You are on page 1of 15

40

Cross Cultural Management

Cultural Dynamics of Corporate Fraud


Douglas M. Watson
Cultural conflict is the underlying cause of differential association and
therefore of systematic criminal behavior.1
Edwin H. Sutherland, 1939-

The Author
Douglas M. Watson is a fraud investigation and corporate security consultant with
the Saudi Arabian Oil Company, Dhahran, Kingdom of Saudi Arabia. He is a Certified Fraud Examiner (CFE), and a Middle East area specialist, with more than 30
years experience conducting and directing investigative operations worldwide.
Prior to taking employment in the private sector, he was the Director of Criminal
Investigations, Headquarters, Air Force Office of Special Investigations (AFOSI),
Washington, DC. He received his BS degree in Law Enforcement from the University of Nebraska-Omaha, his MBA from the University of Utah, and his PhD with
honours in Applied Management and Decision Sciences from Walden University,
Minneapolis, MN.
In Japan, US businessmen insulted their hosts by refusing traditional gift giving,
because they likened the practice to accepting bribes.
In Brazil, a US shipping company, knowing that crate handlers would steal 10%
of their shipments, split each shipment into two crates: the first crate would
haul about 90% of the goods, the second 10%. The Brazilian crate handlers
quickly learned to take the second crate and leave the first intact. From the US
companys perspective, acceding to economic extortion was a business
improvement.
In China, a manager of a US specialty products company caught an employee
stealing. The manager followed the companys US domestic policy for dealing
with such matters and turned the employee over to Chinese provincial
authorities, who executed him.
These vignettes reveal the impact that culture has on ethical dilemmas. From a
business perspective, the most important ethical dilemmas have to do with fraud
or abuse, which cost companies about 6% of their total annual revenue.2 Each of
the vignettes set forth above is a documented case previously reported in the Harvard Business Review.3 Each tells its own story of cultural differentiation. Differences can be relatively minor (insult) or relatively major (death).
It depends on perspective.

Volume 10 Number 1 2003

41

It seems intuitive that people will interpret fraud based on what they believe
constitutes fraud. This is not a new idea, but it does pose an intriguing quandary.
The Organisation for Economic Cooperation and Development (OECD), in its effort to quash international corruption, discovered that what constitutes a corrupt
act in one culture may be considered a legitimate, or at least a quasi-legitimate
business or social practice in another.4 The quandary is not with defining fraud,
per se. Rather, it is how one goes about interpreting the parameters embraced by
the definition. This raises the notion that the concept of fraud is a complex philosophy composed to two opposing thoughts.5 The first is that fraud definitions
cross cultures and are universal. This espouses the belief that most people can recognise fraud for what it is, and most people generally agree that a certain fraud is
a fraud. The second thought is that fraud cannot ultimately be understood apart
from the cultural context in which it occurs, and generalisations must refer to the
culture and subjective values of those who define it.
There may be truth in both of those thoughts. However, the idea that some
people view fraud differently may not be a simple case of ignorance of the law is
no excuse. Rather, it may be the result of something ingrained in the social nature of human interaction, particularly when cultural heritage is introduced into
the equation.
Transparency International (TI), a coalition created to identify and publicise
fraud and corruption, reported a 1994 research project carried out by the Independent Commission Against Corruption (ICAC) in New South Wales, Australia.
The ICAC wanted to determine the kind of conduct public sector employees would
judge as corrupt, and identify those factors that might keep them from reporting
it. The survey results revealed:
Individual respondents different sharply on their views of what was - or was not
- corrupt. [It] is important for all who are interested in minimising corruption
to realise that what any one public sector employee understands as corrupt
may not be shared by his or her colleagues. This lack of commonality of
understanding adds to the difficulty of combating corruption.6
TI sponsored further research, which extended the results of the ICAC survey. They concluded that corruption and fraud, as global phenomena, are perceived by the global population with that same lack of commonality. This suggests
that when assessing fraud - particularly in a multicultural setting - one could say,
one persons fraudster is anothers enterprising entrepreneur.
7

It all has to do with perspective.


