that overconfidence has a strong impact on investment
Received 20 February 2012 decision. In this experimental study using a within-subject design and an asset allocation
Received in revised form 18 July 2012 problem, we detail this relationship by introducing a stage of judgment (initial knowledge
Accepted 20 July 2012
about the assets to invest in) and valuation (forecasts to be made) before the investment
Available online 28 July 2012
decision. We also examine the role of expertise by comparing a group of bankers (20 loan
officers) and a group of students (64), control in the role of risk aversion, and implement
JEL classification:
different measures of overconfidence (miscalibration in two formats – the BTA effect and
D03
D81
the illusion of control). Our results show that no differences were observed between bank-
G00 ers and students in the degree of overconfidence. However, overconfidence seems to deter-
mine decision-making in a different way across the two groups. Concerning students, we
PsycINFO classification: observed that overconfidence influenced general tasks such as global knowledge of the
2260 assets but when it came to investing, risk aversion had a major effect. In contrast, bankers
2340 were strongly influenced by their overconfidence. For them, it mainly affected specific
tasks (valuation and investment choices) but, surprisingly, risk aversion had no effect on
Keywords:
Overconfidence investment decision. Our results suggest that introducing an assessment stage in the deci-
Expertise sion process is an aid to understanding the differences between experts and novices.
Investment decision Ó 2012 Published by Elsevier B.V.
Loan officer
Banker
Risk aversion
Risk perception
Forecasting

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1116
2. Literature overview and theoretical design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1117
2.1. Overconfidence and financial decision-making . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1117
2.2. Research on experts and laymen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1117
2.3. Research on the relation between overconfidence and risk aversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1117

⇑ Corresponding author.
E-mail addresses: jeromelambert.sd@gmail.com (J. Lambert), veronique.bessiere@univ-montp2.fr (V. Bessière), gilles.ngoala@univ-montp1.fr
(G. N’Goala).
1
Address: Montpellier Research Management, Université Montpellier 2, Place Eugène Bataillon Bât. 19 – Case Courrier 028, Montpellier Cedex 5 34095,
France.

0167-4870/$ - see front matter Ó 2012 Published by Elsevier B.V.

http://dx.doi.org/10.1016/j.joep.2012.07.007
1116 J. Lambert et al. / Journal of Economic Psychology 33 (2012) 1115–1128

2.4. The model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1118

3. Experimental design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1118
3.1. Participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1118
3.2. Task and procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1119
3.3. Assets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1119
3.4. Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1119
3.4.1. Overconfidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1119
3.4.2. Risk aversion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1120
3.4.3. Judgment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1120
3.4.4. Valuation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1120
3.4.5. Investment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1121
4. Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1121
4.1. Overconfidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1121
4.2. Judgment, valuation and investment decision. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1121
4.3. The understanding of the investment decision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1121
4.4. The effects of overconfidence on judgment and valuation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1123
4.5. The role of overconfidence on the investment decision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1123
5. Discussion and conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1124
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1125
Appendix A. Survey items . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1125
Appendix B. PART 2: Asset valuation and investment decision. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1126
Appendix C. 1st stage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1126
Appendix D. 2nd stage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1126
Appendix E. 3rd stage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1127
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1127

1. Introduction

The human tendency to be overconfident has been widely documented in psychology and has become a central feature in
economics and behavioral finance. Many facets of overconfidence have been examined2 through the tendency to overestimate
our own knowledge (in particular in miscalibration studies), our abilities compared to others (better-than-average effect), or the
degree to which we control future events (illusion of control).
In parallel, numerous studies have confronted experts’ and novices’ decision-making process (Newell & Simon, 1972;
Shanteau, 1992a,b; Andersson, 2004). Bedard and Chi (1993) defined expertise as a number of years of experience or prac-
tice. It has been shown that experts seldom outperform novices but possess an efficient information-processing (Andersson,
2004; Camerer & Johnson, 1991; Chase & Simon, 1973; Newell & Simon, 1972). In this literature, loan officers have been
insufficiently studied (Andersson, 2004; Sarasvathy, Simon, & Lave, 1998; Von Holstein 1972).
In this paper, we analyze the impact of overconfidence along the dynamic process of decision-making in different stages
(judgment, valuation, decision). We considered an investment decision between different assets, which are differently
known, perceived, judged and evaluated by the participants. Our within-subject experiment also focused on the role of expe-
rience by analyzing two groups: 64 students (Bachelor’s and Master’s Degree levels) and 20 loan officers.
We divided the study into three stages: the judgment, the valuation and the investment decision. We supposed that over-
confidence could directly impact these stages, but also the investment decision through the judgment and the valuation pro-
cess. As a control variable for the investment decision, we included a measure of risk aversion.
The main contributions of our paper to the literature can be summarized as follows. Firstly, people can be overconfident in
different ways and this implies different types of measures. In this study, we use four measures. Two of them derive from prior
studies that have emphasized the sensitivity to elicitation methods in miscalibration tests, so we have integrated two types of
frame in the estimation of confidence (a probabilistic and a frequency judgment). The two others are designed to capture the
better-than-average effect and the illusion of control. We have also computed a composite index based on a factor analysis of
these four measures. The second goal is to study more closely how overconfidence influences the decision-making process, by
comparing a direct impact on investment choices and a mediated impact through a judgment/valuation stage of the assets in
which participants could invest. The third goal is to contribute to the literature on expertise, by examining the above relation-
ships between overconfidence and decision for professionals (bankers) and non-professionals (students).
In all, to sum up our main results, overconfidence can be clearly observed among experts and non-experts. However,
when we detail its effect on decision-making by including a judgment/valuation stage before the investment choice, we doc-
ument important differences.
For students, we observe that overconfidence influences general judgmental tasks such as global knowledge of the assets
but, when they have to invest in assets, risk aversion has a major effect. In contrast, bankers are strongly influenced by their

2
See in particular Hilton, Régner, Cabantous, Charalambides, and Vautier (2011), Moore and Healy (2008), Larrick, Burson, and Soll (2007), Klayman, Soll,
Gonzalez-Vallejo, and Barlas (1999), Alicke, Klotz, Breitenbecher, Yurak, and Vredenburg (1995).
 J. Lambert et al. / Journal of Economic Psychology 33 (2012) 1115–1128 1117

overconfidence. For them, overconfidence mainly affects specific tasks (valuation and investment choice) and risk aversion
has no effect on investment decision.
The rest of the paper is organized in four sections. In Section 2 we present an overview of the literature and our theoretical
model. Section 3 describes the design of the experiment. Section 4 presents the results. Section 5 discusses the results and
concludes our work.

2. Literature overview and theoretical design

2.1. Overconfidence and financial decision-making

The literature on overconfidence has broadly investigated its impact on financial decision-making, mainly focused on
financial markets and trader behavior. In this area, three major consequences of overconfidence have been documented:
overconfidence causes too much trade (Barber & Odean, 2000; Glaser & Weber, 2007; Kim & Nofsinger, 2002; Odean,
1999; Statman, Thorley, & Vorkink, 2004), excessive volatility (Daniel, Hirshleifer, & Subrahmanyam, 1998; Gervais & Odean,
2001) and a combined phenomenon of under-and overreaction to information (Daniel & Titman, 1999; Daniel et al., 1998;
Glaser & Weber, 2007; Lee & Swaminathan, 2000). In corporate finance, important impacts have been observed such as over-
investment or preference for debt financing (Malmendier & Tate, 2005).
Despite much research that has covered the impact of overconfidence on the financial decision involving various financial
agents, few studies have focused their attention on bankers. Sarasvathy et al. (1998) compared entrepreneurs with bankers
in a variety of problems including financial risk. Their results tend to demonstrate strong differences between the two
groups. Entrepreneurs seem to accept the risk whereas bankers try to reduce and control it.

