Content uploaded by Sebastian Sattler
Author content
All content in this area was uploaded by Sebastian Sattler on Aug 05, 2015
Content may be subject to copyright.
Incentives and Inhibitors of Abusing Academic
Positions: Analysing University Students’
Decisions about Bribing Academic Staff
Peter Graeff
1
, Sebastian Sattler
2,3,
*, Guido Mehlkop
4
and Carsten Sauer
5
Abstract: In a web-based survey, we presented university students (n¼2,287) with vignettes in which
they had to decide whether to abuse the position of student assistant for personal gain. Referring to
the existing literature, we tested whether the emergence of corruption increases when benefits from
corruption and the probability of the occurrence of those benefits rise and whether corruption
decreases when costs and the probability of their occurrence increase. These four factors were varied
within a factorial survey design. Moreover, social norms against corrupt practices were measured. We
tested whether such norms reduce the willingness to commit these practices. Our study is the first to
simultaneously scrutinize these influences of corruption in universities. The results suggest that students
deliberate on the pros and cons of corrupt behaviour. Higher risks or costs of corrupt activities appear
to be significant deterrents. Higher benefits or success probabilities increase the likelihood of engaging
in corrupt activities. However, these factors are less important than the social norms against corruption.
As a consequence, our results imply that curbing corruption by monitoring and sanctioning might be
less effective than stimulating social norms against corruption or strengthening the validity of fairness
norms.
Introduction
The occurrence of corruption is a severe problem for
universities because it degrades the quality of education.
The accurate assessment of student performance is
important for both universities and society. Grades are
the most important indicator of students’ skills. Paying
bribe money for grades and thereby transgressing
university norms is a criminal offense and a serious
failure of universities and their staff. Fair grading is
crucial for the equality of educational opportunities and
student motivation to behave fairly. If students know
that they can purchase grades, their incentive to learn is
reduced (Heyneman, 2011). Furthermore, if information
about corruption within the university becomes public, a
loss of reputation is likely.
Heyneman, Anderson, and Nuraliyeva (2008) assumed
that corrupt practices can be more detrimental in the
educational field than in executive areas of the state
because negative effects for young people might emerge.
Learning that corruption can be a means to success
might be an imprint for students’ entire lives. Moreover,
students might transfer this experience to their future
work career. This consequence bears potential external-
ities, as students belong to a country’s forthcoming elite
(Heyneman, 2004). As a result, the degradation of
university students’ values system could lead to serious
future consequences (Bruhn et al., 2002; Andrei et al.,
2009). Eroded norms and ethical principles might not be
restored when the students join the work force.
Additionally, if corruption becomes pervasive within a
university, that institution is no longer contributing to
1
Christian-Albrechts University Kiel, Institute of Social Science, 24118 Kiel, Germany;
2
University of Cologne, Institute
for Sociology and Social Psychology, 50939 Cologne, Germany;
3
Cologne Graduate School in Management,
Economics and Social Sciences, 50932 Cologne, Germany;
4
University of Erfurt, Faculty of Economics, Law and Social
Sciences, 99089 Erfurt, Germany and
5
Bielefeld University, Collaborative Research Centre (SFB) 882, 33615 Bielefeld,
Germany
*Corresponding author. Email: sattler@wiso.uni-koeln.de
P.G. and S.S. contributed equally to this work.
European Sociological Review VOLUME 30 NUMBER 2 2014 230–241 230
DOI:10.1093/esr/jct036, available online at www.esr.oxfordjournals.org
Online publication 13 December 2013
ßThe Author 2013. Published by Oxford University Press. All rights reserved.
For permissions, please e-mail: journals.permissions@oup.com. Submitted: September 2012; revised: September 2013; accepted: October 2013.
at DZM University Koeln on July 8, 2014http://esr.oxfordjournals.org/Downloaded from
the selection of future leaders based on the university’s
standards of impartiality (Noah and Eckstein, 2001).
According to Heyneman, Anderson, and Nuraliyeva
(2008), this selection is the key task of these institutions.
They argued that no other institution could compensate
for such drawbacks. Additionally, if academic grades are
not awarded by individual effort and achievement, but
by simply paying for them, the allocation system is
undermined and no human capital or labour market
productivity is accumulated (idem (id.); Heyneman,
2009). As a consequence, important professional pos-
itions might be assigned to individuals who paid for an
unfairly obtained signal (certificate) to the labour market
rather than the most competent candidates. For example,
one might consider the devastating consequences if the
position of a surgeon is occupied by an incompetent
person who gained his professional position through
corruption (Rumyantseva, 2005). Employers run the risk
of hiring less-skilled staff in such a situation and must
devise selection mechanisms and means of protection
(Bardhan, 1997; Heyneman, 2009).
Academic cheating has become a topic of great interest
in social science research (Michaels and Miethe, 1989;
Cochran et al., 1999; Blankenship and Whitley, 2000;
Jordan, 2001; Vowell and Chen, 2004; Macfarlane, Zhang
and Pun, 2012), but corruption at colleges or universities
is seldom empirically investigated in the literature. Some
research has addressed prevalence rates, but few studies
have analysed the causes of corruption at the individual
level.
In countries with rampant corruption, such as many
eastern European states and several Asian states, corrup-
tion occurs as often in universities as in other public
institutions (Heyneman, Anderson and Nuraliyeva,
2008). The inefficiency of the educational systems (e.g.
slowly working administrations) in these countries
provides opportunities for corruption (Temple and
Petrov, 2004). Bribes are often given to bypass bureau-
cratic procedures. A survey in several eastern European
countries revealed that 79–84 per cent of students ‘[...]
were aware of the practice of illegal bribes to gain
admission’ and that 28–36 per cent ‘thought that
admission test scores could be changed’ (Heyneman,
Anderson and Nuraliyeva, 2008). One out of every five
Bulgarian, Croatian, and Serbian students and two out of
every five Moldovan students reported the use of illegal
methods to obtain admission to their university (id.). A
survey of Kazakh students revealed that 7 per cent of the
teachers demanded gifts or bribes in 2001 (3.3 per cent
in 2005; id.). Of the surveyed Serbian students, 20–30
per cent reported bribing faculty for a grade (very)
frequently (id.). Andrei et al. (2009) analysed corruption
within Romanian universities (e.g. passing exams after
bribing academic staff). The prevalence rates found in
these East European countries and states of the former
Union of Soviet Socialist Republics (USSR) might differ
from those in Western countries given the specific
conditions in the countries investigated (e.g. low or
erratically paid salaries in the academic field or poor
prospects after retiring from the university).