Faced with the ambiguity of cross-cultural relativism, one question keeps rising to the surface. The question is: How can a business executive operating in an
international multicultural setting evaluate dilemmas associated with allegations
of bribery, conflict of interest, embezzlement, management fraud, or even indus-

42

Cross Cultural Management

trial espionage, especially when employees at various levels of the same company
may view those situations from their own different perspective?
The new economy adds new complexity. Electronic B2B and B2C
(business-to-business and business-to-consumer) economies tend to compound
ethical dilemmas. More than 230 countries have joined the electronic marketplace
with domain name designators (i.e. .uk for the United Kingdom, .sa for
Saudi Arabia, etc.).8 Instead of resolving culturally driven misunderstandings, the
anonymity of a multicultural Internet creates a faade of security. Both sides of a
transaction tend to interpret their activities solely on their respective milieu. B2B,
and even more so, B2C, bring disparate players into direct interaction. The result
is that the electronic transmissions of a multibillion-dollar G-7 corporation become essentially indistinguishable from the electronic transmissions of an insolvent third-world teenager.
This is the crux of the problem: If what I think is fraud is not considered so in
my operating environment, how do I know what to do? Do I stand by my convictions, or do I compromise my positions and go with the flow?
That quandry provided the impetus for original research to test the theory
that cultural heritage affects attitudes toward business-related fraud, and as a consequence, influences ethical dilemma decision-making.
A representative sample of white-collar employees of a large multinational
corporation (MNC) located in the Middle East evaluated scenarios describing five
business frauds - bribery, conflict of interest, embezzlement, management fraud,
and industrial espionage - against a 7-point Likert-type scale that measured levels
of approval.9
MNCs are companies that implement business strategies in production, marketing, finance and staffing that transcend national borders, and as such frequently encounter multiculturalism.10 The MNC participating in this study is
unique in the Middle East in that it has implemented a comprehensive Westernoriented business ethics programme, which includes a statement of what constitutes a conflict of interest, as well as a broad-based explanation of what constitutes good business ethics practices.11
For comparison purposes, the participants were asked to categorise themselves according to pre-defined world cultures and according to age, gender, education, and occupation.12 The sample was representative of a very diverse
population. The participants classified themselves into nine different world cultures: African, Australian/New Zealander, Asian, Caribbean/South American,
European, Mediterranean, Middle Eastern, North American, and Pacific Islander.
Figure 1 depicts the response percentages of each of the cultures. The ages of the
participants ranged from 21 to 60 years old. For the purpose of the research, they
were categorised into four age groups, as reflected in Figure 2. The largest repre-

Volume 10 Number 1 2003

43

sentation was persons in the age group 41-50, which consisted of 45% of those
who responded. Seven percent were between 21-30 years old, 21% were 31-40
years old, and the remaining 27% were between 51-60 years old.

Cutural Heritage
1. European
2. African
3. Australian/
New Zealander
4. Asian
5. Caribbean/
South American
6. Mediterranean
7. Middle Eastern
8. North American
9. Pacific Islander

1
12%

9
17%

2
7%
3
9%

8
19%
7
6%

4
14%
6
13%

5
3%

Figure 1: Distribution of cultures

Particant Age Grouping


50%
45%

45%

40%
35%
27%

30%
25%
20%
15%
10%

21%

7%

5%
0%
21-30

31-40

41-50

Figure 2: Age Distribution

51-60

44

Cross Cultural Management

The education level of the participants was also diverse. While 19% did not
have a college education, 52% were college graduates, and 29% held postgraduate degrees. Since the participants, in the main, were well educated, it was
expected they would have the knowledge to distinguish and interpret the scenarios. Figure 3 depicts the education distribution.

