1TOJ
1TOJ
Open Access
Public Accountants’ Perceptions of Ethical Work Climate: An Exploratory
Study of the Difference Between Partners and Employees within the
Instrumental Dimension
Howard Buchan*
State University of New York at Oneonta, Division of Economics & Business, Ravine Parkway, Oneonta, NY 13820
Abstract: The purpose of this study is to consider whether the perceptions of the instrumental dimension of ethical work
climate of partners (owners) of public accounting firms differ from those of employees. Perceptions of ethical climate are
based on the theory developed by Victor and Cullen [1, 2]. Professionals from five public accounting firms located in
New York State participated in the study. Findings suggest partners’ perceptions of the instrumental dimension of ethical
work climate differ from those of employees.
Victor and Cullen [2] argue that ethical climates are mul- pany profit and team interest. Coefficient alpha used to as-
tidimensional and represent normative control systems sess instrument reliability ranged from .69 to .85.
molded by societal norms, organizational form and certain
firm specific factors. For example, based on the findings of Relationship of Ethical Work Climate to Other Con-
structs
their initial study, the authors suggest that all firms form a
“base level” of caring ethical climate. Arguably, this is Wimbush, Shepard and Markham [4] attempted to repli-
somewhat reflective of societal norms. Moreover, this cate the dimensions found by Victor and Cullen [2] by test-
seemed to be the more preferred climate by workers. ing the ECQ in sub-units of a multi-unit organization. Rather
The authors suggest that organizational form also influ- than five dimensions, results of their analysis supported three
ences the development of ethical climate type. For example, (i.e., caring, independence and instrumental) empirically
bureaucratic and economic theories predict certain relation- determined by Victor and Cullen. Factor analysis did not
ships between the normative and technological/structure of distinguish between law and code and rules. Moreover, a
an organization. Ethical climate theory also is largely based new factor labeled “service” was identified. They suggest
on normative systems and, therefore, should be influenced that this factor may not have surfaced in Victor and Cullen’s
by the form of organization. factor analysis if no market forms of governance were in-
cluded in their sample. The association between organiza-
Subsequent Testing of the ECQ tional form and ethical climate type (Victor and Cullen’s
predictability hypothesis) was also of interest in this investi-
Cullen, Victor and Bronson [3] also examined the psy-
chometric properties of the ECQ. Based on results of the gation. For example, the authors tested whether employees
of retail stores perceived an instrumental climate; results
earlier studies [1, 2], the questionnaire was modified to in-
showed little support for that hypothesis.
clude 4 items for each of the nine hypothesized climate di-
mensions. Across 12 organizations which included four ac- Wimbush et al. [10] were among the first to explore the
counting firms, 1,167 individuals were sampled. Unlike the relationship between ethical climate and ethical behavior
initial assessment of the ECQ, results suggested seven cli- considering multiple levels of analysis. Ethical behavior was
mate types with self-interest and efficiency from the “ego- measured by responses to four vignettes that were based on
ism” criterion, friendship/team interest and social responsi- actual events (e.g., lying, stealing). The authors hypothesized
bility from the “benevolence” criterion, and personal moral- that the five ethical dimensions measured by the ECQ (i.e.,
ity, rules, and laws and professional codes from the “princi- caring, law and code, rules, independence and instrumental)
ple” criterion. Table 2 shows both the hypothesized dimen- and ethical behaviors would vary among organizational sub-
sions [1] and a summary of those empirically determined units and that those high on the four ethical dimensions (car-
(italicized in Table 2) in their three validation studies [1-3]. ing, law and code, rules and independence) would also dem-
The only dimensions unidentified at that point include com- onstrate high levels of ethical behavior. Results showed
modest support (12 of 20 relationships were significant) at raised as an issue, perhaps cultural as well as company spe-
the individual level and no significant relationships at the cific factors influenced the observed results.
district level.