In a survey of English students, 2 per cent of the
respondents reported that they had attempted ‘to obtain
special consideration by offering or receiving favours
through, for example, bribery, seduction, corruption’
(Newstead, Franklyn-Stokes and Armstead, 1996). In the
same survey, 29 per cent of the participants also admitted
that they had ‘to mark each other’s work [the authors:
after exams] more generously than it merits’ when they
came ‘to an agreement with another student or students’.
The same behaviour occurred approximately twice as
often (65 per cent) in an earlier study (Franklyn-Stokes
and Newstead, 1995). Studies have also been conducted
on academic staff members who accepted or demanded a
corrupt offer. A study published in Nature revealed that
behaviour such as ‘relationships with students, research
subjects, or clients that may be interpreted as question-
able’, which might also include corruption, happens only
seldom (Martinson, Anderson and de Vries, 2005). Only
1.4 per cent of 3,247 surveyed scientists stated that they
had engaged in such behaviour. In a small randomized-
response survey among academic economists, 0.4 per cent
stated that they had at least once ‘accepted sex, money, or
gifts in exchange for grades’, and 4.3 per cent expected
that other faculty members in economics had done so
(List et al., 2001). These studies provide insight into the
prevalence rates of corruption but do not provide
theoretically based explanations for its occurrence.
This lack of research is surprising given the deleterious
consequences of corruption and the many opportunities
to behave corruptly in the educational field. For instance,
discretionary powers can be abused when academic
reviewers are in the position of certifying the results of
students or colleagues (Rumyantseva, 2005) or when
commission members or deans offer academic jobs
(Heyneman, 2004).
The current study investigates corruption by analysing
the acceptance of bribes in return for changing grades.
1
Bribery is a common form of corruption and comprises
a social exchange of payments and services (Morris,
2011) at the expense of other students, the university
management, or the professor. This scenario could occur
at universities or colleges. Professors often hire student
assistants to (pre-)correct exams, which provides oppor-
tunities to manipulate the exams or grading. Students
might understand this process and consider bribing the
INCENTIVES AND INHIBITORS OF ABUSING ACADEMIC POSITIONS 231
at DZM University Koeln on July 8, 2014http://esr.oxfordjournals.org/Downloaded from
student assistant to avoid failing the exam or to obtain a
good grade.
In theory, such situations are typically described via a
three-actor model that captures the corrupt exchange
(Banfield, 1975; Eisenhardt, 1989; Graeff, 2005) as
follows: A principal (here, the professor) disposes
decisional leeway to an agent (here, the student assist-
ant). The agent abuses his/her position in favour of a
client (here, the student) to realize his/her and the
client’s gains against the principal’s rules.
Owing to the involvement of several actors and their
preferences, corruption becomes an interdependent behav-
iour. The principal and agent typically work in the same
organization and are beholden to the same organizational
rules. The principal represents the interests of the
organization or adheres to universal norms, whereas
the agent might be inclined to favour his/her (and his/
her client’s) private gains at the expense of the organiza-
tion. The agent and client must each consider the
behaviour of the other and trust each other concerning
the exchanged goods or services (e.g. bribe money and
grade change).
The present article focuses on the decision-making
process of the agent only, i.e. the student assistant who
sells better grades for money.
2
According to theoretical
discussions of the reasons for corruption, two major
factors may influence the decision-making process
related to committing a corrupt act: situational condi-
tions (i.e. bribe money and fines) and social norms (cf.
Klitgaard, 1988; Graeff, 2005; Lambsdorf, 2007; Rose-
Ackerman, 2010; Bicchieri and Muldoon, 2011). These
two factors are crucial in explanations of deviant
behaviour in the sociological literature (Kroneberg,
Heintze and Mehlkop, 2010; Mehlkop and Graeff,
2010; Tittle et al., 2010). The distinction between these
factors follows the idea of different goals that can be
pursued, such as increasing one’s own resources or
acting in a socially appropriate way (Lindenberg, 2011).
In this vein, Hechter and Opp posited (2001, p. 404)
that ‘evidently, when social scientists address normative
phenomena they focus on some combination of behav-
ioral regularities, oughtness, and punishment or reward.
[...] Behavior, punishment or reward, and oughtness
can be considered separately; only by doing so can their
mutual dependence be determined [...]. This will permit
researchers to disentangle the complex relations between
these elements, their causes, and their consequences
[...]’. Even if it is assumed that the student assistant is
not able to consider all relevant consequences (typically
denoted by the idea of bounded rationality, Simon,
1997), these parameters might play a role in decision-
making about whether to participate in a corrupt deal
(Roy and Singer, 2006). Therefore, these parameters are
included in our theoretical model.
Our study aims to determine the effects of both of the
above factors. The first factor pertains to the deliberation
on the characteristics of a corrupt situation. It can be
assumed that individuals who enter or create a corrupt
situation deliberate on the advantages (benefits) and
disadvantages (costs) and the probability of their
occurrence.
According to the corrupt promises between agent and
client, the student assistant (agent) expects something
from the bribing student (client) in return for abusing
his/her position. In the present study, the benefit to the
agent is the amount of payment he or she receives from
the client in return for the illegal action of changing the
grade. We expect the following:
H1a: Higher bribes increase the willingness to engage in
corrupt transactions.
Because the success of the corrupt transaction depends
on the agent and the client behaving as they promised in
the form of a chronologically staggered exchange of
goods, trust becomes an important element in the
decision-making process (Dasgupta, 1988; Coleman,
1990; Graeff, 2005). The specific trust of the agent
(that the client will pay after the agent changes the
grade) is concomitant with the probability of realizing
the benefits (success probability).
3
Therefore, we further
hypothesize the following:
H1b: The higher the agent’s trust in the client (which is
equivalent to a higher subjective success probability), the
higher the likelihood of willingness to engage in a
corrupt transaction.
Corruption also entails costs and risks that can have a
profound impact on an actor’s decision. The agent’s
abuse of his/her discretionary power might be detected.
Here, the extent of the monitoring activities of the
professor (principal) might be an important factor. The
agent might know the detection probability, invest
energy in assessing it, or simply guess. If the corrupt
deal is discovered, there might be costs in terms of
sanctions (e.g. a fine and/or other punishment).