Education Level Distribution

Post-Graduate
Degree

29%
College Graduate

52%
Non Graduate

19%
Figure 3: Education distribution

Seventy-six percent of the participants were male, and 24% were female.
They represented more general sectors of the business, with 17% handling management positions, 50% involved with technical jobs, 24% having professional occupations, and 9% performing administrative tasks. Figure 4 and 5 reflect these
demographics.
Discussion of the Results
The participants answered 15 scenarios presented as ethical dilemmas in the form
of an anonymous questionnaire.13 The scenarios were designed to elicit attitudes
to five frauds (bribery, conflict of interest, embezzlement, management fraud, and
industrial espionage). Attitudes are made up of three integral and measurable attributes, known as recognition, feeling and action.14 Recognition involves the ability to identify with or accept a specific type of behaviour. It is an indicator of the
knowledge a person has about a subject or idea. Feeling incorporates the strength
and intensity evoked by and toward a specific type of behaviour. It is an indicator
of how the subject or idea is valued. Action equates to an engagement of response
resulting in a specific type of behaviour. It is an indicator of how a person might
respond to a situation. Fifteen scenarios were used in order to present one dilemma for each of the three attitude attributes for the five frauds. The results

Volume 10 Number 1 2003

45

Gender Distribution
76%
80%
60%
40%

24%

20%
0%
Male

Female

Figure 4: Gender distribution

Occupation Distribution
50%
40%
30%

50%

20%
10%

24%

17%

9%

0%
Management

Technical

Professional

Figure 5: Occupation distribution

Administrative

were obtained by using Analysis of the Variance (ANOVA) at the .05 level of significance (a = .05) of the summated means of the responses.15
First, the data was analysed to determine the survey groups overall attitude
to fraud. This was done by determining the range of approval of the scenarios by
the participants. The chart depicted in Figure 6 reflects that fraud in general was

46

Cross Cultural Management

Africa
AusNZ
Asia
CarSoAm
Europe
M ed
M idEast
NoAm
PacIs
Aggregate

50%

40%

30%

35.1%
32.9%

20%
11.4%
7.1%

10%

2.8%

4.3%

6.4%

0%
Stro ng
Ap p ro val

Stro ng
D isap p ro val

R aw Sco res

Figure 6: Fraud approval range, by score according to culture.

Attitudes to Fraud

Disapprove
19

18.54
18.5

18

17.54
17.5

Mean Scores

Mean Score

re C oas
u na
t as
a Percen tag
o f th
e
ScoreSco
Count
Percentage
ofethe
Total

60%

17.36

17

16.5

16.08

16.14

16

15.5

15

14.5

Approve

Bribery

Mgt Fraud

Embezzle

Figure 7: Attitudes to fraud.

Conflict

Ind Esp

Volume 10 Number 1 2003

47

DIfferentiation of Fraud Scores by Culture

5.8

5.5
5.4

5.
61
74
5.
53
52

5.
50
64

5.6
5.
39
44

Mean Score
Mean
Score

5.7

5.
71
11

5.
76
41

5.9

5.
79
89

5.
88
51

6.0

5.
94
16

6.
02
22

6.1

5.3
5.2
5.1
5.0
Pac Is

Asian

Aus/NZ

Med

Europe

Mid East

African

No Am

Car/SoAm

Fraud
Mean

Culture Cluster

Figure 8: Fraud scores according to culture.