2.2. Research on experts and laymen

In this paper, we focus our attention on the differentiation between laymen and experts. Such opposition will be studied
concerning the degree of overconfidence and decision-making. In past literature, differences have been observed between
these two populations both on their level of overconfidence and their performance ability.
In field experiments, a ‘‘process–performance paradox’’ has been cited by Camerer & Johnson (1991) whereby ‘‘individuals
with vast knowledge about a domain nevertheless are unable to render highly accurate predictions in that arena’’ (Önkal, Yates,
Simga-Mugan, & Öztin, 2003). In some domains such as stock prices and earnings forecasts, professionals do not outperform
laymen (Von Holstein, 1972; yates, McDaniel, & Brown, 1991) Önkal et al., 2003 confirmed these results by analyzing the
judgment accuracy between experts and novices in foreign exchange rates forecasting. They observed that many novices
made more accurate predictions than experts. However, Glaser and Weber (2007) found that experienced investors are more
accurate than novice investors when they estimate their past portfolio performance.
Overall, the analysis of the effect of expertise on decision-making has shown ambiguous results. Several studies docu-
ment the fact that experts are prone to overconfidence (for example Russo & Schoemaker, 1992). Glaser, Langer, and Weber
(2007) show that professional traders possess a higher degree of overconfidence than students but emphasize the fact that
‘‘the question of the strength of professionals’ biases compared to that of non-professionals is difficult to answer’’.

2.3. Research on the relation between overconfidence and risk aversion

Empirical and theoretical studies have analyzed the determinant of risk-taking or the impact of overconfidence in port-
folio choices. Numerous formats have been used to measure attitude against risk – the majority of studies conclude that
there is a negative relationship between overconfidence and risk aversion.

Risk
Aversion

MEDIATED EFFECT (relation 2)

Overconfidence Judgment/ Investment

(relation 3) Valuation (relation 4) Decision

DIRECT EFFECT (relation 1)

Fig. 1. Overconfidence and decision – direct and indirect (mediated) relationship.

1118 J. Lambert et al. / Journal of Economic Psychology 33 (2012) 1115–1128

Risk aversion has been expressed in different ways such as an allocation between a risky and a safe investment (Brockhas,
1980; MacCrimmon & Wehrung, 1990; Simon, Houghton, & Aquino 1999; Nosić & Weber, 2008) or such as herding behavior
(Diamond, 1991; Hirshleifer & Thakor, 1992; Menkhoff, Schmidt, & Brozynski, 2006).
Theoretical models such as that of Odean (1998) found that overconfident traders possess more risky portfolios than ra-
tional traders. This risk-taking behavior results from an undiversified portfolio. Similar results have been observed by De
Long, Shleifer, Summers, and Waldmann (1990), Benos (1998), Gervais and Odean (2001), Odean (1999); Barber & Odean
(2001) and Kim & Nosfinger(2002)). Moreover, these studies showed that overconfident traders had a lower performance
level than rational traders.
Menkhoff et al. (2006) provided a survey of 117 German fund managers and analyzed the relationship between experi-
ence, risk-taking and overconfidence. The authors found that overconfidence (defined as miscalibration) decreases with
experience as does risk-taking.
Despite these persuasive results, some other studies found no relations between overconfidence and attitude toward risk,
in particular Kirchler & Maciejovsky (2002), who measured risk attitude with the cash-equivalent method. Similar results
have been observed in empirical studies and theoretical models (Frascara, 1999; Heath & Tversky, 1991; Li, 2002; Prender-
gast & Stole, 1996).

2.4. The model

Our model supposes that overconfidence positively influences investment decision, in line with numerous studies. But
this influence could be direct or indirect through an additional stage devoted to judgment and valuation. Fig. 1 depicts both
influences.
Overconfident people can express overconfident judgment or valuation but do not necessarily act upon them when they
make a decision. In this case, the influence of overconfidence on decision would not be mediated by judgment. So we tested a
direct relationship between overconfidence and decision (relation 1 in Fig. 1) but also a relation which is mediated by a judg-
ment/valuation procedure (relation 2). We also integrated risk aversion as a control variable (and examined the relationship
between overconfidence and risk aversion), in line with the literature that has pointed out its role on investment decision.
According to the literature, risk aversion drives negatively and directly the level of investment in risky assets.
Moreover, we tested whether these relationships between overconfidence and decision are influenced by expertise.
According to the literature, experts are prone to overconfidence, just as people in general, and their overconfidence influ-
ences their decision-making. But its effect might not be so direct because they are used to acting through procedures, par-
ticularly for important decisions such as investment choices. They are used to integrating valuation processes in their
professional decision-making. Therefore, the impact of their overconfidence on decision-making could be much better med-
iated through a valuation stage compared with non-professionals.

3. Experimental design

3.1. Participants

Our within-subject experiment included two types of participants: a group of students and a group of bankers. The first
group was composed of 64 business students from Montpellier University (46 were male). Due to their sparse background in
Economics and Finance (39 students were in Bachelor’s Degree groups and 25 students were from a Master’s program spe-
cialized in New Technology Management), they were assimilated as novice loan officers (Andersson, 2004; Hoffman, Shad-
bolt, Burton, & Klein, 1995).

Part 1: Part 2: Asset study

Psychological & investment
traits
Judgment

Overconfidence
& Risk Aversion
Measures 3-stage
Evaluation decision
process

Investment
Decision

Fig. 2. Experimental process.

 J. Lambert et al. / Journal of Economic Psychology 33 (2012) 1115–1128 1119

The second group consisted of 20 loan officers from major French banks with a variety of experience (mean = 7.75; stan-
dard-deviation = 4.16). 11 loan officers were male.

3.2. Task and procedure

While the instructions were being explained, participants could ask questions. Afterwards they were asked to fill out the
survey (detailed in the Appendix A), which was divided into two parts. The experiment took 1 h to complete.
Due to the complexity of organizing a seminar session with all the bankers, we decided to elaborate the questionnaire on
a laptop and arrange individual sessions with each loan officer. We considered that an appointment to participate in the
experiment might evoke sufficient interest for the study so we did not offer any compensation (Andersson, 2004; Hodgkin-
son, Bown, Maule, & Glaister, 1999). Each question was similar for different bankers and took the same time to fill out. Soft-
ware was added to establish when the time had elapsed in order to manage their time equally and to be homogeneous with
the students’ settings.
For the students, we organized four sessions during the same morning. They had to fill out the same survey as that of the
bankers (few items were different due to the specificity of the bankers’ experience), with the same amount of time to answer.
Each student received a compensation of €15.
The different steps of the experiment are summarized in Fig. 2.
In the first part, the questionnaire measured overconfidence and risk aversion. Then, the second part was designed to de-
tail the stages preceding the decision: (1) a judgment stage with a general perception of each asset (initial risk perception
and knowledge), (2) a valuation stage with forecast estimations for each asset, and (3) the investment choice. These different
stages, namely the judgment/valuation stage before the investment stage, were designed to recreate the model described in
Fig. 1, and we also wanted to emulate circumstances from decision-making in real life.