4
In the
absence of a chance of getting caught, >60 per cent of
students in four East European countries would cheat
(Heyneman, Anderson and Nuraliyeva, 2008). This
finding indicates the importance of the deterrent effects
of sanctions. Based on this discussion, we postulate the
following components of hypothesis 1:
H1c: Higher sanctions decrease willingness to engage in
corrupt transactions.
232 GRAEFF ET AL.
at DZM University Koeln on July 8, 2014http://esr.oxfordjournals.org/Downloaded from
H1d: A higher detection probability leads to a decreased
willingness to engage in corrupt transactions.
The second factor is the normative dimension of
corruption. Corruption ‘[...] is behaviour which devi-
ates from the formal duties of a public role because of
private-regarding [...] pecuniary or status gains; or
violates rules against the exercise of certain types of
private-regarding influence’ (Nye, 1967, p. 419). If a
student assistant sells a grade, this action can be
considered as a violation of the principals’ regulations
and rules of conduct or as a norm deviation aimed at
augmenting private gain. Such behaviour clearly contra-
dicts university codes of conduct (Braxton and Bayer,
1999) and sound scientific practices (Macfarlane, Zhang
and Pun, 2012). In this scenario, norms of fairness are
transgressed. These norms might be informally enforced
by academic staff or fellow students. Corrupt behaviour,
therefore, is linked to normative considerations and
touches on society members’ shared expectations con-
cerning appropriate and inappropriate behaviour.
Therefore, the agent’s moral conscience comes into
play. It is assumed that agents who are considering a
corrupt action should have a feeling ‘that one ought not
to do so’ if they have internalized such normative
principles. The violation of normative principles can lead
to a bad conscience or feelings of guilt, which function
as psychological costs (Cochran et al., 1999; Posner and
Rasmusen, 1999; Opp, 2013). Heyneman, Anderson and
Nuraliyeva (2008) found that 20–35 per cent of the
surveyed students in Bulgaria, Croatia, Moldova and
Serbia would feel bad after cheating and that 11–14 per
cent considered corrupt behaviour ‘unacceptable’. In
Franklyn-Stokes and Newstead’s (1995) study, 13 per
cent of the respondents reported that they had refrained
from offering favours because they perceived such
behaviour as immoral/dishonest, and 13 per cent
perceived such behaviour as unfair to other students.
Additionally, other studies on cheating, in general, or
plagiarism, in particular, found that moral beliefs were
associated with lower levels of misconduct (Diekhoff
et al., 1999; Sattler, Graeff and Willen, 2013). In line
with Detert, Sweitzer, and Trevino (2008) and
Heyneman (2009, 2011), we hypothesize the following:
H2: The weaker is agents’ oughtness of a social norm
(which can be interpreted as a form of ‘moral dis-
engagement’), the more likely they are to make unethical
and corrupt decisions.
The following analysis systematically tests the power of
the deliberation on incentives, such as bribe money, fines,
and their respective probabilities, and on social norms.
The present study explains corruption in universities, an
area that is lacking in existing research.
Method
Participants and Survey Procedure
From January to March 2012, we conducted a web
survey among German university students to assess the
influences of situational components of the decision-
making process and social norms on corruption. We
used data from the third wave of the longitudinal
FAIRUSE study.
5
A total of 3,020 students received pre-
notification letters via post from their universities. These
letters provided information about the survey topics
(study conditions, learning strategies, satisfaction with
university, etc) and the participation procedure as well as
the voluntary nature and anonymity of the study.
Because we had no personal information (such as
addresses) on the students and because the universities
did not have access to the data, the full anonymity of the
respondents was preserved. A data security officer
monitored the adherence to data security. One week
after receiving the letters, students received an email
invitation that included a personal link to the question-
naire. Thereafter, up to two reminders were sent. We
also provided incentives for completers, with the aim of
attaining high participation rates and data quality
(Go
¨ritz, 2006). Students could choose between money
sent via postal mail or PayPal, a voucher for a well-
known online retailer, and donations to United Nations
Children’s Fund (UNICEF) or Amnesty International (5
Euro each). Overall, 2,466 students (response rate ¼81.2
per cent) responded. Of these students, 2,308 (retention
rate ¼93.6 per cent) completed the survey. Owing to
item non-responses and dropouts, 2,287 cases were
considered in our analysis.
Dependent Variable
The students evaluated vignettes (Jasso, 2006; Wallander,
2009) that described a corrupt situation. After being
provided with the vignette text (see next section),
students were asked to place themselves in the position
of the student assistant (agent) and decide whether they
would accept the corruption offer of another student
(client), as described using the following question: ‘If
you were the assistant, would you change the grade?’
Answers were measured on a 10-point Likert scale that
ranged from ‘definitely not’ (coded as 0) to ‘definitely
yes’ (9). Similar measures of respondents’ willingness/
intention to engage in deviant behaviour (in academics)
after presenting a vignette (Nagin and Paternoster, 1993;
INCENTIVES AND INHIBITORS OF ABUSING ACADEMIC POSITIONS 233
at DZM University Koeln on July 8, 2014http://esr.oxfordjournals.org/Downloaded from
Tibbetts and Myers, 1999; Rettinger, Jordan and
Peschiera, 2004) or without vignettes (Beck and Ajzen,
1991; Tittle et al., 2010) were successfully applied in
previous research.
Manipulation of Independent Variables
Using a Factorial Survey Design
We applied a factorial survey design and provided vignettes
with hypothetical corruption situations. Vignettes are a
useful tool for testing theoretical presumptions (Jasso,
2006). The factorial survey technique was used because it is
difficult to directly observe the behaviour under investiga-
tion when it occurs in secrecy, as corruption typically does.
In such a situation, vignettes are a research tool that allows
the collection of data on people’s decisions in a hypothet-
ical situation (Hastie and Dawes, 2001; Murdock and
Stephens, 2007). Evidently, the variation of vignettes is not
equivalent to the manipulation of features in the real
world. Vignettes are, however, as Rettinger and Kramer
(2009: p. 297) state, ‘a good substitute for similar
manipulations in the real world when the latter are not
possible’. Moreover, the inferences that students draw when
making decisions about whether to engage in corruption as
described in the vignette can be assumed to be rather
similar to those made in the real world. Therefore, this
technique might be useful for studying an individual’s
decisionsinsituationsofcorruption.
