08D2-08 - Using the theoretical concepts throughout the text, analyze the distribution of wealth in America, focusing on how trends in wealth disparity impacts your work as a public administrator. Compare this to wealth distribution in at least one other

Reproduced with permission of M.E. Sharpe, Inc. in the format electronic usage via Copyright Clearance Center.

__________ 11 The Impact of Demographic, Professional, and Organizational Variables and Domain on the Moral Reasoning of Public Administrators Debra W. Stewart and Norman A. Sprinthall Most studies of managerial ethics focus on the attitudes and values of the respon dents, and the lion's share of the empirical work has looked at American busi ness managers rather than at public administrators. This chapter reports on a study of moral reasoning rather than simply on ethical attitudes or values, and explores empirically that reasoning in samples of public administrators and pub lic administration graduate students. Variables thought to explain variation in aspects of ethics include age, gender, seniority (Hodgkinson 1971); education (Purcell 1977); function and level of responsibili and 1984), an et aJ. 1989; Trevino 1 ch of these variables ty (Hunt, Wood, and Chonko 1989; Harris 1989; Posner d organizational context (Con ner and Becker 1975; Hunt 986). In this study we explore the relationship between ea and levels of moral reasoning among public administrators. Our research allows us to look at these relation ships in three ethical domains and to consider the aggregate relationships as well.

This study represents an effort to move beyond case study and admonition to ward understanding the way moral reasoning and ethical behavior are influenced in organizations.

Assumptions and Background This essay assumes that the public administrator is a moral agent. While we recognize that the development of individual agency is not a simple process (McDonald and Victor 1988), we assume that it is an essential feature of life in 205 206 DEBRA W. STEWART AND NORMAN A. SPRINTHALL public-sector organizations (Stewart 1985; Denhardt 1988). But to agree with Denhardt (1988) that the ethical administrator has a responsibility to utilize moral assessments is to leave unanswered the question of how to conduct that moral assessment.. In an earlier essay we argued for the need to conduct more research on the systems of ethical reasoning that public administrators may em ploy in resolving ethical dilemmas. We also presented a new instrument to assess stages of moral reasoning exhibited by public administrators (Stewart and SprinthalI 1991). It was based on the Kohlberg (1984) theory that moral develop ment occurs in a specific sequence of stages, and that ethical or moral judgment is neither fixed nor relativistic but rather forms a sequence of ethical models. We argued previously that the higher order models are more democratic and just in their comprehensiveness and thus are compatible with the ideals of public ser vice (Stewart and Sprinthall 1991). The instrument we developed assesses levels of moral reasoning among individuals in a public administration context. Data and Methods Detailed information on the development of the instrument, the Stewart Sprinthall Management Survey (SSMS), is reported elsewhere (Stewart and Sprinthall 1991). The survey is based on Lawrence Kohlberg's finding that moral development occurs in a specific series of stages across cultures. By studying empirically the system of thinking that people employ to deal with moral questions, Kohlberg identified an invariant sequence of stages of moral growth that ranges from a straightforward concern about self to a stage focused on the application of universal moral principles such as those that relate to justice and equality (Kohlberg 1984). James Rest developed the Defining Issues Test (DIT) as an objective adaptation of Kohlberg's interview (Rest 1986). Like the Kohlberg instrument, the DIT presents general moral dilemmas and classifies individuals according to the arguments they invoke to solve these dilemmas. The SSMS parallels the DIT but is designed to assess nloral reasoning evoked in a public-sector management context. The dilemmas provided in the SSMS deal with the actual ethical quandaries individuals encounter in public administration. In that context the SSMS reflects reasoning across five stages of moral develop ment. The stages are as follows:

Stage 1: Concern for Obedience and Punishment. To avoid punishment one must be obedient-fear of punishment is a major motivator.

Stage 2: Concern for Cooperation and Reciprocity in a Single Instance. Cooper ati ve interactions are entered because each party has something to gain. "Let's make a deal." It is the exchange that makes it fair. Bargains are struck to achieve self-interest. Materialism predominates.

Stage 3: Concern for Enduring Personal Relationships. Maintaining of good relationships over time is valued; approval of others is important. Be kind and MORAL REASONING OF PUBLIC ADMINISTRATION 207 considerate and you will get along with others. Engage in reciprocal role taking; social conformity is the highest value.