3.3. Assets

Our experimental assets are based on 20 stocks listed on international markets. We defined two classes of assets: familiar
and unfamiliar stocks (consistent with Ganzach, 2000) with 10 stocks in each group. Within these two groups, we also se-
lected stocks with a high and a low degree of risk, along with 10 stocks in each group. Familiar stocks were chosen from well-
known companies (such as Rolls-Royce). Unfamiliar stocks were initially selected according to our judgment and we also
modified their names to assure their unfamiliarity (e.g. Cobham became Haboham). To determine the degree of actual risk,
stocks were selected according to their beta. We considered that a low-risk stock has a beta below 0.8, while a high-risk stock
has a beta above 1.2. Associated with the assets’ name and beta, we collected their previous 5 quarterly results, describing
the volatility linked to the beta.3 So, we were able to study whether the investment choice was driven by a general judgment
(in particular risk perception) conveyed by the name of the company, or by a valuation of risk derived from past volatility. As a
whole, we studied 20 assets: five were well-known with high-risk, five well-known with low-risk, five were unknown with high
risk, and five were unknown with low risk.

3.4. Measures

3.4.1. Overconfidence
We measured four different forms of overconfidence: miscalibration in interval estimates with probabilistic and fre-
quency judgment, better than average effect (BTA) and illusion of control. The first calibration measure, with a probabilistic
judgment, comprised 10 questions. Each participant provided a range of possible values for every question (i.e. an interval
estimate), which had only one correct numerical answer. Thereafter, he/she had to comment on his/her confidence as to the
question’s accuracy. To be precise, the participant had to indicate a subjective probability that his/her interval was correct.
The level of overconfidence was calculated by subtracting each participant’s actual proportion of correct answers from that
same person’s respective average of self-assessed confidence. Below 0, the participant was underconfident and above 0, the
participant was overconfident. The subjective probability methodology has been used in a number of other studies (Alpert &
Raiffa, 1982; Busenitz & Barney, 1997; Campbell, Goodie, & Foster, 2004; Fischhoff, 1977; Fischhoff, Slovic, & Lichtenstein,
1977; Gigerenzer, Hoffrage, & Kleinbölting, 1991; Juslin, Laukka, Liljeström, Västfjäll, & Lundqvist, 2011; Nelson & Narens,
1980; Teigen & Jorgensen, 2005) and has generally shown a lower level of overconfidence, compared to the objective prob-
ability methodology (in which a given probability – e.g. 90% – is assigned, see for instance Russo & Schoemaker, 1992).
The 10 questions in this test were selected in order to avoid too difficult or too easy questions because subjects typically
appeared to be overconfident for difficult questions (percentage of correct answers below approximately 75%), but under-
confident for easy questions (Lichtenstein & Fischhoff, 1977; Griffin & Tversky, 1992). These questions were elaborated after
pre-experimental face-to-face meetings with the bankers in which we interviewed them to better understand their domain-
specific issues and professional environment. For students, the questions were initially sampled by means of a preliminary
study with a test group.

3
We selected stocks for which past earnings volatility was consistent with their betas (high volatility associated with high beta).
1120 J. Lambert et al. / Journal of Economic Psychology 33 (2012) 1115–1128

Table 1
Overconfidence measures and risk aversion – Mean and SD.

N = 84 (20 bankers, 64 students) Bankers Students t-Test

Miscalibration (probability judgment) 19% (14%) 17% (19%) .450
Miscalibration (frequency judgment) 5% (16%) 6% (20%) .160
Better than average effect (BTA) 18% (21%) 17% (27%) .112
Illusion of control 26% (20%) 23% (32%) .462
Overconfidence index 16% (12%) 15% (16%) .483
Risk Aversion 28,500€ (18,469€) 35,689€ (25,467€) 1.156

None of the t-tests between bankers and students are significant. All the overconfidence variables are significantly different from zero, at least at the 1%
level.
In order to compute the overconfidence index, we converted the five-point scale into percentage for BTA and illusion of control (1 = 0%, 2 = 25%, 3 = 50%,
4 = 75%, 5 = 100%).
A mean above 0% shows overconfidence (with a maximum at 100%) and a mean below 0% shows underconfidence (with a maximum at 100%). At 0%, the
subject is neither over nor underconfident.

Following the 10 questions, participants were asked to estimate the number of correct answers. By comparing this number
with the actual number, we measured the second form of overconfidence: miscalibration by frequency judgment (Cosmides &
Tooby, 1996; Gigerenzer et al., 1991; Granhag, Strömwall, & Allwood, 2000; Sniezek & Buckley, 1991). The interpretation was
similar to the previous measure, but this form of confidence was designed to capture more retrospective confidence.
To measure the BTA effect, we replicated the well-known example of Svenson (1981) who found that 90% of drivers esti-
mated that they were above average in their driving ability. We added two questions to estimate the ability of participants
compared to the others. The average of a 5-item scale captured the BTA and was transformed into a percentage.
Our last form of overconfidence measured the illusion of control with three questions adapted from Simon et al. (1999)
and Langer and Roth (1975). We added two questions from Dumont, Schwarzer, and Jerusalem (2000) that focus on the con-
cept of self-efficacy (Bandura, 1977a). As well as the BTA variable, a 5-item scale measured the illusion of control and was
converted into a percentage.
All those four confidence measures combined general and professional questions (as detailed in the appendix). Our meth-
odology was globally designed to capture different forms of confidence. This multifaceted form of overconfidence has been
recently highlighted by Hilton et al. (2011) and Fellner & Krügel (2012).
Finally, to test overconfidence as a whole, we created an aggregated measure of these four variables. We weighed each
variable in an index by using its eigenvalue according to the methodology of Hayward and Hambrick (1997).4

3.4.2. Risk aversion

The subjects had to allocate an amount ranging from 0€ to 100,000€ between a risky and a safe investment. A participant
was considered as very risk averse when having a high amount in a safe investment. This measure replicated the Nosić and
Weber (2008) study and is also used by a major French bank to measure risk aversion.

3.4.3. Judgment
This stage took place before participants got any information about assets. Here we captured initial judgments about the
degree of knowledge and risk perception, solely based on the name of each asset. Each participant was required to state his/
her level of knowledge and risk perception. Both variables were on a 5-point scale ranging from 1 – zero knowledge (low degree
of risk) to 5 – well known assets (high degree of risk).5 The methodology of a single-item instrument to measure such variables
has been widely used (Ganzach, 2000; Payne, 1975; Slovic, 1967; Weber, Anderson, & Birnbaum, 1992; Weber & Milliman, 1997).

3.4.4. Valuation
After forming a judgment based on initial perceptions, participants were given information to produce forecasts. They had
to estimate the future result for each company on the basis of the five consecutive previous results. Then they were asked to
give an estimation of (1) the accuracy of their forecasts and (2) the difficulties they encountered in completing the forecasts
(Du & Budescu, 2007), ranked on a 5-point scale from low to high difficulty (or accuracy). We should notice that the accuracy
of the estimation was not required in order to compute performance but to improve each participant’s awareness in the dif-
ficulty of the estimation, which was our key variable here. Indeed, the core interest was to examine what type of risk per-
ception yields the decision: a risk perception based on initial and general knowledge (that is, judgment, at the previous
stage) or a risk perception based on a forecasting process and a valuation of past volatility (at this stage).