In the between-subjects design, each student was
randomly assigned to one vignette. The vignette
provided a story about a student assistant who has the
opportunity to change the grade of a fellow student in
exchange for bribe money. In the quasi-experiment, the
levels of the following four theoretically derived dimen-
sions were varied systematically (see Table 1 for a
description of the vignette): (i) the benefit B(the
amount of money received for changing the grade),
(ii) the probability qof obtaining the benefit (a function
of the briber’s trustworthiness), (iii) the costs Cif
detected (the fine), and (iv) the likelihood pof getting
caught (the probability that the supervising professor
(principal) reviews the grading of the student assistant
and detects the corrupt act). The vignette universe
consists of 500 vignette combinations (5
3
4). We added
25 (5 5) vignettes in which only qand Bvaried to
examine situations in which only benefits occur; thus,
the deterrence component of the decision was set to zero
(P¼0 and C¼0). This set-up implies that a positive
correlation exists between pand C. Owing to the
factorial setting, there was no correlation between
the other factors. In total, 525 vignettes were used, and
each vignette was evaluated by an average of four
respondents.
Measuring Normative Beliefs
In addition to the vignettes, respondents were asked
about their normative beliefs related to the investigated
corrupt behaviour. The measures used here refer to the
concept of the oughtness of social norms.
6
This concept
is based on norm internalization and consists of three
elements, as follows: norm importance (NI) describes the
degree to which the norm is perceived as important;
Table 1 The vignette text and the variation of the decision parameters in the factorial survey design
Vignette text:
A student assistant is advised by his professor to pre-correct final exams and to recommend the grades. A student
who knows that she has failed the exams contacts him. She asks him whether he can review her exam in such a
way that she will pass it with a good grade. She offers to pay him ‘B’ Euro in return as soon as he changes the
grade.
According to his assessment of the students’ trustworthiness, she will give him the money with a ‘q’ per cent chance.
If the professor learns that the student assistant has changed the grade, it is reported to the authorities, and he has to
pay a fine of ‘C’ Euro.
The student assistant knows that it is the professor’s habit to check a ‘p’ per cent sample of all exams that were pre-
corrected by the student assistant.
Dimension Levels Coding
B-Benefits in Euro (bribe money) 25, 50, 150, 600, 3,000 25, 50, 150, 600, 3,000
q-Success probability (per cent 5, 25, 50, 75, 100 .05, .25, .5, .75, 1
C-Costs in Euro (fine) 0, 25, 50, 150, 600, 3,000 0, 25, 50, 150, 600, 3,000
p-Detection probability (per cent) 0, 5, 10, 20, 40 0, .05, .10, .20, .40
234 GRAEFF ET AL.
at DZM University Koeln on July 8, 2014http://esr.oxfordjournals.org/Downloaded from
norm obedience (NO) measures how strongly the
respondents feel obliged to follow the norm; norm
deviance (ND) represents the evaluation of transgressing
the norm of incorrupt behaviour. Table 2 provides an
overview of the operationalization.
Control Variables
The statistical models controlled for gender and age,
which might affect the willingness to engage in corrup-
tion. A few studies (Swamy et al., 2001) imply that
women are less involved in corruption than men owing
to the socialization of women or the different use of
negative social capital by men. Because the empirical
evidence of such studies refers to aggregated attitude
measures or macro-data (Alatas et al., 2009) and might
not be valid at the individual level, this suggestion
should be tested with a micro-data sample. Gender was
coded 0 for females and 1 for males. In the present
sample, 61.6 per cent of the participants were female.
We also controlled for age, as older students might be
more familiar with the proceedings in the educational
field (Andrei et al., 2009), and more extensive knowledge
of university processes could facilitate corrupt actions.
The median age group in the present sample was 22–23
years.
Results
The majority of participants (73.5 per cent; Mean ¼0.65;
SD ¼1.468) reported that they strongly oppose engaging
in corruption (see Figure 1). However, a share of
students was more or less willing to seize the oppor-
tunity and change the grade in exchange for bribe
money, as described in the vignette. Only 0.6 per cent
(scale points 8 and 9) had a relatively high tendency to
accept the offer.
Figure 2 provides an overview of how the willingness
to accept the corruption offer depends on the variations
across the vignette outcomes. An increase in the amount
of bribe money increased the willingness to accept
the offer (Panel A). An increased probability of
receiving the money after changing the grade also had
a positive impact (Panel C). Higher fines (Panel B) in
the case of detection and an increase in the detection
probability (Panel D) decreased the willingness. Bivariate
analysis of variance indicated that all relationships
(vignette levels) were significantly different from zero
(P< 0.001).
7
These results are in line with hypotheses
H1a to H1d.
Table 2 Operationalization of three elements of the ‘oughtness’ of a social norm and descriptive results
Element of
the oughtness
of a social
norm
Questions Answer options Mean SD
NI If you were in the position of a student
assistant, how important is it for you to
make true proposals for grades due to
moral reasoning?
1¼not important at all,
7¼very important
6.22 1.499
NO Is this statement true for you? As student
assistant, I should make correct proposals
for grades in all circumstances.
1¼never, 7 ¼always 6.52 0.909
ND How do you evaluate changing proposals for
grades bought for money? I consider this
to be ...
1¼not objectionable at all,
7¼very objectionable
6.45 1.355
73.5
12.3
5.4 2.3 2.1 1.9 1.4 0.5 0.3 0.3
0
25
50
75
100
Percentage of respondents
0
definitely
not
1 2 3 4 5 6 7 8 9
definitely
yes
Willingness t o change a grade in exchange for bribe money
Figure 1 Distribution of the willingness to change a grade
in exchange for bribe money (n¼2,287)
INCENTIVES AND INHIBITORS OF ABUSING ACADEMIC POSITIONS 235
at DZM University Koeln on July 8, 2014http://esr.oxfordjournals.org/Downloaded from
In the next step, nested ordinary least squares (OLS)
regression models were estimated to simultaneously test
the impacts of all factors on the willingness to accept the
corruption offer. Model 1 in Table 3 shows that all four
experimentally varied decision parameters showed the
expected sign. Higher benefits (P< 0.001) and higher
probabilities of attaining those benefits (P< 0.001)
significantly increased the willingness to enter into a
corrupt exchange. Increasing costs (P< 0.005) and an
increasing detection probability (P< 0.001) both
decreased this willingness. In Model 2, the control
variables of sex and age were added. Age (P< 0.617) had
no effect, whereas men were more prone to engage in
corruption (P< 0.002).
Adding the norm variables to the model (see Model 3)
led to a remarkable increase in explained variance.