Another interesting feature of this study was considera-
Fritzsche [12] classified subjects employed by a high tion of the potential relationship between organizational cli-
technology firm by climate type based on responses to the mate and misconduct. The author points out that: “ethical
ECQ. While not entirely clear, I assume that respondents climates are embedded in the organizational climate which is
were assigned to the category receiving the highest score on embedded in the organizational culture and thereby affecting
the scale. The author was interested in the relationship be- behavior differently. The difference, we believe, lies in the
tween the predominant climate type perceived by the indi- level of specificity of the criterion.”(p. 333). This view is
vidual and responses (ethical or unethical) to vignettes pos- certainly consistent with Victor and Cullen’s [2] approach to
ing various ethical dilemmas. Topics dealt with included ethical climate.
bribery, coercion, deception, theft and unfair discrimination.
Findings suggest significant relationships between orga-
Approximately 60% of the subjects were classified as laws
nizational and ethical climate and misconduct, with ethical
and professional code and generally chose the ethical path.
climate explaining the greatest percentage of the variance.
However, the response to the bribery vignette proved to be
Contrary to expectations, differences in perceived ethical
the exception; approximately 50% of the subjects were will- climate between departments were not found. However, this
ing to make payment.
was a relatively small company (138 employees and 97 us-
In a similar study, Barnett and Vaicys [13] investigated able responses) with the majority of respondents working in
the potential impact of ethical climate on behavioral inten- the production department. The dominant ethical climate was
tions. The authors also used the Multidimensional Ethics rules and regulations. As the authors point out, this result
Scale (MES) as a measure of ethical judgment. A survey may be typical of organizations dominated by production
instrument was mailed to a random sample of 1,000 mem- related activities.
bers of the American Marketing Association; 207 usable
responses were received (20.7%). Because this represents Development of Hypothesis
data from (potentially) 207 different organizations, the data The purpose of this study is to determine whether owners
cannot be aggregated to determine existence of a predomi- (partners) perceptions of the instrumental dimension of ethi-
nant climate type, sub-climates within an organization, and cal work climate differed from those of other levels in the
so on. firm. Even though no study has apparently considered the
However, the potential relationship between perception specific issue in a public accounting setting, prior research
of ethical climate and behavioral intentions at the individual does provide some guidance. For example, Finn, Chonko
level remains a central building block for creating a multi- and Hunt [15] found that “upper echelon” CPAs with the
dimensional/multi-level model. Another boundary of this highest levels of income identified fewer ethical issues than
particular study relates to the composition of the sample. The those with lower levels of income. Because the study fo-
majority were marketing managers with mean annual house- cused on top management, the authors suggest future re-
hold incomes of approximately $105,000. search should examine perceptions of employees at “lower
levels.”
Factor analysis suggested four ethical climates which
Barnett and Vaicys [13] labeled as: self-interest, Forte [8] also found a relationship between management
team/friendship, social responsibility and rules/code. Failure levels and perceptions of ethical issues. Attitudes of those
to more closely duplicate earlier studies may relate to the with lower levels of managerial experience and income to-
sample restrictions discussed above. The survey instrument wards ethical issues involving stakeholder relations were less
included an (arguably) unethical act. Subjects completed the positive than those with higher levels of income and experi-
MES as a measure of ethical judgment and were asked to ence.
indicate the likelihood they would engage in the behavior Cullen et al. [3] point out that the loci of analysis (e.g.
described in the scenario if in a similar situation. Results did individual, local and cosmopolitan) can combine in different
not support a direct effect of perceived ethical climate on ways depending on the structure and unique characteristics
behavioral intentions. However, they suggest ethical climate of the organization. As such, the study considered the rela-
moderates the ethical judgment-intention relationship. tionship between ethical criteria (e.g., egoistic, benevolent
Vardi [14] also considered the relationship between ethi- and principled) and organizational commitment. While the
cal climate and behavior. Unlike Barnett and Vaicys [13], the research considered two types of organizations, of interest
author focused on one organization, a metal products com- here is the study of accounting firms: two were offices of
pany located in northern Israel. Subjects included employees multinational firms and two were small local firms. Findings
from all four departments: production, production services, suggest a positive relationship between the benevolent and
marketing and administration. Unlike other studies, the sur- principled dimensions and organizational commitment. Con-
vey instrument asked subjects to indicate the frequency of sistent with prior research [16, 17] perceptions of egoistic
misconduct in the plant. climates were negatively related to commitment.