4
Hayward & Hambrick measure CEO overconfidence while explaining corporate acquisitions. They use three indicators to estimate CEO overconfidence and
derive a composite measure of overconfidence from a factor analysis of the three indicators. They then integrate this index in their empirical tests.
5
For each construct, we used numerical values (1, 2, 3, 4 and 5) to represent the scale points (from very low = 1 to very high = 5, from strongly disagree = 1 to
strongly agree = 5, etc.). The ‘distance’ between items ‘1’ and ‘2’ is likely to be the same as between items ‘3’ and ‘4’, etc. We also developed symmetric scales
which include middle points (neither agree - nor disagree = 3 for instance). For convention, when a Likert scale is symmetric and equidistant, the resulting data
can generally be assumed to be interval scaled.
 J. Lambert et al. / Journal of Economic Psychology 33 (2012) 1115–1128 1121

Table 2
Correlation matrix, standardized coefficient (prob.).

N = 84 1 2 3 4 5 6
1. Miscalibration – probability 1
2. Miscalibration – frequency .676 (.000) 1
3. Better than average .247 (.000) .174 (.000) 1
4. Illusion of control .264 (.000) .264 (.000) .291 (.000) 1
5. Overconfidence index .393 (.000) .250 (.000) .866 (.000) .607 (.000) 1
6. Risk aversion .222 (.000) .057 (.020) .136 (.000) .051 (.038) .200 (.000) 1

3.4.5. Investment
During the final stage participants had the possibility of investing in the 20 assets. Following the methodology of Weber,
Senmorgen, and weber (2005)we asked subjects to imagine they had won 100,000€ on the lottery and were committed to
investing this money. They had the opportunity to invest it (or part of it) in the assets they had selected from the 20, but
were informed that this placement was risky. The non-invested money was automatically lodged in a safe depository which
yielded an annual rate of 3%. We computed the sum of money invested in each asset to test our model.

4. Results

4.1. Overconfidence

Table 1 presents the degree of overconfidence among the four different measures of overconfidence and the composite
index. Consistent with the previous literature, our sample shows overconfidence whatever the measures used6 (Biais, Hilton,
Mazurier, & Pouget, 2005; Glaser & Weber, 2007; Teigen & Jorgensen, 2005; Nosić & Weber, 2008). However, we observed a
significant difference between the two kinds of miscalibration (t-test: 7.048⁄⁄⁄, p < 0.001) as mentioned by Gigerenzer et al.
(1991). No differences were observed between loan officers and students (t-tests are not significant), in contrast to Glaser
et al. (2007) who studied traders. A possible explanation could be that loan officers are less overconfident than traders. This
may be the case, according to the abundant literature on traders’ overconfidence, but our study is not designed to provide evi-
dence for such differences. This contradiction may also have risen from the task characteristics. In their experiment, Glaser et al.
measure overconfidence in a forecast context while we focused ours on the different aspects of overconfidence (miscalibration,
BTA and illusion of control). Traders are experts in trend estimates on financial markets and Glaser et al. based their measure of
overconfidence on this particular task.
In addition, both bankers and students were risk averse: they decided to invest very little money in the risky asset as
shown in Table 1.
In Table 2, from the total sample (N = 84), we documented strong correlations between the overconfidence index and its
four measures: Miscalibration – probabilistic judgment, Miscalibration – frequency judgment, Better than average effect
(BTA), and Illusion of control. All the Pearson correlations are significant at the 1% level. In addition, we observed a significant
correlation between risk aversion and overconfidence: r = 0.20 (p < 0.001). This negative coefficient means that strong over-
confidence is associated with a low risk aversion.

4.2. Judgment, valuation and investment decision

The judgment/valuation stage of the experiment is composed of three variables: knowledge and risk perception, based on
the name of each company (judgment), and the perceived difficulty of estimating forecasts (valuation). Investment is the
amount invested in each asset. In Table 3, we examined differences between loan officers and students. We noted that stu-
dents tended to invest more than bankers, whereas loan officers had less difficulty of forecasting than students (t-tests are
significant at 1% for these two variables solely).
Students and loan officers did not differ in initial judgments, i.e. through knowledge and risk perception (p > 0.05). Fur-
thermore, we found a negative correlation between asset knowledge and risk perception (r = 0.64, p < 0.01 – see Table 4).
This result is consistent with those of Sachse, Jungermann, and Belting (2012), who find that the higher the experience of
investors is, the lower the perceived investment risk. In the participants’ mind, high asset knowledge was often associated
with low risk. In their initial judgments, they strongly relied on the name of each company which seemed to enhance a sort
of halo effect – be it positive or negative. The two steps of the asset evaluation process were also associated: difficulty of
forecasting was significantly correlated with risk perception (0.234, p < 0.01) and asset knowledge ( 0.228, p < 0.01).

4.3. The understanding of the investment decision

We hereafter present the results of the determinants of the investment decision excluding overconfidence which will
be introduced below. For each participant and for each asset (that is for 1680 observations), we tested a linear causal

6
We also computed an equally weighted index for overconfidence (instead of using the percentage of explained variance). The results did not differ.
1122 J. Lambert et al. / Journal of Economic Psychology 33 (2012) 1115–1128

Table 3
Judgment, valuation and investment decision variable – Mean and SD.

N = 84 Bankers (N = 20) Students (N = 64) t-Test

Knowledge 2.29 (1.44) 2.37 (1.57) .956
Risk perception 2.88 (1.22) 2.98 (1.31) 1.330
Difficulty of forecasting 2.69 (1.25) 2.91 (1.09) 3.467***
Investment per agent 28,940€ (23,901€) 57,620€ (28,838€) 3.336***

Standard-deviations are reported in parentheses.

Asterisk indicate significance level of this test:
***
1%.

Table 4
Correlation matrix, standardized coefficient (prob.).

N = 84 1 2 3 4 5 6
1 Overconfidence index 1
2 Risk aversion .200 (.000) 1
3 Asset knowledge .075 (.002) .034 (.158) 1
4 Forecasting difficulty .118 (.000) .006 (.803) .228 (.000) 1
5 Risk perception .040 (.104) .003 (.898) .645 (.000) .234 (.000)
6 Investment decision .020 (.402) .066 (.007) .324 (.000) .187 (.000) .246 (.000) 1

Table 5
Impact of the judgment/valuation and risk aversion on the investment.