Stage 4: Concern for Law and Duty. Authority maintains morality; everyone in society is obligated and protected by the law; respect for the authority of law is part of one's obligation to society.

Stage 5: (P): Principled Reasoning .. This mode of reasoning envisions the mind of a hypothetical rational person-what agreement would a hypothetical group of rational people accept? Impartiality is central. Democratic principles of justice and fairness are the core values.

In the SSMS the individual dilemmas deal with three domains of administra tive decision making: promotion, with the attendant issues of affirmative action and patronage; procurement, with the ever-present concerns of conflict of inter est; and data base management, with the related issues of data file integrity. In each domain the SSMS provides opportunity for the respondent to reflect upon the mode of reasoning employed. The total SSMS scores represent averages across all three administrative domains. For an individual respondent the stage scores reflect the tendency for that individual to select reasons that correspond to different levels of moral development. Analysis The instrument was administered to three separate samples in North Carolina:

graduate students in public administration (N = 75), local government managers attending a series of executive training programs (N = 136), and city and county managers and assistant managers (N = 190). Across all three samples the pattern of distribution is similar, with the most commonly selected reasons falling in stage 4 and the principled stage. There was a clear pulling toward "law and duty" reasoning with almost 40 percent of the respondents in each sample, reflecting principled reasoning. This figure corresponds to the level of moral reasoning found in the U.S. population as a whole for individuals who have completed college. However, it is significantly higher than that reported for the southern region (Sapp 1985).

As shown in Table 11.1, the percentages indicate the number of persons whose reasoning could be classified according to stage type. Thus, 40 percent of the graduate student sample, or thirty subjects, employed principled reasons in resolution of the administrative dilemmas. Thirty-five of these subjects used stage 4, or the legal code, as their means of resolving the issues. The number of subjects according to stage in the other two groups is similar. Since the survey measure is a recognition test, the percentages reflect the subjects' ability to identify stage type reasons from a list of choices. It has been shown in other research on moral reasoning that such an approach actually overestimates by one stage the level of reasons that the subject can produce de novo or can actually 208 DEBRA W. STEWART AND NORMAN A. SPRINTHALL Table 11.1 Total SSMS Stage Score Comparison across All Groups Students Local Gov't Mgr. (%) City/County Mgr. (%) Stage Scores (N=136) 1+2 5 6 7 3 5 7 10 4 46 45 42 5 40 39 38 Note: The percentages vary due to rounding and a few unclassified subjects. employ in real dilemma situations (Rest 1986; Koh1berg 1984). Thus, a person who identifies a preference for principled solutions may actually employ stage 4, or the legal code, in a real situation. Similarly, a preference for stage 4 would likely shift closer to stage 3, social conformity, in an actual administrative di lemma. The base rate for our sample is similar to college educated adults in general and is higher than adults in general from the South and Southeast. The actual level of reasoning, however, is probably one stage lower than the level obtained by a recognition test. We will now tum to an analysis of possible differences within our sample according to demography, organizational responsi bilities, and context. Demographic Factors In this analysis we look at the impact of a set of demographic factors that have been thought to influence the kinds of ethical choices individuals are likely to make and the factors they are likely to consider important in this decision-mak ing process. As Tables 11.2 through 11.5 suggest, there are no significant differ ences that can be attributed to gender, race or education level across any of the three samples.

Carol Gilligan (1982) has speculated specifically on the importance of gender differences and level of ethical reasoning .. She has argued that women tend to reason about moral conflict based on a notion of morality as care, while men reason through such conflicts based on morality as justice. Our results, however, show quite clearly that males and females are virtually the same in the numbers who identify principled or justice-based reasoning. These results are also in line with recent meta-analyses indicating that gender is irrelevant as a factor in moral reasoning (Walker 1988). With all four demographic factors, only age turned out to be significant, and that for only one of the three groups, the graduate student sample. This finding may be an artifact of graduate school admission and a small n in that category. Usually, age after college does not bear a significant relation ship to ethical reasoning (Rest MORAL REASONING OF PUBLIC ADMINISTRATION 209 Table 11.2 Comparison of Principled Reasoning by Gender for Each Group Gender Students (%)Local Gov't. Mgr. (%) City/County Mgr. (%) (N=136) (N=190) Male 39 39 (N=95) 38 (N= 174) Female 42 39 (N=41) 39 (N=16) Table 11.3 Comparison of Principled Reasoning by Race for Each Group Students (%) Local Gov't. Mgr. (%) City/County Mgr. (%) Race (N=O) (N=136) Black 35 40 White 40 38 (N=186) *Race information was not available for the graduate student group.