Norms were stronger predictors in the corruption
situation of the current vignette than were benefits,
costs, and their respective probabilities. However, only
two norm variables reached significance, which partially
confirms H2. NI was not significant (P< 0.181), but NO
(P< 0.001) and deviance (P< 0.001) demonstrated sig-
nificance. A comparison of the standardized coefficients
revealed much stronger effects of NO and deviance
relative to any other variable. The significant gender
effect vanished after including norms into the equation
(P< 0.299). This result suggests that the significant
gender effect in the current sample was a spurious
reflection of norms.
We also tested for interaction effects between the
situational variables and norms (not reported here, but
available on request), as there is some evidence for such
interaction effects in the empirical modelling of deviant
behaviour (Mehlkop and Graeff, 2010; Kroneberg,
Heintze and Mehlkop, 2010). However, no significant
interaction effects could be found, with one exception:
NO and detection probability interacted positively
(P< 0.044). Owing to the singularity of the result, we
do not discuss it in detail here.
Discussion
There is a body of literature that addresses academic
dishonesty among students (for an overview, see
Diekhoff et al., 1999; Whitley, 1998; McCabe,
Trevinoand Butterfield, 2001; Murdock and Stephens,
2007). Our study contributes to this literature by
focusing on corruption among students—a seldom-
scrutinized type of academic dishonesty. This form of
deviant behaviour can only occur if there is a position
that can be abused by an agent to favour a client.
.2
.4
.6
.8
1
Willingness
25 50 150 600 3000
Benefits [Euro]
A
.2
.4
.6
.8
1
Willingness
25 50 150 600 3000
Costs [Euro]
B
.2
.4
.6
.8
1
Willingness
525 50 75 100
Success probability [%]
C
.2
.4
.6
.8
1
Willingness
0 5 10 20 40
Detection probability [%]
D
Figure 2 Mean willingness to change a grade in exchange for bribe money on a 10-point scale with the anchors 0: ‘definitely
not’ and 9: ‘definitely yes’ in relation to the amount of bribe money (Panel A), the probability of receiving the benefits after
changing the grade (C), the fine if the grade change was detected (B), and the detection probability (D). The error bars indicate
the standard errors of the mean (SEM; n¼2,287)
236 GRAEFF ET AL.
at DZM University Koeln on July 8, 2014http://esr.oxfordjournals.org/Downloaded from
Because there are many positions in academic fields that
provide opportunities to behave in a corrupt manner, it
can be assumed that corruption occurs at colleges and
universities. The present findings support this assump-
tion. However, only some students were strongly willing
to engage in corruption.
The major aim of this study was to analyse why
students engage in corrupt situations in universities. By
clarifying this behaviour, the current work may help to
reveal influences that may also explain corruption in
other life domains, as we assume that we have referred to
a general model of corrupt behaviour. We employed a
factorial survey design with vignettes within a web survey
among a large random sample of university students.
The results showed that increasing benefits in terms of
offering more bribe money and increasing the probabil-
ity of receiving this money led to significantly greater
willingness to behave corruptly, confirming H1a and
H1b. In line with H1c and H1d, increasing the costs of
corruption in terms of higher fines and a higher
probability of being detected decreased this willingness.
8
These variables presented as strong predictors of deviant
behaviour in previous studies as well (among others, see
Gibbs, 1975; Nagin and Pogarsky, 2001; Bushway and
Reuter, 2008; Sattler, Graeff and Willen, 2013; for mixed
evidence, see McCarthy and Hagan, 2005; Mehlkop and
Graeff, 2010). Our findings are also in line with the
results of studies that showed that poor economic
conditions might lead to corruption, such as the earlier
mentioned studies in East Europe and the former USSR
(Andrei et al., 2009).
In the present study, the three dimensions of the
‘oughtness’ of social norms concerning corruption
exhibited different strengths in explaining corrupt deci-
sions. Although no effect was found for the degree to
which people assessed the normative regulation as
important (contradicting H2), NO and deviance were
highly important (confirming H2). The more strongly
respondents felt obliged to follow the norm of incorrupt
behaviour and the more they disapproved of transgress-
ing such norms, the less willing they were to behave
corruptly. The present results are in line with previous
studies that found similar effects of internalized norms
(Grasmick and Bursik, 1990; Nagin and Paternoster,
1993; Tibbetts and Myers, 1999; Tittle et al., 2010;
Sattler, Graeff and Willen, 2013).
9
Referring to rational
choice theory, these effects can be explained in terms of
different costs or sanctions (Posner and Rasmusen,
1999). For example, the internalization of norms can
lead to the anticipation of shame, which is a ‘form of
potentially self-imposed, or reflective punishment’
(Grasmick and Bursik, 1990).
10
According to our results,
the oughtness of a social norm is more important in
decision-making than deliberations about the situational
factors, i.e. bribe money and fines or their respective
probabilities. No effects were found for the control
variables of gender and age.
One may expect that situational conditions and norm
predispositions can reinforce or prohibit each other in
their impact on the decision to become corrupt. One
may propose, for instance, that people with low NO are
more strongly affected by high incentives than others.
Such types of hypotheses imply interactions between the
situational decision parameters (e.g. the amount of bribe
Table 3 OLS Regression models (using Huber–White
corrections) to explain the willingness to change a
grade in exchange for bribe money (standardized
coefficients, t-values in parentheses)
a
Predictors Model 1 Model 2 Model 3
Success
probability (q)
0.079*** 0.078*** 0.078***
(3.88) (3.83) (4.20)
Benefits (B) 0.116*** 0.113*** 0.106***
(4.91) (4.80) (4.87)
Detection
probability (p)
0.093*** 0.093*** 0.089***
(4.68) (4.66) (4.74)
Costs (C) 0.054** 0.053** 0.050**
(2.84) (2.78) (2.80)
Sex (0 ¼female,
1¼male)
0.066** 0.020
(3.09) (1.04)
Age 0.011 0.004
(0.50) (0.19)
NI 0.025
(1.34)
NO 0.305***
(8.91)
ND 0.189***
(5.78)
Intercept (B-value) 0.549*** 0.519*** 5.201***
(8.67) (4.94) (13.49)
N 2,287 2,287 2,287
Adjusted R-squared 0.031 0.035 0.206
F17.18 12.80 27.76
Probability > F0.000 0.000 0.000
Test of predictors Block 1 Block 2 Block 3
R-squared 0.033 0.037 0.209
Change in R-squared 0.004 0.171
F 17.18 4.56 56.05
Block df 4 2 3
Probability > F0.000 0.011 0.000
*P < 0.05, **P < 0.01, ***P < 0.001.