Regression analysis N = 1680 Bankers (N = 400) Beta Students (N = 1280) Moderating role of expertise – Bankers vs.
(Participants  Assets) (prob.) Beta (prob.) students (prob.)
Knowledge .40 (.000) .25 (.000) .011
Risk perception .08 (.169) .05 (.126) .393
Difficulty of forecasting .12 (.010) .13 (.000) .091
Risk aversion .013 (.779) .056 (.032) .246

We report the significance of the t value in parentheses. The significance of this test is at 5%.

relationship in which four variables (knowledge, risk perception, and difficulty of forecasting, plus risk aversion as a control)
were presumed to directly influence investment decision. To assess multicollinearity in the regression analysis, we looked at
the tolerance (i.e. percentage of variance in the predictor that cannot be accounted for by the other predictors) and the var-
iance inflation factor (VIF): for each construct, tolerance values were much larger than .10 (between .57 and .99) and VIF
were much smaller than 10 (between 1.00 and 1.74). A moderator variable, the level of expertise (bankers vs. students),
was seen as a variable that altered the strength of these causal relationships. So, we simultaneously measured the four causal
relationships for two levels of expertise (bankers vs. students). In the regression analysis, we were interested in making ex-
plicit comparisons of one level against another (bankers vs. students).7 The results are shown in Table 5.
Both groups seemed very sensitive to the assessment stage in their investment choices of two measures – that is – for the
‘‘knowledge’’ component (based on the name of the company) and the ‘‘difficulty of forecasting’’. But the initial ‘‘risk percep-
tion’’ (also based on the name) had no effect on investment decision for both populations, probably because the assessment
stage implied a more reliable way of evaluating risk via the difficulty of forecasting. Both groups invested a higher amount in
the assets which they considered having good knowledge of (positive coefficients) and low difficulty in estimating forecasts
(negative coefficients).
We tested the significance of expertise by regressing investment on each dependent variable (4 variables in Table 5) and
by introducing expertise as a dummy (bankers vs. students) moderator variable. The expertise had an impact on the relation-
ship between the knowledge and the investment decision. The effect was significantly more important for bankers (b = 0.40)
than for students (b = 0.25). However, we did not observe any moderating effect for the other independent variables (rela-
tionships between ‘‘difficulty of forecasting’’ and investment and between ‘‘risk perception’’ and investment). Even if we did
not observe a moderating effect between the ‘‘difficulty of forecasting’’ and investment, the significance at 9% allowed us to
establish a slight difference between bankers and students. We also documented no effect of risk aversion for loan officers,
although they are professionals for whom risk is expected to be a main decision driver. Risk aversion had a significant impact

7
We also checked on whether our results held when controls were made for within-subject (84 participants) and within-asset effects (20 assets). In our
regression analysis, we integrated control variables (name of the participants and name of the assets) as random factors, whereas the level of expertise (bankers
vs. students) was seen as a fixed factor. What interested us was to what extent the random factors (participants and assets) accounted for variance in the
dependent variable and, thus, could alter other effects. In fact, our results were similar.
 J. Lambert et al. / Journal of Economic Psychology 33 (2012) 1115–1128 1123

Table 6
The effects of overconfidence on judgment and valuation.

Regression analysis N = 1680 Bankers (N = 400) Students (N = 1280 Moderating role of

(Participants  assets) Beta (prob.) Beta (prob.)) expertise – Bankers vs.
students (prob.)
Knowledge .024 (.638) .085 (.002) .279
Risk perception .21 (.014) .024 (.394) .032
Difficulty of forecasting .12 (.000) .10 (.000) .000

We report the significance of the t value in parentheses. The significance of this test is at 5%.

on student investment. However, the effect of risk aversion on investment did not significantly differ between bankers and
students (p > .05).

4.4. The effects of overconfidence on judgment and valuation

In this section we present the results for relation 3 in Fig. 1. To test the implication of overconfidence on judgment and
valuation and to analyze the role of expertise, we used an index of overconfidence based on similar items between bankers
and students. The results are presented in Table 6. They highlight the fact that overconfidence has an impact on each variable
but differs depending on the population.
For the bankers, overconfidence had a negative impact on risk perception whereas for the students no relation was ob-
served. In contrast, the subjective knowledge of the assets was driven by the level of overconfidence only for students. The
difficulty of forecasting was influenced by the level of overconfidence and by the level of expertise. In relation 4, a similar
impact of the difficulty of forecasting on investment had been documented but here we observed that the impact of this var-
iable was stronger for bankers than for students (respectively b = .12 and b = .10).
When taken together, relations 3 and 4 showed that the global decision process was partially similar among both experts
and novices, but also provided evidence on differences between them. For both, the difficulty of forecasting was a main driver
of the investment decision, meaning that they were both sensitive to a valuation stage. But overconfidence made this valu-
ation stronger for bankers (in Table 6 the ‘‘difficulty of forecasting’’ variable has the strongest coefficient for them and the im-
pact of overconfidence on this variable is significantly stronger than for students). For both, the initial knowledge of the
company was also a main driver of the investment decision, meaning that they were also very sensitive to an initial judgment
(without any valuation process) in their decision, even if overconfidence drove this perception for students solely (Table 6).

4.5. The role of overconfidence on the investment decision

We tested relations 1 and 2 in Fig. 1. In a primary analysis (relation 1), we found that overconfidence had a direct impact
on the investment decision for bankers solely (b = .165⁄⁄⁄). For them, the amount invested in assets was positively correlated
with the degree of overconfidence, consistent with Odean (1998) and Glaser and Weber (2007). No such relationship was
observed among students.
To further investigate the relationship between overconfidence and investment, we introduced the judgment and valu-
ation variables as mediators (knowledge, risk perception and difficulty of forecasting), following the methodology of Baron
and Kenny (1986). After the introduction of mediators, the effect of overconfidence on the level of investment remained sig-
nificant for bankers only, but lower (b = .128⁄⁄) than observed in the direct relationship. Bankers appeared to cope with their
risk attitude better than they handled their overconfidence.
On the contrary, students selected the assets according to their risk aversion and this strongly drove the investment deci-
sion, regardless of their overconfidence. Overconfident students seemed to judge according to their overconfidence but did
not directly invest on the basis of this assessment – they seemed rather to invest in accordance with their risk aversion. For
them, overconfidence had no direct or mediated effect on investment because the decision to invest was dominated by risk
aversion. Overconfidence had only an indirect effect as noted above: it positively influenced knowledge whereas knowledge
positively affected investment decision; and overconfidence negatively influenced difficulty of forecasting while difficulty of
forecasting had a negative impact on investment decision.
Our findings tend to complete our initial model and our results are given in Figs. 3 and 4.
In the student sample, we were not able to demonstrate any mediating effect of knowledge and difficulty of forecasting
since there was no previously significant relation8 (relation 1) between the independent (overconfidence) and dependent var-
iable (investment decision). In contrast, in the bankers’ sample, there was a direct effect of overconfidence on investment deci-
sion. However, the mediation effect – only possible through difficulty of forecasting – was just partial: when relations 3 and 4
were controlled following Baron and Kenny (1986), the previously significant relation between overconfidence and investment

8
For Baron and Kenny (1986), ‘‘a variable function is a mediator when it meets the following conditions: (a) variations in levels of the independent variable
significantly account for variations in the presumed mediator (i.e., Path a), (b) variations in the mediator significantly account for variations in the dependent variable
(i.e., Path b), and (c) when Paths a and b are controlled, a previously significant relation between the independent and dependent variables is no longer significant, with
the strongest demonstration of mediation occurring when Path c is zero.’’
1124 J. Lambert et al. / Journal of Economic Psychology 33 (2012) 1115–1128

Forecasting Difficulty Risk Aversion

-.13
-.10
-.056

.085 .25
Overconfidence Asset Knowledge Investment Decision

ns

ns ns
Risk Perception

Fig. 3. Students’ model of investment decision (beta, p < 0.05).