Table 11.4 Comparison of Principled Reasoning by Education for Each Group Students (%) Local Gov't. Mgr. (%) City/County Mgr. (%) 1. Less than BA (N=136) 40 39 2. BA or more 40 39 36 3.

MA or more 40 40 Table 11.5 Comparison of Principled Reasoning by Age for Each Group Gender Students (%) Local Gov't. Mgr. (%) City/County Mgr. (%) (N=136) 25-30 35 37 38 37 39 38 41-50 44 40 37 51+ 57 39 41 *Difference significant at 0.0002 Eta 0.49 210 DEBRA W. STEWART AND NORMAN A. SPRINTHALL Table 11.6 Comparison of Stage Score by Level of Responsibility for Local Government Manager Sample and County/City Manager Sample Combined on Total SSMS Stage Score 1+2 3 4 5 Manager (%) 7 9 43 38 Ass't Mgr. (% 6 11 41 39 Level of Positio ) Dept. Head (N=52) 7 7 46 40 n Div. Head (%) (N=31) 6 7 46 36 Supervisor (%) 6 6 44 40 Note: Co option. lumns do not total 100 percent due to rounding error and meaningless response Table 11.7 Comparison of Mean Principled Reasoning Scores by Line vs. Staff for . Local Government Managers, City/County Managers, and All Managers Line Staff Local City/County Local City/County Gov't. Mgr. Mgr. All Mgrs. Gov't. Mgr. Mgr. All Mgrs. (N=150) (N=72) (N=40) (N=112) SSMS Level P Reasoning Functional and Organizational Responsibilities A second set of factors that were found to affectthe ethical thinking in organiza tions may be grouped under the rubric of functional and organizational responsi bility. While these variables may be operationalized in a variety of ways, this study looked at level of responsibility, line versus staff, and functional task as factors that might explain variation. Tables 11.6 and 11.7 display the findings for level of responsibility and line vs. staff.

Table 11.6 indicates that across levels of responsibility within organizations, a combination of the local government manager and the city/county manager sam ples, there is no expression of principled reasoning that can be attributed to level of responsibility in position. The modest differences that do appear seem unre lated to level of position. MORAL REASONING OF PUBLIC ADMINISTRATION 211 Table 11.7 suggests that there is no relationship between principled reasoning and line versus staff position in either of the sample groups of local government managers or city/county managers.

In a third exercise we simply reviewed the job titles of all respondents who indicated a propensity to select principled reasons (p = 50 or more). This inspec tion revealed no pattern that might be attributed to functional responsibility; that in making their decisions city managers, their assistants, planners, budget directors, and personnel managers appeared equally likely or unlikely to consider stage 5 reasons. Organizational Context Finally, we considered whether the context within which the decision maker was located would affect the likelihood of identifying principled reasons in resolving ethical dilemmas. The context factors that we considered relevant to local-level managers were the type of jurisdiction where they are located, the size of their jurisdiction, and whether or not their organization has a code of ethics. First we combined the two samples and considered all managers together.

As Table 11.8 the simple context of city vs. county vs. region clearly has no impact on the likelihood of selecting principled reasons in decision mak ing. The mean response is roughly equivalent across cities, counties, and regions.

Table 1 1.9 displays stage score response across small, m edium, and large cities and counties. There are no significant differences that can be attributed to size of jurisdiction, whether city or county. The final context factor was the existence of a code of ethics the jurisdiction .. The city/county manager survey provided data on that question.

Table 11.10 reports no significant difference in the mean response scores that can be attributed to whether or not a jurisdiction has a code of ethics. Difference in Ethical Domains The analysis thus far reveals no differences that can be attributed to commonly cited demographic, organizational, or contextual factors as the reasons selected by public managers for resolution of ethical quandaries. The data do reveal differences across ethical domains. Since statistically significant positive rela tionships hold between each of three stories and the total SSMS scores, the total SSMS score measures overall capacity to function at a particular stage or level.