Note: ‘a’ Due to the skewness of the dependent variable, all models were
replicated using negative binomial regression. Because all results were equal
in regard to substantive effects, OLS results are presented, which are easier
to interpret.
INCENTIVES AND INHIBITORS OF ABUSING ACADEMIC POSITIONS 237
at DZM University Koeln on July 8, 2014http://esr.oxfordjournals.org/Downloaded from
money) and the norm variables. However, only a single
interaction between norm variables and any of the other
decision parameters was found in the present study.
This study has several limitations. First, we investi-
gated corruption using a factorial survey rather than by
observing decisions in real situations. However, this
approach has several improvements to simple survey
techniques without experimental settings, such as the
reduction of socially desired responding (Alexander and
Becker, 1978; Choong, Ho and McDonald, 2002; Wason
et al., 2002). Second, there are multiple possibilities for
creating levels of vignette dimensions. Different vignette
set-ups might result in different results. We attempted to
construct a realistic vignette set-up that was guided by
theoretical reasoning and numerous discussions with
colleagues. The range for the amount of bribe money
and the fine were based on the consideration that
students were able to earn 10 Euro per hour in typical
student jobs. The maximum of 3,000 Euro is, therefore,
equal to 300 working hours. The success probability
ranged from a very low chance (5 per cent) to the
theoretical maximum (100 per cent). Values <5 per cent
might be unrealistic. We limited the detection probabil-
ity to 40 per cent because it was assumed that values >40
per cent might lead to immediate refusals of the
corruption opportunity. Further studies should investi-
gate other set-ups, but the replication of our results
might also produce insightful clues regarding the theory
and methods of corruption research.
The present study may help to develop strategies to
fight corruption, which should be a task of the whole
educational system, but particularly of the specific
university or college. First and foremost, all strategies
for improving the internalization of compliance norms
(e.g. by statements of honesty), university ‘courts’, or
honour codes should be at the top of the list
(Heyneman, 2011). These strategies are not prevalent
in all universities, but they can help to prevent corrup-
tion. If these strategies promote norm internalization,
the adherence to an internalized norm provides an
internal reward, whereas violations lead to psychological
costs (Coleman, 1990).
With regard to the situational parameters for decisions
about corruption (such as bribe money and fines), it
might be advisable to focus on mechanisms of detection
at both the student and the teacher levels. These
mechanisms are typically found to be a crucial factor
in preventing crimes in general (Becker, 1968; Gibbs,
1975; Nagin and Pogarsky, 2001; Dahlba
¨ck, 2003;
Bushway and Reuter, 2008).
Although the present results were consistent and
matched the theoretical expectations, future research is
needed. One specific demand results from the fact that
the study investigated the decision of a hypothetical
agent but did not focus on the client or the principal.
Additionally, a scenario involving a corrupt network was
not analysed. Future research could broaden the under-
standing of corruption in universities by investigating
the incentives and inhibitors of other positions in
corruption situations. This study was, however, a first
attempt to approach this topic and, therefore, addressed
a clear-cut situation. The results suggest that future
research with similar experimental settings could prom-
ise interesting outcomes.
Notes
1Other types of corruption in the educational field
are described in Heyneman, Anderson, and
Nuraliyeva (2008) and Heyneman (2009).
2Strategic games were not investigated here. Such
settings have been analysed in laboratory experi-
ments (Abbink, 2006).
3By only focusing on the agent’s decision, we do not
investigate the strategic problem of trust between
the agent and the client (such as mentioned in
Graeff, 2005). In the current vignettes, the respond-
ents (agents) learned the success probabilities in
terms of the client’s trustworthiness (see Methods
section). Thus, the respondents did not assess their
levels of trust in the client. There are other ways
to model trust as a decisional factor. A rather
prominent approach is to provide the agent with a
first-mover advantage in which he/she receives the
bribe money first and can decide whether to change
the grade or not (cf. Abbink, 2006).
4We do not analyse a situation in which corruption
partners abuse their position to blackmail the other
partner.
5For the initial sample of this study, a three-stage
random selection procedure was used. Therefore,
four German universities were selected. Within the
universities, 175 students were sampled from 14
randomly drawn disciplines with n
students
>175 and
300 students from all other disciplines with n
students
<175. This resulted in 11,000 eligible participants
(2,750 per university) from 138 academic
disciplines.
6After conducting our survey, we found that a paper
to which we referred concerning the oughtness
measure (Lindenberg, Joly and Stapel, 2011) had
238 GRAEFF ET AL.
at DZM University Koeln on July 8, 2014http://esr.oxfordjournals.org/Downloaded from
been retracted. There is evidence that Stapel, the
third author, manipulated the data presented in the
paper. His co-authors had no knowledge of Staple’s
manipulation. The operationalization of the ought-
ness of social norms used in the current study was
guided by the paper of Lindenberg et al. We assume
that the soundness of the theoretical concept is not
affected by the manipulated data. Moreover, the
current operationalization of oughtness is in line
with the notion of norms provided, e.g. by Hechter
and Opp (2001). However, future research is needed
to validate the measure. The paper by Wiegel et al.
(2013) is one example that shows that the concept
of the ‘oughtness’ of social norms holds predictive
validity. Our paper is another such example.
7Results are available on request.
8One anonymous reviewer has provided an interest-
ing argument on this point: the probability of being
caught depends on the professor’s efforts to monitor
the student assistant’s behaviour (see Tsebelis, 1990
for the argument in general). In our article, we were
interested in how this probability alters the re-
spondent’s decision. By varying the probability of
being caught, the design implicitly reflects the fact
that principals must make a decision about how
much time and effort to allocate to detecting
cheaters. As only the agent’s decision is investigated
in the current vignette, the principals’ monitor-
ing effort is exogenously given. However, the
corruption situation can also be modelled using a
game-theoretic approach in which the professor’s
effort is affected by the past and by the expected
prospective behaviour of the student assistant.
However, this analysis is beyond the article’s scope
and should be taken into consideration in future
research.
9Some studies did not find such effects or found
mixed evidence for internalized norms, i.e. Beck and
Ajzen (1991).
10 In cases in which the student assistant and the
student offering bribe money are in a close
relationship (such as friends or relatives), one
might assume that the student assistant might feel
obliged to help the student by changing her grade.