Forecasting Difficulty Risk Aversion

-.12
-.12
ns

ns .40
Overconfidence Asset Knowledge Investment Decision

.128

-.21 ns
Risk Perception

Fig. 4. Bankers’ model of investment decision (beta, p < 0.05).

decision remained highly significant (p < 0.01) but the t value decreased slightly (3.34 vs. 2.88). In other words, the bankers’ dif-
ficulty of forecasting partially – and not totally – mediated the influence of overconfidence on investment decision.

5. Discussion and conclusion

One important result in this study is that the degree of overconfidence is similar between experts and novices, whatever
the measures. This result is in opposition to Glaser et al. (2007) who document the fact that traders are more overconfident
than students, but we did not reproduce the Glaser et al. experiments and measures. These variations across studies devoted
to overconfidence have been recently highlighted by Fellner & Krügel, 2012 who concluded: ‘‘only further empirical research
can settle speculations about why different methods to assess overconfidence create divergent findings’’. (p.152)
The most important result that we have documented is that the two groups are not similarly influenced by their overcon-
fidence. Three major differences have emerged from the experiment, regarding (1) the mediator effect of an assessment stage
before making the decision, (2) the distinction between general judgment and specific valuation and (3) the role of risk
aversion.
Our intuition had been that professionals are used to implementing procedures before making a decision (in particular for
important decisions such as investment decisions), and in fact we observed, only for bankers and not for students, a signif-
icant mediated relationship between overconfidence and decision (relation 2) consistent with our intuition.
In addition, when we closely look into how overconfidence influences the three components of judgment and valuation
(relation 3), and how those components constitute a decision (relation 4), we can observe that bankers are more influenced
by valuation, and students by initial and general judgment. In fact, we have documented that which concerns students only,
a significant combined relation 3 and 4, which indicates that initial judgment (‘‘knowledge’’ based on the name of the com-
pany) is important in their decision process. Moreover, when analyzing the difficulty of forecasting in the same manner, we
 J. Lambert et al. / Journal of Economic Psychology 33 (2012) 1115–1128 1125

have shown that both groups are significantly influenced (relations 3 and 4 combined), but this impact is stronger for bank-
ers (as shown by the introduction of expertise as a moderator variable in the analysis).
This analysis is consistent with the literature which has pointed out that experts are prone to overconfidence in their
decision-making (in particular traders on financial markets). Our experiment reinforces this result here by showing that
overconfidence has a powerful impact on the way experts implement valuation. This result can also be interpreted with re-
spect to the literature which established a positive link between overconfidence and task difficulty (Lichtenstein & Fischhoff,
1977; Griffin & Tversky, 1992). Another important conclusion is that experts act upon their judgments: their investment
choices are related to their valuation.
Finally, we have noted the fact that non-experts are strongly influenced by their risk aversion, whereas bankers are not.
This is not due to a difference in the degree of risk aversion (no differences are observed). Students seemed to be overcon-
fident in their initial judgment when they perceived the name of the company, and also when they evaluated the difficulty of
forecasting. But when it came to decision-making they did not rely on their opinion concerning the assets, whatever their
degree of overconfidence. Their investment choices were only driven by their risk aversion. This finding is in opposition
to previous results which document a direct link between overconfidence and decision in experimental settings involving
students (such as Glaser & Weber, 2007; Odean, 1998; or Maciejovsky & Kirchler, 2002). It appears that the introduction
of an additional stage in the experiment, namely a valuation process, might question their initial judgment (strongly deter-
mined by their degree of overconfidence). Their initial judgment would be then put in doubt, and this ‘‘think twice’’ in invest-
ment decision seems to be guided by risk aversion.
In conclusion, our results suggest that introducing an assessment stage in the decision process helps in the understanding
of the differences between experts and novices. The use of professionals as participants in our experiment guaranteed some
interesting findings. However, experts are not always accessible and are sometimes reluctant to answer long questionnaires.
In our experiment, only 20 professionals participated (as compared with 64 students) and provided us with 400 investment
decisions (1280 from students). This resulted in unequal sample sizes. Further research is therefore needed to overcome this
limitation and to reinforce our general conclusions. Additional studies should investigate the stability of these findings over
different samples, different types of experts along with different tasks and methods for measuring overconfidence.

Acknowledgements

The authors gratefully acknowledge the comments by Denis Hilton and two anonymous reviewers. The authors also
thank Yoav Ganzach and Laurent Villanova, and the participants at MRM research seminar for their helpful comments on
earlier versions of this paper. An earlier version of the paper was presented at the SPUDM Conference in 2011.

Appendix A. Survey items

PART 1: Overconfidence and risk aversion measurement

(bankers’ version – some student questions were different)

Miscalibration (probability judgment)

For each of the following questions, provide a low and high estimate and the percentage of confidence.
How many deputies are there in the French Parliament?
How many failures took place in France in 2009?
What was the inflation rate in Europe in 2008 (percentage)?
How many countries are in the OECD?
How many Olympic records were beaten in Beijing?
How many wage-earners worked in the banking sector in France in 2008?
2/3 of corporate loans are under a certain amount, which one (in Euros)?
How many banks were listed on Euronext on September 1st 2009?
What was the highest value of oil (light sweet crude)?
In 2008, what was the trade balance in France (in billions)?
Miscalibration (frequency judgment): According to you, how many answers are correct?
Above average effect
You will rate the following questions from 1 (below the average) to 5 (above the average):
How do you rate your driving skills?
In comparison to your colleagues, how do you estimate your risk analysis?
In comparison to your friends, how do you estimate your general culture?
Illusion of control
Score the following from 1 (do not agree) to 5 (fully agree):
I think I can anticipate a customer’s risk

(continued on next page)

1126 J. Lambert et al. / Journal of Economic Psychology 33 (2012) 1115–1128

I can anticipate fraud from a customer

My valuations are fully accurate
It is easy for me to be focused on my objectives and to reach my goals
Whatever happens, I can handle it
Risk aversion
Imagine you just won €100,000 on the lottery. You receive the following financial offer from a reputable bank:
Investment A: It is equally possible that you could lose half of the amount invested and double the money invested
Investment BÒ: Your investment will yield 3%/year (guaranteed)
What is the amount you would invest in investment A?

Appendix B. PART 2: Asset valuation and investment decision

In this part you will be confronted with an asset valuation. It is important that your answers reflect your real behavior.
Please answer all questions.

Appendix C. 1st stage

We present here a list of stocks in which you could invest. State your level of knowledge and your risk perception for
EACH asset.
Knowledge: 1 (no knowledge) ? 5 (very good knowledge).
Risk perception: 1 (very low) ? 5 (very high).

Assets
AZR. Ltd. Coustco Sale Corp. Microsoft Corporation
Adobe Systems Incorporated Danone Paccar Inc.
Abron Products Inc. Deutsche Telekom N Rolls-Royce
Bouygues Galedia Inc. Siemens
Haboham Hursarco Corp. Midathern Co.
Coca Cola Co. Heed Matin CP Woyerhaser Co.
Colgate Logitech International S.A

Appendix D. 2nd stage

We present here the last 5 quarterly results. You will estimate the future result and the difficulty you had in reaching this
estimation.
Difficulty of the estimation: 1 (very easy) ? 5 (very difficult).