However the strength of one's capacity to function at the principled reasoning level appears to vary across domains. A few recent empirical studies provide background on this point.

In the measurement of moral judgment there are often differences in levels of reasoning according to the actual dilemma content. Developmental psychologists usually refer to such differences as examples of cognitive décalage-systematic 212 DEBRA STEWART AND NORMAN A. SPRINTHALL Table 11.8 Comparison of Mean Principled Reasoning across Cities, Counties, and Regions City County Region SSMS (N=117) (N=4) Mean Principled 39% 38% 42% Reasoning Table 11.9 Comparison of Stage Scores across Size of Jurisdiction for Cities and Counties County City Greater Greater than Less than 25,000 than Less than 10,000 25,000 25,000 100,000 100,000 10,000 25,000 25,000 Stage % % % % % % Score (N-35) 1+2 6 8 7 6 5 7 3 13 8 8 7 7 11 4 42 42 43 45 45 41 5 37 38 38 40 39 38 Table 11.10 Mean Stage Scores on SSMS Comparing Jurisdictions with and without Codes of Ethics Yes Code (%) No Code (%) Stage Scores On SSMS (N=58) (N=30) 1+2 6 8 3 11 9 4 42 43 5 37 39 gaps in reasoning levels by problem areas (Kohlberg 1980). An important deter mining factor appears to be how close to real experience is the dilemma content as well the extent to which there are known and articulated positions. Thus, MORAL REASONING OF PUBUC ADMINISTRATION 213 when adult subjects were confronted with real-life issues as opposed to third party hypothetical problems, the reasoning level declined (Walker, deVries, and Trevethan 1987). Similarly, in a study with adolescents the reasoning level declined when the content of the dilemmas shifted from abstract ques tions such as, Should a person in a foreign country steal a drug to save his spouse? to everyday issues such as obedience to one's parents versus loyalty to one's own peer group (Gilligan et al. 1971). In the latter case, the level of reasoning declined by almost one stage for both male and female subjects.

Rest (1986) has noted that the ability to identify principled reasons involves a complex process of conlprehension, awareness of consequences, selection of courses of action that are consistent with such principles, and enough per sonal strength to withstand criticism. Such factors develop as a result of experience in dealing with controversial issues and in taking action. This means experience dealing with issues that have not been rehearsed or hashed out and that are personally invol ving-that is, close to real life-and that levels employed will be lower than with experiences assessed through stan dard or more familiar dilemmas.

There is another source of possible domain discrepancy. Jennifer Hochschild (1981) found that different beliefs about distributive justice prevail in different domains of life. Her in-depth interviews with a sample of rich and poor respon dents in the Standard Metropolitan Statistical Area of New Haven showed that people may use egalitarian norms when they address issues in the-socializing or political domain, but they shift to differentiating no rms in the economic domain (Hochschild 1981, 48,49, 82). Thus they come to different conclusions in differ ent domains because they are applying different principles of justice (Hochschild 1981). This type of discrepancy could only be assessed in our case through the creation of an additional set of principled reasons based on different justice concepts. Thus far, the enormous number of studies in the moral development distributive justice format does provide us with a ready source of comparison in that general moral domain. That data base (Kuhmerker 1991) continues to grow both within this country and cross-culturally. This means that the differences we find in levels of moral development on the specific issues faced by professionals in the field of public administration can be compared to general levels in the adult population and with appropriate cultural translations to public administra tors in other countries as we ll. 1 In the study reported in this paper we can say that there appears to be a shift in the preference for a principled mode of reasoning across the three stories posed in the SSMS instrument, from over 50 percent stage scores for principled reasoning across all three samples for the first story, to less than 30 percent selection of principled responses across each of the three samples for the third story. Table 11.11 displays these data. We believe this suggests that there are different ethical decision domains and that some domains are more likely to elicit principled reasoning than others. 214 DEBRA W. STEWART AND NORMAN A. SPRINTHALL Table11.11 Comparison within Stories and across Groups on Stage Score Students Local Gov't Mgr. (%) City/County Mgr. (%) (N=136) (N=190) Stage Scores Story 1 Story 2 Story 3 Story 1 Story 2 Story 3 Story 1 Story 2 Sto ry 3 1+2 2 7 7 2 6 9 2 6 10 3 8 7 3 5 12 2 8 17 2 4 36 45 58 39 39 62 35 33 59 5 52 42 28 52 43 22 51 42 20 The three stories are as follows:

Story 1: Promotion Bob was hired to revitalize a somewhat lackluster division in a state agency.