Factor analysis of the translated version of Victor and Peterson’s [9] study of business professionals found that
Cullen’s Ethical Climate Questionnaire produced three sig- self-interest and company profit (egoist criteria can combine
nificant factors: rules, caring and instrumental. Several stud- to form instrumental climate) were the only dimensions that
ies support five dimensions which include those found in this correlated positively with unethical behavior. This approach
study plus laws and code and independence. Even though not is consistent with Wimbush and Shepard’s [17] suggestion
4 The Open Ethics Journal, 2009, Volume 3 Howard Buchan
that the other climate types support ethical intentions. Thus, alpha was .71. Prior studies provide support for this dimen-
instrumental climates that emphasize individual self-interest sion and report acceptable levels of reliability (e.g., Flannery
and company interests above all others are most likely to and May [18], =.81; Kelley and Dorsch [19], =.73; and
foster unethical behavioral intentions. Wimbush et al.’s [10] Wimbush, Shepard and Markham [10], >.72).
empirical study discussed earlier found some support for this
assertion. Social Desirability Bias
The foregoing leads to the following hypothesis: Beck and Ajzen [20] illustrate the difficulties encoun-
tered in ethics research and acknowledge the criticisms sur-
Partners (owners) perceptions of instrumental ethical rounding self-report data, but suggest “there are few, if any,
work climate will differ from those of employees practical alternatives that could provide equally interesting
RESEARCH METHODOLOGY and detailed information about an individual. The practice of
relying on self-reports is thus likely to continue, even though
Research Design it is well recognized that such reports may be biased by ten-
dencies to furnish socially desirable responses and to deny
The proposed study is a nonexperimental cross-sectional
holding socially undesirable attitude or performing socially
design using a survey questionnaire. Human subjects are
undesirable behavior.” (p. 291).
required and the protocol was approved by the appropriate
institutional review board. The Marlowe-Crowne Social Desirability Scale [21] has
been widely used in personality research to test for the pres-
Sample and Procedure ence of this type of response bias. Several shorter versions of
Five public accounting firms operating primarily in one the standard 33-item scale have been developed. Confirma-
state and located in the northeast agreed to participate in this tory factor analysis [22, 23] and testing of various versions
study. Access to professionals is often restricted and gener- provide support for certain sub-scales. Fischer and Fick [23]
ally requires use of convenience samples. Even though the found that the scale developed by Strahan and Gerbasi [24]
sample is nonprobabilistic, subjects are representative of the to have high internal consistency and was strongly correlated
domain of interest and analysis of results will consider the with the standard Social Desirability Scale (SDS). Thus, this
boundary conditions and other potential limitations. The ex- 10-item scale was used in this study as a test for response
tent of participation and the administration of the survey bias. Reliability of the scale was evaluated by computing
instrument varied among the firms. Cronbach’s alpha. The degree of correlation with the instru-
mental scale was used to assess potential bias.
Administration of the Survey Instrument
DATA ANALYSIS, DISCUSSION AND CONCLUSION
Potential subjects received a letter inviting them to par-
ticipate in the study. The letter adhered to guidelines estab- Sample Characteristics
lished by the human subjects institutional review board,
Table 4 presents the distribution of staffing levels of the
briefly described the study and offered assurance that re-
participants in the five firms. Overall, participation of the
sponses were confidential and anonymity was guaranteed.
owners of the firms illustrates strong support for the project.
Response Rate Partners/principals represent 30 percent of the subjects and
employees 70 percent.
Table 3 summarizes the surveys distributed and received.
The overall response rate was 65.6% resulting in a final Table 4. Position in the Firm
sample size of ninety-five responses.