disapproved of in 79.8% of the responses to the scenarios, with 57.3% of those


strongly disapproving. Conversely, 20.2% of the responses were uncertain or expressed more tolerant views of fraud, with 12.3% of those strongly approving. The
relative position of the individual cultures is reflected in the chart for comparison.
Then, the data was analysed to determine the participants attitudes toward
the five individual frauds, irrespective of cultural heritage. Figure 7 shows that
they disapproved most of industrial espionage, followed in order by conflict of interest, embezzlement, management fraud, and finally, bribery.
Next, the data was examined to determine whether or not the participants
differentiated in their attitudes to fraud according to their cultural heritage. Fraud
construct scores were compiled and compared according to the corresponding culture clusters. The mean was computed from the summed scores for each culture.
This is referred to as the culture cluster score. For comparative purposes, an aggregate mean score for all respondents was also computed, and is called the fraud
mean.
Examination of Culture as a Correlate
Although all the cultures recorded varying attitudes toward fraud to some extent,
four cultures - Mediterraneans, Australian/New Zealanders, Asians, and Pacific Is-

48

Cross Cultural Management

landers - significantly differed in their degree of approval from the others. Figure
8 reflects the differentiation of fraud scores between the cultures.
One feature of ANOVA (a = .05) is that it allows for subset comparison. In
this instance, four separate subsets of equivalency emerged: (a) North Americans
and Caribbean/South Americans; (b) Africans and Middle Easterners; (c) Asians,
Australians/New Zealanders, and Europeans; and (d) Pacific Islanders. This shows
that even though there is differentiation among certain cultures, the difference
may not necessarily amount to an isolation of each culture from the other. Instead, what appeared in this study was a tendency of some cultures to cluster together and interpret fraud similarly (see Figure 9).

6.1

Caribbean
South American

6.0

5.9

Mean Score

Mean Scores

5.8

North
American

African

Middle
Eastern

European

Fraud Mean

5.7
Mediterranean
Australian
New Zealander

5.6

5.5

Asian
Pacific
Islander

5.4

5.3

Culture Cluster

Figure 9: Culture clusters.

The data provided a basis to believe that cultural heritage does affect attitudes to fraud in general. Now the question remained to see if cultural heritage affected attitudes to the specific types of fraud.
To answer this question, fraud construct scores for each fraud type were compiled for cross-cultural comparison. ANOVA revealed that whereas the different
cultures did view the types of fraud somewhat differently, the variation was not
statistically significant (see Figure 10).

Volume 10 Number 1 2003

49

Score Analysis

Culture
Cluster

Bribe

ConInt

Embez

MgtFrd

IndEsp

African

Mean
StDev

5.6552
1.6799

6.1609
1.3193

5.7931
1.2489

5.2874
1.4027

6.5287
0.9121

Australian/
New Zealand

Mean
StDev

4.9630
1.9569

5.5278
1.5953

5.7685
1.2727

5.2685
1.3130

6.1481
1.0205

Asian

Mean
StDev

5.0234
1.9758

5.8304
1.5146

5.6433
1.4510

5.0117
1.7691

6.0234
1.3717

Caribbean/
So. American

Mean
StDev

5.7778
1.6896

6.1667
1.1607

5.9722
1.1126

5.8611
1.4160

6.3333
1.2112

European

Mean
StDev

5.4808
1.8875

5.8077
1.4218

5.9103
1.2976

5,4423
1.4313

6.1795
1.0871

Mediterranean

Mean
StDev

5.1449
1.9261

5.7246
1.7222

5.8116
1.3941

5.7500
1.6772

6.3611
1.0533

Middle East

Mean
StDev

5.4667
1.8476

5.9830
1.4466

5.9490
1.3668

5.3137
1.5171

6.2824
1.0586

Mean
StDev

5.8128
1.5181

5.9406
1.1919

5.7215
1.2144

5.9772
1.2040

6.2557
1.1138

In other words,
some cultures may espouse a different perspective about fraud
in a macro sense, but
when a particular type
of fraud is isolated for
personal evaluation,
the employees in this
study disapproved of
the scenarios with
relative equality, regardless of their cultural heritage.
Examination of Traditional Crime Correlates