Then, the student assistant could find himself in a
dilemma, in which he must obey the professors’
orders and meet the expectations of his fellow
students. To avoid the influence of such
relationships, the vignette was formulated in neutral
language without mentioning any relationship
between the student assistant and the student.
However, the influence of close relationships on
corruption should be investigated in future research.
Acknowledgement
All authors contributed to, read, and approved the
manuscript.
Funding
German Federal Ministry of Education and Research
[grant number 01PH08024 to S.S.] and the Rectorate
Scholarship of Bielefeld University [grant number
3521.01 to S.S.]. The funding agencies influenced neither
the aims of the study nor the interpretation of the
results. The views expressed here do not necessarily
reflect the policies of the funders. We are fully liable for
the integrity of the data and the correctness of the data
analysis.
References
Abbink, K. (2006). Laboratory experiments on corrup-
tion. In Rose-Ackerman, S. (Ed.), International
Handbook on the Economics of Corruption.
Cheltenham, UK: Edgar Elgar, pp. 418–437.
Alatas, V. et al. (2009). Gender and corruption: insights
from an experimental analysis. Southern Economic
Journal,75, 663–668.
Alexander, C. S. and Becker, H. J. (1978). The use of
vignettes in survey research. Public Opinion
Quarterly,42, 93–104.
Andrei, T. et al. (2009). A simultaneous equation model
for estimating corruption in higher education. Acta
Oeconomica,59, 411–439.
Banfield, E. C. (1975). Corruption as a feature of
governmental organisation. Journal of Law and
Economics,18, 587–605.
Bardhan, P. (1997). Corruption and development: a
review of issues. Journal of Economic Literature,35,
1320–1346.
Beck, L. and Ajzen, I. (1991). Predicting dishonest
actions using the theory of planned behavior. Journal
of Research in Personality,25, 285–301.
Becker, G. S. (1968). Crime and punishment: an
economic approach. Journal of Political Economy,
76, 169–217.
INCENTIVES AND INHIBITORS OF ABUSING ACADEMIC POSITIONS 239
at DZM University Koeln on July 8, 2014http://esr.oxfordjournals.org/Downloaded from
Bicchieri, C. and Muldoon, R. (2011). Social Norms. In
Edward, N. Z. (Ed.), The Stanford Encyclopedia of
Philosophy (Spring 2011 Edition). <http://plato.stan
ford.edu/archives/spr2011/entries/social-norms/seen>
[accessed 3 September 2012].
Blankenship, K. L. and Whitley, B. (2000). Relation of
general deviance to academic dishonesty. Ethics and
Behavior,10, 1–12.
Braxton, J. M. and Bayer, A. E. (1999). Faculty
Misconduct in Collegiate Teaching. Baltimore: Johns
Hopkins University Press.
Bruhn, J. G. et al. (2002). Moral positions and academic
conduct: parameters of tolerance for ethics failure.
The Journal of Higher Education,73, 461–493.
Bushway, S. and Reuter, P. (2008). Economists’ contri-
bution to the study of crime and the criminal justice
system. In Tonry, M. (Ed.), Crime and Justice: A
Review of Research. Chicago: University of Chicago
Press, pp. 389–452.
Choong, P., Ho, S. and McDonald, R. A. (2002). An
examination of the effects of social desirability bias
on business ethics: research results. International
Journal of Management,19, 3–9.
Cochran, J. K. et al. (1999). Shame, embarrassment, and
formal sanction threats: extending the deterrence/
rational choice model to academic dishonesty.
Sociological Inquiry,69, 91–105.
Coleman, J. S. (1990). Foundations of Social Theory.
Cambridge: Harvard University Press.
Dahlba
¨ck, O. (2003). Analyzing Rational Crime – Models
and Methods. Dordrecht: Kluwer.
Dasgupta, P. (1988). Trust as a commodity. In
Gambetta, D. (Ed.), Trust: Making and Breaking
Cooperative Relations. Oxford: Basil Blackwell,
pp. 49–72.
Detert, J. R., Sweitzer, V. L. and Trevino, L. K. (2008).
Moral disengagement in ethical decision making: a
study of antecedents and outcomes. Journal of
Applied Psychology,93, 374–391.
Diekhoff, G. M. et al. (1999). College cheating in Japan
and the United States. Research in Higher Education,
40, 343–353.
Eisenhardt, K. M. (1989). Agency theory: an assessment
and review. Academy of Management Review,14,
57–74.
Franklyn-Stokes, A. and Newstead, S. E. (1995).
Undergraduate cheating: who does what and why?
Studies in Higher Education,20, 159–172.
Gibbs, J. P. (1975). Crime, Punishment, and Deterrence.
Amsterdam: Elsevier.
Go
¨ritz, A. S. (2006). Incentives in web surveys: meth-
odological issues and a review. International Journal
of Internet Science,1, 58–70.
Graeff, P. (2005). Why should one trust in corruption?
The linkage between corruption, trust, norms, and
social capital. In Lambsdorff, J., Graf, M. and
Schramm, M. (Eds.), The New Institutional
Economics of Corruption. London: Routledge,
pp. 21–42.
Grasmick, H. G. and Bursik, R. J. (1990). Conscience,
significant others, and rational choice: extending the
deterrence model. Law and Society Review,24,
837–861.
Hastie, R. and Dawes, R. M. (2001). Rational Choice in
an Uncertain World: The Psychology of Judgment and
Decision Making. Thousand Oaks, CA: Sage.
Hechter, M. and Opp, K. D. (2001). What have we
learned about the emergence of social norms?. In
Hechter, M. and Opp, K. D. (Eds.), Social Norms
New York: Russell Sage Foundation, pp. 394–415.
Heyneman, S. P. (2004). Education and corruption.
International Journal of Educational Development,
24, 638–648.
Heyneman, S. P. (2009). The concern with corruption in
higher education. In Gallant, T. B. (Ed.), Creating
the Ethical Academy: A Systems Approach to
Understanding Misconduct and Empowering Change.
New York: Routledge and Kegan Paul, pp. 13–27.
Heyneman, S. P. (2011). The corruption of ethics in
higher education. International Higher Education,62,
8–9.
Heyneman, S. P., Anderson, K. H. and Nuraliyeva, N.
(2008). The cost of corruption in higher education.
Comparative Education Review,51, 1–25.
Jasso, G. (2006). Factorial survey methods for studying
beliefs and judgments. Sociological Methods Research,
34, 334–423.