Staff 32 34
Firm Distributed Usable Responses
Senior 14 15
A 13 13 Manager/Supervisor 15 16
B 12 12 Senior Manager 5 5
C 40 19 Partner/Principal 29 30
D 55 33 Total 95 100
E 40 18
Total 160 95 Table 5 presents the areas of specialization within the
participating firms. Audit combined with the audit and tax
category represents 64 percent of the total number of sub-
INSTRUMENTATION jects. What appears somewhat unusual is the large percent-
age of respondents who identified tax as their area of spe-
Instrumental Ethical Climate cialization.
The seven questionnaire items developed by Victor and Table 6 presents the average years of experience by gen-
Cullen [2] were used to measure instrumental climate (Ap- der. Females represent 45 percent of the total respondents,
pendix I). Reliability of the scale, measured by coefficient which may reflect recent trends in public accounting. The
Public Accountants’ Perceptions of Ethical Work Climate The Open Ethics Journal, 2009, Volume 3 5
average number of years of experience for females is consid- A one-way ANOVA (Table 8) was used to test the hy-
erably less than that of males. However, this does not seem pothesis. The independent variable, position in the firm, in-
unusual given current trends. cluded five levels: staff, senior, manager/supervisor, senior
manager and partner/principal. The dependent variable was
Table 5. Area of Specialization the mean of the items measuring instrumental ethical cli-
mate. The ANOVA was significant, F(4, 90)=2.86, p=.03.
Area Number Percent Follow-up tests were conducted to evaluate pairwise dif-
ferences among means. Because variances among the groups
Audit 50 53 ranged from .50 to 1.51 and the results of the Levene Test of
Tax 31 32 Equality of Variance, F(4, 90)=.89, p=.473, I chose not to
assume that the variances were homogeneous and conducted
Tax & Audit 14 15 post hoc comparisons using the Dunnett’s C test, a test that
Total 95 100 does not assume equal variances among the groups. There
was a significant difference between the partner/principal
group and the staff and senior groups.
Data Analysis and Results
The correlation (r= -.114) of the short-version of the
Table 7 presents the climate mean for the five positions Marlowe-Crowne scale with the instrumental scale was not
in the firm. Notably, the partner/principal group’s mean significant.
score is the lowest among the categories.
Female 45 47 6.2
Male 50 53 13.7
Total 95 100
Source Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared
APPENDIX 1
Instrumental Ethical Climate
Ethical Climate Questionnaire
INSTRUCTIONS
We would like to ask you some questions about the general climate in your Firm. Please answer the following in terms of
how it really is in your Firm, not how you would prefer it to be. Please be as candid as possible; remember, all your responses
will remain strictly anonymous.
Please indicate whether you agree with each of the following statements about your company. Please use the scale below
and write the number which best represents your answer in the space next to each item.
To what extent are the following statements true about your Firm?
Completely Mostly Somewhat Somewhat Mostly Completely
false false false true true true
0 1 2 3 4 5
Public Accountants’ Perceptions of Ethical Work Climate The Open Ethics Journal, 2009, Volume 3 7
E1 In this Firm, people protect their own interests above all else._____
E2 In this Firm, people are mostly out for themselves._____
E3 There is no room for one’s own personal morals or ethics in this Firm._____
E4 People are expected to do anything to further the Firm’s interests, regardless of the consequences._____
E5 People here are concerned with the Firm’s interests-to the exclusion of all else._____
E6 Work is considered substandard only when it hurts the Firm’s interests._____
E7 The major responsibility of people in the Firm is to control costs._____
Adapted from Victor and Cullen (1988).
REFERENCES [14] Vardi Y. The effects of organizational and ethical climates on
misconduct at work. J Bus Ethics 2001; 29: 325-37.
[1] Victor B, Cullen J. A theory and measure of ethical climates in [15] Finn D, Chonko L, Hunt S. Ethical problems in public accounting:
organizations. Res Corp Soc Perform Policy 1987; 9: 51-71. The view from the top. J Bus Ethics 1988; 7: 605-16.