30/06/08 30/09/08 31/12/08 31/03/09 30/06/09

AZR. Ltd. 1080 1091 494 657 875
Adobe Systems Incorporated 199 486 199 161 287
Abron Products Inc. 134 87 291 197 354
Bouygues 1066 1076 842 502 984
Haboham 158 137 113 107 137
Coca Cola Co. 1864 1790 1954 1836 2012
Colgate 540 610 610 673 697
Coustco Sale Corp. 431 466 428 383 465
Danone 1254 1267 1273 1290 1287
Deutsche Telekom N 4947 4852 4990 4980 4835
Galedia Inc. 409 521 545 509 479
Hursarco Corp. 149 3665 112 223 653
Heed Matin CP 1015 987 979 944 951
Logitech International S.A 146 302 398 89 33
Microsoft Corporation 4796 4751 4820 4762 4757
Paccar Inc. 525 577 380 290 196
Rolls-Royce 676 1567 2200 2590 3089
 J. Lambert et al. / Journal of Economic Psychology 33 (2012) 1115–1128 1127

2nd stage (continued)

30/06/08 30/09/08 31/12/08 31/03/09 30/06/09
Siemens 3126 2960 2642 2197 2503
Midathern Co. 811 756 614 748 699
Woyerhaser Co. 66 107 885 155 58

Appendix E. 3rd stage

You have just won €100,000 on the lottery and have the possibility (not the obligation) of investing the amount you wish
in these stocks. The stock returns are uncertain. The rest of the amount will be invested in a guaranteed fund (3%/year).

References

Alicke, M. D., Klotz, M. L., Breitenbecher, D. L., Yurak, T. J., & Vredenburg, D. S. (1995). Personal contact, individuation, and the better than average effect.
Journal of Personality and Social Psychology, 68, 804–825.
Alpert, M., & Raiffa, H. (1982). A progress report on the training of probability assessors. In D. Kahneman, P. Slovic, & A. Tversky (Eds.), Judgment under
uncertainty: Heuristics and biases. Cambridge university press.
Andersson, P. (2004). Does experience matter in lending? A process-tracing study on experienced loan officers’ and novices’ decision behavior. Journal of
Economic Psychology, 25, 471–492.
Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice Hall.
Barber, B. M., & Odean, T. (2000). Trading is hazardous to your wealth: The common stock investment performance of individual investors. Journal of Finance,
55, 773–806.
Barber, B. M., & Odean, T. (2001). Boys will be boys: Gender, overconfidence, and common stock investment. Quarterly Journal of Economics, 116, 261–292.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical
considerations. Journal of Personality and Social Psychology, 51, 1173–1182.
Bedard, J., & Chi, M. T. H. (1993). Expertise in auditing. Auditing: A Journal of Practice and Theory, 12, 21–45.
Benos, A. V. (1998). Aggressiveness and survival of overconfident traders. Journal of Financial Markets, 1, 353–383.
Biais, B., Hilton, D., Mazurier, K., & Pouget, S. (2005). Judgmental overconfidence, self-monitoring, and trading performance in an experimental financial
market. Review of Economic Studies, 72, 287–312.
Brockhaus, R. H. (1980). Risk-taking propensity of entrepreneurs. Academy of Management Journal, 23, 509–520.
Busenitz, L., & Barney, J. (1997). Differences between entrepreneurs and managers in large organizations: Biases and heuristics in strategic decision making.
Journal of Business Venturing, 12, 9–30.
Camerer, C. F., & Johnson, E. J. (1991). The process-performance paradox in expert judgment: How can experts know so much and predict so badly? In K. A.
Ericsson & J. Smith (Eds.), Toward a general theory of expertise: Prospects and limits (pp. 195–217). New York: Cambridge University Press.
Campbell, W. K., Goodie, A. S., & Foster, J. D. (2004). Narcissism, confidence, and risk attitude. Journal of Behavioral Decision Making, 17, 481–502.
Chase, W. G., & Simon, H. A. (1973). The mind’s eye in chess. In W. G. Chase (Ed.), Visual Information Processing (pp. 215–281). New York: Academic Press.
Cosmides, L., & Tooby, J. (1996). Are humans good intuitive statisticians after all? Rethinking some conclusions from the literature on judgment under
uncertainty. Cognition, 58, 1–73.
Daniel, K., & Titman, S. (1999). Market efficiency in an irrational world. Financial Analysts Journal, 55, 28–40.
Daniel, K., Hirshleifer, D., & Subrahmanyam, A. (1998). Overconfidence, arbitrage, and equilibrium asset pricing. Journal of Finance, 53, 921–965.
De Long, J. B., Shleifer, A., Summers, L., & Waldmann, R. J. (1990). Positive feedback investment strategies and destabilizing speculation. Journal of Finance, 45,
379–395.
Diamond, D. (1991). Reputation acquisition in debt markets. Journal of Political Economy, 97, 828–862.
Du, N., & Budescu, D. V. (2007). Does past volatility affect investors’ price forecasts and confidence judgments? International Journal of Forecasting, 23,
497–511.
Dumont, M., Schwarzer, R., & Jerusalem, M. (2000). French adaptation of the general self-efficacy scale. Berlin University, <http://www.userpage.fu-
berlin.de/~health/french.htm>.
Fellner, G., & Krügel, S. (2012). Judgmental overconfidence. Three measures, one bias? Journal of Economic Psychology, 33, 142–154.
Fischhoff, B. (1977). Perceived informativeness of facts. Journal of Experimental Psychology: Human Perception and Performance, 3, 349–358.
Fischhoff, B., Slovic, P., & Lichtenstein, S. (1977). Knowing with certainty: The appropriateness of extreme confidence. Journal of Experimental Psychology:
Human Perception and Performance, 3, 552–564.
Frascara, J. (1999). Cognition, emotion and other inescapable dimensions of human experience. Visible Language, 33, 74–87.
Ganzach, Y. (2000). Judging risk and return of financial assets. Organizational Behavior and Human Decision Processes, 83, 353–370.
Gervais, S., & Odean, T. (2001). Learning to be overconfident. Review of Financial Studies, 14, 1–27.
Gigerenzer, G., Hoffrage, U., & Kleinbölting, H. (1991). Probabilistic mental models: A Brunswikian theory of confidence. Psychological Review, 98, 506–528.
Glaser, M., & Weber, M. (2007). Overconfidence and trading volume. Geneva Risk Insurance Review, 32, 1–36.
Glaser, M., Langer, T., & Weber, M. (2007). On the trend recognition and forecasting ability of professional traders. Decision Analysis, 4, 176–193.
Granhag, P. A., Strömwall, L. A., & Allwood, C. M. (2000). Effects of reiteration, memory, and feedback on eyewitnesses’ confidence. Applied Cognitive
Psychology, 14, 397–420.
Griffin, D., & Tversky, A. (1992). The weighing of evidence and the determinants of overconfidence. Cognitive Psychology, 24, 411–435.
Hayward, M., & Hambrick, D. C. (1997). Explaining the premiums paid for large acquisitions: Evidence of CEO hubris. Administrative Science Quarterly, 42,
103–127.
Heath, C., & Tversky, A. (1991). Preferences and beliefs: Ambiguity and competence in choice under uncertainty. Journal of Risk and Uncertainty, 4, 5–28.
Hilton, D. J., Régner, I., Cabantous, L., Charalambides, L., & Vautier, S. (2011). Do positive illusions predict overconfidence in judgment? A test using interval
production and probability evaluation measures of miscalibration. Journal of Behavioral Decision Making, 24, 117–139.
Hirshleifer, D., & Thakor, A. V. (1992). Managerial conservatism, project choice, and debt. Review of Financial Studies, 5, 437–470.
Hodgkinson, G. P., Bown, N. J., Maule, A. J., & Glaister, K. W. (1999). Breaking the frame: An analysis of strategic cognition and decision making under
uncertainty. Strategic Management Journal, 20, 977–985.
Hoffman, R. R., Shadbolt, N. R., Burton, A. M., & Klein, G. (1995). Eliciting knowledge from experts: A methodological analysis. Organizational Behavior and
Human Decision Processes, 62, 129–158.
1128 J. Lambert et al. / Journal of Economic Psychology 33 (2012) 1115–1128