Soon after becoming division he held a meeting with all division personnel and announced that all future promotions would be based on demon strated merit and affirmative action. The patronage practice of the former director would be discontinued. Bob issued a written statement to confirm this new policy.

About a month later Bob ~s boss told him that he expected Joe J an individual on Bob's staff, to be promoted. Joe was a marginally effective white male and there were several other employees in the division much more de serving of promotion. Bob pointed out his boss all of the reasons for not wanting to promote Joe Jones at that time. But the boss responded that he really would like to see Joe promoted and that Bob's ability to create more promotional opportunities for his staff in the future (new positions, successful job reclassification, etc.) depended on his cooperation in this situation.

Story 2: Friends in Government In our agency, as in many others, private vendors are hired provide goods and services. I deal directly with these representatives and have become good friends with one representative. Our wives have become friends and our fami lies enjoy one another. I occasionally join him for lunch and he picks up the check as a company expense. This has always been acceptable in our organiza tion. However, a dilemma arose recently when my friend invited my family and me to join his family and other friends at his beach cottage for a week. He was going to pay for the food and drink, and since the others were private uld be a great trip an en by the press as a p to my compa his firm was going to write it pff. We knew it wo d we really wanted to go. I also knew that it could be se ayoff for a large contract that my agency had just awarded ny. MORAL REASONING OF PUBLIC ADMINISTRATION 215 Story 3: Data Re-creation Jack heads the Management Information System section for his agency.

Through no fault of Jack's some benefit payment data were accidentally de leted from the agency computer file. There was no way to retrieve these data the system. The local governments charged with inputting these data originally complained bitterly about the burden of this task, so the agency's top management wanted to keep the recent data loss quiet. Top management told Jack to devise a scheme to recreate the data based on the assumption that certain relationships existed between the data elements. But Jack argued that this would result in some people receiving more benefits than they should and others receiving fewer. Top management in the agency felt that to meet pay ment deadlines of local government, there was no choice. Jack was told to re-create the file as best he could.

Respondents are asked to read the stories and decide what plan of action they would follow. Then they are asked to review a list of possible considerations and indicate which they would consider most important in the situation.

Clearly, the domain does make a difference across all three samples. While our conclusions on this point are preliminary, we suggest that the difference can be explained in terms of familiarity with the domain and depth of individual experience in resolving ethical quandaries in the area .. The promotion story raises issues that have been under intensive discussion in public administration over the past twenty years. The competition between "merit" and "equity" claims is a standard topic in all public-sector arenas from the formal classroom or training session to the office coffee klatsch. More important, it has been a topic on the agenda of society at large for more than two decades. Accordingly, we find mean principled scores above 50 percent for all three groups.

The "Friends in Government" dilemma is by contrast a less intensively dis cussed social issue for the general public.

There has been significant public media attention focused on conflict-of-interest issues regarding governmental procurement and other practices; however, the intensity of the public debate pales by comparison with the affirmative action-merit controversy. Still, within public agencies serious attention is given to the issue of conflict of interest through agency training programs, written policies, and simple admonitions from agency officials. It is reasonable to expect that respondents in this study have strong familiarity with the issue and have had opportunities to discuss and to reflect on a dilemma similar to the one raised in "Friends in Government .." Their scores on principled reasoning of 47 percent, 43 percent, and 42 percent, respec tively, across the three groups support this assumption.

The third story, "Data Re-creation," is the least familiar of the three. The problem posed, that of simulating a large and lost data set, is a relatively new problem in the annals of public administration. The problem is occasioned by a technology known, but probably not well understood, by most respondents. 216 DEBRA W. STEWART AND NORMAN A. SPRINTHALL Clearly this is not a topic discussed by the broader society. Rarely would such a dilemma be encountered by public managers. The specialized character of this problem makes the attendant ethics an unlikely topic for discussion in agency training seminars. In short, in identifying issues in this story, respondents are left without the benefit of discussions, either within the agency or in society, that would provide the background for reflection and decision. Hence we find across all three groups that the mean score on principled reasoning was at the highest (28 percent) in the student group, followed by 22 percent and 20 percent for the other groups. It may be unsettling to note that the average percentage of princi pled reasoning on this dilemma is only slightly higher than the average reasoning of junior high school students on hypothetical dilemmas (Rest 1986).