The data from this


study was also compared and contrasted
Pacific Is.
Mean 4.8819
5.5833
5.5694
5.0625
5.8750
with traditional crime
StDev 2.0308
1.5412
1.5158
1.6532
1.3759
correlates of age, education, gender and ocFigure 10: Scores of Individual frauds. (Standarddeviations noted in italics
cupation. Only age
revealed significant
differentation. Across cultures, younger respondents (those under 31 years old)
were more tolerant to fraud. Equally interesting, those in age group 41-50 were
more tolerant of bribery, conflict of interest, and embezzlement than were the
other respondents, except those under age 31 (Figure 11).
No. American

Scholars generally agree that there is some correlation between age and
fraud.16 However, there are differences of opinion. Criminologist R.L. Akers believes the age relationship is not clear-cut. Even though younger people may act
more vicariously in ages - not just the age - that influence deviant behaviour. Peer
associations change with age. Akers found that these changes closely parallel the
age-pattern of changes in crime. Simply being a certain age does not mean someone is more disposed to fraud than one of a different age.17
In this study those respondents in age group 41-50 recorded more tolerance
to fraud than all but the under-31 group. This trend appeared equally across all
cultures. This is important, because the 41-50 age group accounted for 45% of the
respondent population. A study commissioned by the Association of Certified
Fraud Examiners found that frauds committed by older perpetrators cause greater
losses than do frauds perpetrated by younger offenders.18 Historical trend statis-

50

Cross Cultural Management

Age 21-30
Age 31-40
Age 41-50
Age 51-60
Mean
StDev

Bribery
4.2099
2.0959
5.4253
1.8870
5.2875
1.9617
5.7021
1.8036
5.1562
1.9371

Conflict
5.3457
1.7760
5.9256
1.4985
5.7425
1.5696
6.0914
1.3215
5.7763
1.5414

Embezzle
5.6420
1.7125
5.8812
1.6188
5.7407
1.7995
5.8230
1.7929
5.7717
1.7309

MgtFraud
4.8148
2.1102
5.3295
2.0093
5.3651
1.8300
5.5811
1.8463
5.2726
1.9489

IndEsp
5.6543
1.8314
6.2720
1.2675
6.2028
1.2912
6.1976
1.4723
6.0817
1.4656

Mean/StDev
5.1333
1.9052
5.7667
1.6562
5.6677
1.6904
5.8791
1.6473

Figure 11: Age as a correlate to fraud. (Standard deviations noted in italics.)

tics from the US Uniform Crime Reports19 show that arrest rates for fraud peak in
the late teens and early twenties, and by age 41 the rate of fraud arrests declines
to half of its peak. The inference is that whereas younger people may commit
more fraud, and older people may commit more expense fraud, the group that
will do the most damage is the one where the cost of fraud and the rate of fraud is
maximised.20
The chart in Figure 12 is a simple representation of this high threat area. The
chart is constructed against the mean age distribution curve for the surveys respondent population. Point 1" is the X-axis point of the youngest employees, in
this case those employees 30 and under. Point 2" is the X-axis point of the age
group with the most representation, in this case the 41-50 age group. Assuming a
linear relationship, the number of fraud crimes decreases with age, and the cost of
fraud increases with age. Point A on the chart is the point where case rate and
loss begin to maximise. Point B is the intersect where the optimum maximisation ends. After Point B fewer people lowers the potential overall costs of fraud,
even though the individual frauds may cost more.
Conclusions and Recommendations
The purpose of this study was to examine the relationship between cultural heritage and attitudes toward fraud. The data indicates that the culturalisation process is complex, with no apparent absolutes. People claiming different cultural
heritage do differ in how they approve of fraud in general. But what is equally significant is how these same people universally evaluated specific types of fraud. A
comprehensive analysis of the results, taking into consideration all the potential
correlates - cultural heritage as well as age, education, gender, and occupation indicates that the respondents may have been influenced by an over-arching corporate culture that tends to assimilate diverse attitudes into a more universal
standard of behaviour.