Jordan, A. E. (2001). College student cheating: the
role of motivation, perceived norms, attitudes, and
knowledge of institutional policy. Ethics and
Behavior,11, 233–247.
Klitgaard, R. (1988). Controlling Corruption. Berkley:
University of California Press.
Kroneberg, C., Heintze, I. and Mehlkop, G. (2010).
The interplay of moral norms and instrumental
incentives in crime causation. Criminology,48,
259–294.
Lambsdorf, J. (2007). The Institutional Economics of
Corruption and Reform. Theory, Evidence, and Policy.
Cambridge: Cambridge University Press.
Lindenberg, S. (2011). Intrinsic motivation in a new
light. Kyklos,54, 317–342.
Lindenberg, S., Joly, J. F. and Stapel, D. A. (2011,
retracted). The norm-activating power of celebrity:
the dynamics of success and influence. Social
Psychology Quarterly,74, 98–120.
240 GRAEFF ET AL.
at DZM University Koeln on July 8, 2014http://esr.oxfordjournals.org/Downloaded from
List, J. A. et al. (2001). Academic economists behaving
badly? A survey on three areas of unethical behavior.
Economic Inquiry,39, 162–170.
Macfarlane, B., Zhang, J. and Pun, A. (2012). Academic
integrity: a review of the literature. Studies in Higher
Education. (online first), 1–20.
Martinson, B. C., Anderson, M. S. and de Vries, R.
(2005). Scientists behaving badly. Nature,435,
737–738.
McCabe, D. L., Trevino, L. and Butterfield, K. D. (2001).
Cheating in academic institutions: a decade of
research. Ethics and Behavior,11, 219–232.
McCarthy, B. and Hagan, J. (2005). Danger and the
decision to offend. Social Forces,83, 1065–1096.
Mehlkop, G. and Graeff, P. (2010). Modelling a rational
choice theory of criminal action: subjective expected
utilities, norms, and interactions. Rationality and
Society,22, 189–222.
Michaels, J. W. and Miethe, T. D. (1989). Applying
theories of deviance to academic cheating. Social
Science Quarterly,70, 870–885.
Morris, S. D. (2011). Forms of corruption. CESifo DICE
Report,2, 10–14.
Murdock, T. B. and Stephens, J. B. (2007). Is cheating
wrong? Students’ reasoning about academic dishon-
esty. In Anderman, E. A. and Murdock, T. B. (Eds.),
The Psychology of Academic Cheating. San Diego, CA:
Elsevier Press.
Nagin, D. S. and Paternoster, R. (1993). Enduring
individual differences and rational choice theories
in crime. Law and Society Review,27, 467–496.
Nagin, D. S. and Pogarsky, G. (2001). Integrating
celerity, impulsivity, and extralegal sanction threats
into a model of general deterrence: theory and
evidence. Criminology,39, 865–891.
Newstead, S. E., Franklyn-Stokes, A. and Armstead, P.
(1996). Individual differences in student cheating.
Journal of Educational Psychology,88, 229–241.
Noah, H. J. and Eckstein, M. A. (2001). Fraud and
Education: The Worm in the Apple. Lanham, MD:
Rowman & Littlefield.
Nye, J. S. (1967). Corruption and political development:
a cost-benefit analysis. American Political Science
Review,61, 417–427.
Opp, K.-D. (2013). Norms and rationality. Is moral
behavior a form of rational action? Theory and
Decision,74, 383–409.
Posner, R. A. and Rasmusen, E. B. (1999). Creating and
enforcing norms, with special reference to sanctions.
International Review of Law and Economics,19,
369–382.
Rettinger, D. A., Jordan, A. E. and Peschiera, F. (2004).
Evaluating the motivation of other students to
cheat: a vignette experiment. Research in Higher
Education,45, 873–890.
Rettinger, D. A. and Kramer, Y. (2009). Situational and
personal causes of student cheating. Research in
Higher Education,50, 293–313.
Rose-Ackerman, S. (2010). The institutional economics of
corruption. In DeGraaf, G., von Maravic, P. and
Wagenaar, P. (Eds.), The Good Cause. Theoretical
Perspectives in Corruption. Opladen: Budrich, pp. 52–72.
Roy, A. and Singer, A. E. (2006). Reducing corruption in
international business: behavioral, managerial and
political approaches. Journal of Economic and Social
Policy,10, 3–24.
Rumyantseva, N. (2005). Taxonomy of corruption in
higher education. Peabody Journal of Education,80,
81–92.
Sattler, S., Graeff, P. and Willen, S. (2013).
Explaining the decision to plagiarize: an empirical
test of the interplay between rationality, norms, and
opportunity. Deviant Behavior,34, 444–463.
Simon, H. A. (1997). Models of Bounded Rationality:
Empirically Grounded Economic Reason. Cambridge,
MA: The MIT Press.
Swamy, A. et al. (2001). Gender and corruption. Journal
of Development Economics,64, 25–55.
Temple, P. and Petrov, G. (2004). Corruption in higher
education: some findings from the states of the
former Soviet Union. Higher Education Management
and Policy,16, 83–99.
Tibbetts, S. G. and Myers, D. L. (1999). Low self-control,
rational choice, and student test cheating. American
Journal of Criminal Justice,13, 179–200.
Tittle, C. R. et al. (2010). Expected utility, self-control,
morality, and criminal probability. Social Science
Research,39, 1029–1046.
Tsebelis, G. (1990). Penalty has no impact on crime: a
game theoretic analysis. Rationality and Society,2,
255–286.
Vowell, P. R. and Chen, J. (2004). Predicting academic
misconduct: a comparative test of four sociological
explanations. Sociological Inquiry,74, 226–249.
Wallander, L. (2009). 25 years of factorial surveys in soci-
ology: a review. Social Science Research,38, 505–520.
Wason, K. D., Polonsky, M. J. and Hyman, M. R.
(2002). Designing vignette studies in marketing.
Australasian Marketing Journal,10, 41–58.
Whitley, B. E. (1998). Factors associated with cheating
among college students: A review. Research in Higher
Education,29, 235–274.
Wiegel, C., Sattler, S., Goeritz, A. S. and Diewald, M.
(2013). Work-Related Stress and Cognitive
Enhancement among University Teachers.
Manuscript submitted for publication.
INCENTIVES AND INHIBITORS OF ABUSING ACADEMIC POSITIONS 241
at DZM University Koeln on July 8, 2014http://esr.oxfordjournals.org/Downloaded from