At the construct validity level, we can say that all the groups, regardless of position, comprehended the sets of reasons in a similar manner across all three situations. The rank order by principled level was always the same, with Story 1 receiving the highest percent of principled responses and Story 3 the lowest percent across nine comparisons. This means that the subjects were reading the story choices in a consistent manner according to level of understanding and familiarity. At a theory construct level it is noteworthy that the second modal stage was always in juxtaposition to the first modal stage. Thus in Story 1, stage 5 was the predominant stage selected, over 50 percent of the time. For the same story, stage 4 was always the second most common choice. Similarly in Story 3, stage 4 was the predominant mode while stage 5 was second.

These findings also indicate that the subjects remained consistent in their ability to select theoretically coherent levels as opposed to random choices.

Behavioral science research in moral/ethical development has shown that a base-line criterion is the ability of an instrument to assess levels of reasoning in an invariant sequence (Kohlberg 1980).The subjects should select reasons that represent their general mode at one particular level and the next level will always be at plus or minus one. The only exception to the overall invari ant sequence in our study is with the lowest stages , for example the combined stage 1 and stage 2 responses, for specific stories. On an overall basis how ever, for all subjects on all three stories, the sequence is invariant (Table 11.1) with the peak at stage 4. These results do support the theoretical valid ity of stage and sequence concepts with empirical outcomes from our sam ples. The reliability of the instrument can be supported by the fact that the measure has been used with four different samples (two separate administra tions for graduate students and two for the groups of public administrators) .. In each case the rank order of story choice for the combined stories has always been the same, which indicates that the concurrent reliability is strong.

We have not yet estimated the stability of the instrument over time. Other re search with the general objective measure of moral/ethical judgment, upon which our instrument was based, has shown test/retest correlations in the +0 .70 range (Rest 1986). MORAL REASONING OF PUBLIC ADMINISTRATION 217 Conclusions It should be noted that surveys such as ours measure the ability to identify levels of ethical reasons and do not assess actual moral behavior in real-world situa tions. Other research (Blasi 1980), however, has shown a consistent behavioral relationship between thought and action. It is certainly clear that if persons cannot process issues at a principled level or cannot recognize democratic princi ples, such persons are much less likely to act in accord with ethical principles .. The ability to recognize such issues is a necessary first step in the process of ethical behavior (Rest 1986). In this sense, a first conclusion from our study indicates that the greatest variability in level of reasoning is derived from the content of the problem situation. If the content is familiar and there has been considerable discussion and analysis of the issues, there is a greater likelihood of higher stage reasoning. The opposite is also true. In unfamiliar situations where little has been discussed or processed, individuals are highly likely to employ less democratic and more self-serving reasons.

A second conclusion is that the usual factors of demography and organiza tional context have almost no influence on the level of moral reasoning. Not withstanding the need for public executives to manifest a broader ethical vision that recognizes interconnectedness and that operates on a longer time horizon (Luke 1991), public executives in this study are no more able to identify princi pled reasoning than mid-level or first-line managers. Perhaps more ironically, the existence of ethical codes provides no influence on the ability to identify ethical reasons. These findings indicate that ethical reasoning and public administration are dynamic and interactive. The levels are independent of status and organiza tion context. Those variables obviously play an important role in many areas of administration but not when focused on issues of ethics.

A third conclusion is that there may be a need to reexamine both graduate and professional in-service education. Development of extensive codes of ethics ap pears to be a sterile enterprise. It may be more profitable to consider dialogue on issues of ethical controversy both during graduate school and throughout profes sional practice. Certainly it is no easy task to learn to identify the ethical prem ises in difficult situations, and clearly it cannot be learned through the lecture method. Open discussion and challenge, on the other hand, can facilitate the growth process even in such a complex area as ethical reasoning (Thoma 1984).