Volume 10 Number 1 2003

51
Age Curve Analysis
High Threat Area

Fraud Cost

"2"

"B"

"A"

"1"

Under 30 Years

Fraud Rate

31-40 Years

41-50 Years

51-60 Years

Figure 12: Cost of fraud/age of fraudster.

The respondents relative delineation of attitudes toward the different types


of fraud may be as much a reflection of the MNCs corporate ethic as they are the
respondents claimed cultural heritage. A companys ethic is integral to its corporate culture. A corporate culture consists of shared assumptions, beliefs, and values of the company that nurture and encourage employees emotional attachment,
commitment, identification, and feelings for the company, so that when employees make decisions and choose alternatives about the company, they take into account the companys goals and objectives.21 A corporate culture assists employees
in making sense of their actions by providing justification for their behaviour.22
Consequently, corporate cultures can guide employees to right behaviour, just as
they can condone wrong behaviour; or in other words - this is how we do things
in this company.
The aggregate results provide a reflection of the respondents attitudes toward how things are done in their MNC, which espouses an anti-fraud posture. Irrespective of demographic variables, almost 80% of the scenario responses reflect
disapproving attitudes to fraud in general. As a group, the respondents were most
disapproving of industrial espionage, followed by conflict of interest, embezzlement, management fraud, and bribery. Both of these findings - the negative general attitude to the scenarios, and the hierarchical ranking of fraud - are important
information for the MNC to use to counter deviant behaviour. However, while

52

Cross Cultural Management

about 80% disapproved, the remaining 20% were either ambivalent to or agreed
with the scenarios.23 The implication is much like a two-edged sword: Most employees know what constitutes fraud, and if confronted with such impropriety,
would take measures to deter, or if necessary report acts of fraud. At the same
time, a relatively large minority might not.
While this study could not definitively answer the elusive question about culturally driven misunderstandings, it has added to the pool of knowledge about
cultural differentiation, and has identified a possible fix - at least in the realm of
traditional business. One lesson for business executives is to establish, emphasise
and maintain a well-formulated awareness programme designed to teach employees how to deal with ethical dilemmas. A focused awareness programme should
increase the employees knowledge base about ethics and fraud. This should result
in greater recognition of deviant conduct, thereby decreasing the potential for undesirable behaviour. Attitudes can be ameliorated through awareness. It is a good
first step when all the employees of a single company understand their companys
perspective. The next step is to understand that others may have a different perspective. It stands to reason that within multicultural business environments, part
of any fraud awareness programme should include cultural sensitivities.
An effective awareness programme also acts as a deterrent. People who may
become disposed to fraud learn that those who supervise them know what constitutes fraud. This decreases the window of opportunity, since the possibility of discovery is increased. After considerable research, the Association of Certified Fraud
Examiners has concluded that a lack of knowledge about fraud by employees ultimately contributes to its occurrence and adds to its cost.24 Armed with this information, business executives, particularly those managing a multicultural employee
workforce in an international environment, will be able to address corporate ethics and business practices
on a more equitable basis.
In the end, it seems
that things may not always
be as they seem to be.
Consider the political satire from the Malaysia New
Straits Times (Figure 13).
Mr West accuses the
East of corruption while
he secretly receives his
own piece of the action.
Figure 13: Corruption (reprint with permission)

Clearly, a lot may depend on your perspective!