[2] Victor B, Cullen J. The organizational bases of ethical work cli- [16] Trevino L, Butterfield K, McCabe D. The ethical context in organi-
mate. Admin Sci Q 1988; 33: 101-25. zations: influences on employee attitudes and behaviors. Bus Ethics
[3] Cullen J, Victor B, Bronson J. The ethical climate questionnaire: Q 1998; 8(3): 447-77.
An assessment of its development and validity. Psychol Reports [17] Wimbush J, Shepard J. Toward an understanding of ethical climate:
1993; 73: 667-74. Its relationship to ethical behavior and supervisory influence. J Bus
[4] Wimbush J, Shepard J, Markham S. An empirical examination of Ethics 1994; 13: 637-47.
the multi-dimensionality of ethical climate in organizations. J Bus [18] Flannery B, May D. Environmental ethical decision making in the
Ethics 1997; 16 (1): 67-78. U.S. metal finishing industry. Acad Manag J 2000; 43: 642-62.
[5] DeConinck J, Lewis W. The influence of deontological and teleo- [19] Kelley S, Dorsch M. Ethical climate, organizational commitment,
logical considerations and ethical climate on sales managers’ inten- and indebtedness among purchasing executives. J Pers Selling
tions to reward or punish sales force behavior. J Bus Ethics 1997; Sales Manag 1991; 11(4): 55-66.
16(5): 497-507. [20] Beck L, Ajzen I. Predicting dishonest actions using the theory of
[6] Cullen J, Parboteeah K, Victor B. The effects of ethical climates on planned behavior. J Res Pers 1991; 25: 285-301.
organizational commitment: A two-study analysis. J Bus Ethics [21] Crowne D, Marlowe D. A scale of social desirability independent
2003; 46: 127-41. of psychopathology. J Consult Psychol 1960; 24: 349-54.
[7] VanSandt C, Shepard J, Zappe S. An examination of the relation- [22] Loo R. Confirmatory factor analysis of the full and short versions
ship between ethical work climate and moral awareness. J Bus Eth- of the Marlowe-Crowne Social Desirability Scale. J Soc Psychol
ics 2006; 68: 409-32. 2000; 140(5): 628-35.
[8] Forte A. Antecedents of managers moral reasoning. J Bus Ethics [23] Fischer D, Fick C. Measuring social desirability: Short forms of the
2004; 51: 315-47. marlowe-crowne social desirability scale. Educ Psychol Measure-
[9] Peterson D. The relationship between unethical behavior and the ment 1993; 53: 417-25.
dimensions of the ethical climate questionnaire. J Bus Ethics 2002; [24] Strahan R, Gerbasi K. Short, homogeneous version of the marlowe-
41: 313-26. crowne social desirability scale. J Clin Psychol 1972; 28: 191-3.
[10] Wimbush J, Shepard J, Markham S. An empirical examination of [25] Vidaver-Cohen D. Moral climate in business firms: a conceptual
the relationship between ethical climate and ethical behavior from framework for analysis and change. J Bus Ethics 1998; 17(11):
multiple levels of analysis. J Bus Ethics 1997; 16(16): 1705-17. 1211-27.
[11] Kohlberg L. Stages and sequences: The cognitive developmental [26] Schein EH. Organization Culture and Leadership. San Francisco:
approach to socialization. In: Goslin D, Ed. Handbook of socializa- Jossey-Boss 1992.
tion theory and research. Chicago, IL: Rand McNally 1969; pp. [27] Gottlieb J, Sanzgiri J. Towards an ethical dimension of decision
347-480. making in organizations. J Bus Ethics 1996; 15: 1275-85.
[12] Fritzsche D. Ethical climates and the ethical dimension of decision [28] Carlson D, Perrewe P. Institutionalization of organizational ethics
making. J Bus Ethics 2000; 24(2): 125-41. through transformational leadership. J Bus Ethics 1995; 14: 829-38.
[13] Barnett T, Vaicys C. The moderating effect of individuals’ percep-
tions of ethical work climate on ethical judgments and behavioral
intentions. J Bus Ethics 2000; 27: 351-62.
Received: November 17, 2008 Revised: January 1, 2009 Accepted: January 5, 2009