A fourth conclusion points to future research. all respondents in the three studies were drawn from North Carolina, and since we know there may be regional differences associated with levels of moral reasoning, more research needs to be done across regions to identify possible differences. In addition, it is important to understand more about the relationship between stage score re sponses and actual behavior in public administrator roles. Finally, more research is needed to understand the most effective methods of ethics training. With the use of SSMS as our starting point, this will fo rm our continuing research agenda. 218 DEBRA W. STEWART AND NORMAN A. SPRINTHALL Note 1. We are currently piloting the SSMS with a small group of public administrators in several districts in Warsaw, Poland. The project ultimately entails comparison of local level public administrators in the United States and Poland. References Blasi, A. 1980. "Bridging Moral Cognition and Moral Action." Psychological Bulletin 88: 1-45.

Conner, P.E., and Becker, B.W. 1975. "Values and the Organization: Suggestions for Research." Academy of Management Journal 18: 550-61.

Denhardt, K.G. 1988. The Ethics of the Public Services: Resolving Moral Dilemmas in Public Organizations. New York: Greenwood.

Gilligan C. 1982. In a Different Voice. Cambridge MA: Harvard University Press.

Gilligan, C.; Kohlberg, L.; Lerner, M.; and Belenky, M. 1971. "Moral Reasoning about Sexual Dilemmas." U.S. Commission on Obscenity and Pornography. Washington, DC: U. S. Government Printing Office.

Harris, J. 1989. "Ethical Values and Decision Processes of Business Students and Busi ness Practitioners: A Review and Extension." Unpublished manuscript, cited in J.

Harris, 1990, "Ethical Values of Individuals at Different Levels in the Organizational Hierarchy of a Single Firm." Journal of Business Ethics 9: 741-50.

Hochschild, Jennifer L. 1981. What's Fair? American Beliefs about Distributive Justice. Cambridge, MA: Harvard University Press.

Hodgkinson, C. 1971. "Organizational Influence on Value Systems." Educational Ad ministration Quarterly 7: 46-55. Hunt, S.D.; Wood, V.R.; and Chonko, L.B. 1989. "Corporate Ethical Values and Organi zational Commitment in Marketing." Journal of Marketing 53: 79-90. Kohlberg, L. 1980 . The Meaning and Measurement of Moral Development. Worcester, MA: Clark University Press. ---. 1984. Essays on Moral Development. Vol 2. New York: Harper and Row.

Kuhmerker, L. 1991. The Kohlberg Legacy. Birmingham, AL: REP Books.

Luke, J. S. 1991. "New Leadership Requirements for Public Managers: From Managerial to Policy Ethics." In Ethical Frontiers in Public Management, edited by J. S. Bowman, 158-82. San Francisco: Jossey-Bass.

McDonald, R.A., and Victor, B. 1988. "Towards the Integration of Individual and Moral Agencies." Business and Professional Ethics Journal 7: 103-18.

Posner, B.2., and Schmidt, W.H. 1984. "Values of the American Manager: An Update." California Management Review 16: 202-16.

Purcell, T.V. 1977. "Do Courses in Business Ethics Pay Off?" California Management Review 19: 50-58. Rest, J. 1986. Moral Development: Advances in Research and Theory, 2d edition. New York: Praeger.

Sapp, G.L. 1985. Moral Development: Modes, Processes and Techniques. Birmingham, AL: REP Books. D. 1985. "Ethics and the Profession of Public Administration: The Moral Responsi bility ofIndividuals in Public Sector Organizations." Public Administration Quarterly 8: 4. Stewart, D., and Sprinthall, N. 1991. "Strengthening Ethical Judgement in Public Ad ministration." In Ethical Frontiers in Public Management, edited by J. S. Bowman, 243-60. San Francisco: Jossey-Bass. MORAL REASONING OF PUBLIC ADMINISTRATION 219 Thoma, S. 1984. "Do Moral Education Programs Facilitate Moral Judgement?" Moral Education Forum 9 (4): 20-25. Trevino, L. K. 1986. "Ethical Decision Making in Organizations: A Person-Situation Interactionist Model." Academy of Management Review 11: 601-17. Walker L. 1988. "The Development of Moral Reasoning." Annals of Child Development 5: 33-78. Walker, L.J.; deVries, B.; and Trevethan, S.D. 1987. "Moral Stages and Moral Orienta tions in Real-Life and Hypothetical Dilemmas." Child Development 58: 842-59.