Volume 10 Number 1 2003

53

Endnotes
1. Sutherland, E.H. (1939) Principles of Criminology, Philadelphia, PA: J.B. Lippincott, Co. Quoted in Caldwell, R.G. (1965). Criminology. New York: The Ronald
Press Company.
2. Geis, G. (1996) Report to the nation on occupational fraud and abuse. Association of Certified Fraud Examiners (ACFE), Austin, TX: ACFE, p.5.
3. Donaldson, T. (1996) Values in tension: Ethics away from home. Harvard
Business Review, 74, 5, pp.48-62.
4. Yannaca-Small, C. (1995) Battling International Bribery. OECD Observer, 192,
pp.16-17.
5. Originally put forth by Beirne, cited in Leavitt, G.C. (1990) Relativism and
cross-cultural criminology: A critical analysis, Journal of Research in Crime and
Delinquency, 27, 1, p.13.
6. Independent Commission Against Corruption (ICAC). (1994, April). Unraveling
corruption, a public sector perspective: Survey of NSW public sector employees
understanding of corruption and their willingness to take action. ICAC Research
Report Number 1. Retrieved January 15, 1998 from the World Wide Web:
http://docserver.ub.rug.nc./eldoc/som/96B24.
7. Transparency International (1996) Transparency International source book: Part
A, analytical framework, and Part B, applying the framework. National Integrity
Systems. Retrieved May 3, 1997 from the World Wide Web: http://www.transparency.de/book/Part_A_B/index.html?fraud#first_hit
8. Information retrieved from ISO 3166 Two-Letter Country Code, Global Domain
Name Guide Network Laboratory. Retrieved July 31, 2000 from the World Wide
Web: http://www.seua.am/netinfo/other-info/index.html
9. The sample consisted of 695 employees who were randomly selected from a
general population of 19,011 by using disproportionate stratified allocations of
specifically defined population clusters corresponding to requisite culture clusters
identified in the MNCs employee base. Four hundred and sixteen (416) employees (60%) representing nine different world cultures responded.
10. Lacey, S.J. (1994, Spring) International business and multinational corporations. Unpublished manuscript, Florida State University. Retrieved June 13, 2000
from the World Wide Web: http://mailer.fsu.edu/~kshelfer/busref/papers/intmulti.html
11. Watson, D.M. (2000) Cross-cultural interpretations of fraud: An attitudinal
study in a Middle Eastern multinational corporation. Dissertation submitted in

54

Cross Cultural Management

partial fulfillment of degree of Doctor of Philosophy, Walden University, Minneapolis, MN, p.19.
12. Of the respondents, one declined to identify a cultural heritage, five declined
to identify a level of education, three declined to identify an occupation, and one
declined to report gender. These anomalies were factored into the analysis.
13. Questionnaire is available upon request, and is incorporated as part of: Watson, D.M. (2000) Cross-cultural interpretations of fraud: An attitudinal study in a
Middle Eastern multinational corporation. Dissertation submitted in partial fulfillment of degree of Doctor of Philosophy, Walden University, Minneapolis, MN.
14. Norland, E. (1991) Attitude measurement, Camping, 1, pp.30-33; Petty,
R.E., Wegener, D.T., Fabrigar, L.R. (1997) Attitudes and attitude change, Annual Review of Psychology, 48, pp.609-648, pp.611, 629-630.
15. Analysis of the variance (ANOVA) was the primary means used to analyse key
data. Responses were subjected to ANOVA at the .05 level of significance (a=.05).
For source documents and analytical databases see: Watson, D.M. (2000) Crosscultural interpretations of fraud: An attitudinal study in a Middle Eastern multinational corporation. Dissertation submitted in partial fulfillment of degree of Doctor of Philosophy, Walden University, Minneapolis, MN.
16. Akers, R.L. (1998) Social learning and social structure: A general theory of crime
and deviance. Boston: Northeastern University Press, pp.346-358.
17. Akers, op.cit., pp.357-358.
18. Geis, op.cit., p.21.
19. Gottfredson, M.R. and Hirschi, T. (1990) A general theory of crime. Stanford,
CA: Stanford University Press, pp.192-193.
20. Geis, G. op.cit., pp.21, 35.
21. Almhelm, E.A. (1992, December 16). A study in organizational culture: The
case of SAMAREC. Unpublished doctoral dissertation, Auburn University, Auburn,
AL, p.54.
22. Ibid., p.55.
23. See Watson, op.cit., pp.184-185.
24. Geis, op.cit., p.35.