Juslin, P. N., Laukka, P., Liljeström, S., Västfjäll, D., & Lundqvist, L. O. (2011). Emotional reactions to music in a nationally representative sample of Swedish
adults: Prevalence and causal influences. Musicae Scientiae,15 174 207. A survey study of emotional reactions to music in everyday life.
Kim, K. A., & Nofsinger, J. R. (2002). The behavior and performance of individual investors in Japan. Review of Financial Studies Conference, Mannheim.
Kirchler, E., & Maciejovsky, B. (2002). Simultaneous Over- and Underconfidence. Evidence from Experimental Asset Markets. Journal of Risk and Uncertainty,
25, 65–85.
Klayman, J., Soll, J., Gonzalez-Vallejo, C., & Barlas, S. (1999). Overconfidence: It depends on how, what, and whom you ask. Organizational Behavior and
Human Decision Processes, 79, 216–247.
Langer, E. J., & Roth, J. (1975). Heads I win, tails it’s chance: The illusion of control as a function of the sequence of outcomes in a purely chance task. Journal
of Personality and Social Psychology, 32, 951–955.
Larrick, R. P., Burson, K. A., & Soll, J. B. (2007). Social comparison and confidence: When thinking you’re better than average predicts overconfidence.
Organizational Behavior and Human Decision Processes, 102(1), 76–94.
Lee, C., & Swaminathan, B. (2000). Price momentum and trading volume. Journal of Finance, 55, 2017–2069.
Li, X. (2002). Performance, herding, and career concerns of individual financial analysts. Working Paper: Vanderbilt University.
Lichtenstein, S., & Fischhoff, B. (1977). Do those who know more also know more about how much they know? The calibration of probability judgments.
Organizational Behavior and Human Performance, 3, 552–564.
Maciejovsky, B., & Kirchler, E. (2002). Simultaneous over-and underconfidence: Evidence from experimental asset markets. Journal of Risk and Uncertainty,
25, 65–85.
MacCrimmon, K. R., & Wehrung, D. A. (1990). Characteristics of risk taking executives. Management Science, 36, 422–435.
Malmendier, U., & Tate, G. (2005). CEO overconfidence and corporate investment. Journal of Finance, 60, 2661–2700.
Menkhoff, L., Schmidt, U., & Brozynski, T. (2006). The impact of experience on risk taking, overconfidence, and herding of fund managers: Complementary
survey evidence. European Economic Review, 50, 1753–1766.
Moore, D. A., & Healy, P. J. (2008). The trouble with overconfidence. Psychological Review, 115(2), 502–517.
Nelson, T. O., & Narens, L. (1980). A new technique for investigating the feeling of knowing. Acta Psychologica, 46, 69–80.
Newell, A., & Simon, H. A. (1972). Human problem solving. Englewood Cliffs, New Jersey: Prentice-Hall.
Nosić, A., & Weber, M. (2008). How risky do I invest: The role of risk attitudes, risk perceptions and overconfidence. Working Paper: University of Mannheim.
Odean, T. (1998). Volume, volatility, price, and profit when all traders are above average. Journal of Finance, 53, 1887–1934.
Odean, T. (1999). Do investors trade too much? American Economic Review, 89, 1279–1298.
Önkal, D., Yates, J. F., Simga-Mugan, C., & Öztin, S. (2003). Professional vs. amateur judgment accuracy: The case of foreign exchange rates. Organizational
Behavior and Human Decision Processes, 92, 169–185.
Payne, J. W. (1975). Relation of perceived risk to preferences among gambles. Journal of Experimental Psychology: Human Perception and Performance., 1,
86–94.
Prendergast, C., & Stole, L. (1996). Impetuous youngsters and jaded old-timers: Acquiring a reputation for learning. Journal of Political Economy, 194,
1105–1134.
Russo, J. E., & Schoemaker, P. (1992). Managing overconfidence. Sloan Management Review, 33, 7–17.
Sachse, K., Jungermann, H., & Belting, J. M. (2012). Investment risk – The perspective of individual investors. Journal of Economic Psychology, 33, 437–447.
Sarasvathy, D. K., Simon, H. A., & Lave, L. (1998). Perceiving and managing business risk: Differences between entrepreneurs and bankers. Journal of Economic
Behavior and Organization, 33, 207–225.
Shanteau, J. (1992a). Competence in experts: The role of task characteristics. Organizational Behavior and Human Decision Processes, 53, 252–266.
Shanteau, J. (1992b). How much information does an expert use? Is it relevant? Acta Psychologica, 81, 75–86.
Simon, M., Houghton, S., & Aquino, K. (1999). Cognitive biases, risk perception, and venture formation: How individuals decide to start companies. Journal of
Business Venturing, 15, 113–134.
Slovic, P. (1967). The relative influence of probabilities and payoffs upon perceived risk. Psychonomic Science, 9, 223–224.
Sniezek, J. A., & Buckley, T. (1991). Confidence depends on level of aggregation. Journal of Behavioral Decision Making, 4, 263–272.
Statman, M., Thorley, S., & Vorkink, K. (2004). Investor overconfidence and trading volume. Working Paper.
Svenson, O. (1981). Are we less risky and more skillful than our fellow drivers? Acta Psychologica, 47, 143–148.
Teigen, K. H., & Jorgensen, M. (2005). When 90% confidence intervals are 50% certain: On the credibility of credible intervals. Applied Cognitive Psychology, 19,
455–475.
Von Holstein, S. (1972). Probabilities forecasting: an experiment related to the stock market. Organizational Behavior and Human Performance, 8, 139–158.
Weber, E. U., & Milliman, R. A. (1997). Perceived risk attitudes: Relating risk perception to risky choice. Management Science, 43, 123–144.
Weber, E. U., Anderson, C. J., & Birnbaum, M. H. (1992). A theory of perceived risk and attractiveness. Organizational Behavior and Human Decision Processes,
52, 492–523.
Weber, E. U., Senmorgen, N., & Weber, M. (2005). Communicating asset risk : how name recognition and the format of historic volatility information affect
risk perception and investment decisions. Risk Analysis, 25, 597–609.
Yates, J. F., McDaniel, L. S., & Brown, E. S. (1991). Probabilistic forecasts of stock prices and earnings: The hazards of nascent expertise. Organizational
Behavior and Human Decision Processes, 42, 145–171.