Columns

TheOmbudsman’sRole ina DisputeResolutionSystem

Mary P Rowe

c ontempotm-y theory and practice suggest that organizations EditorJs Note: In alternating issues, should design and build dispute reso- A’egotiation Journal features a regu- hnion systems—rather than just one or lar column on the subject of’ ‘dispute another structure— systems design,” a concept initially in circumstmces where people wiU be proposed by William L. Ury, Jeanne working together or dealing with each M. Brett, and Stephen B. Goldberg in other over time. Reviewrof the success their 1988 book, Getting Disputes of (proliferating) ombuds offices’ sug- Resoived: Designing Systems to Cut gests that this kind of office is both a the Costs of Conj7ict (San Francisco: desitable and cost-effective element in Jossey Bass). Brett and Ury are serv- an efficient dispute resolution system. ing as coordinators of this column, This column focuses on the ombuds- which is aimed at serving as a forum man who works within an organiza- for the ongoing exchange of ideas tion. Much of the discussion, however, about dispute systems design. is equally appropriate for ombudsmen who serve clients such as citizens, stu- dents, newspaper readers, patients, ator, informal fact-finder, upward- vendors, taxpayers, etc. feedback mechanism, consultant, I define an internal ombudsman as problem prevention device and a neutral or impartial manager within change agent; and whose office is an organization, who may provide located outside ordinary line manage- informal and confidential assistance ment structures. to managers and employees in resolv- An often-quoted sentence about ing work-related concerns; who may ombudsmen states that “ombudsmen serve as a counselor, informal go- may not make or change or set aside between and facilitator, formal medi- a law or policy; theirs is the power of

Mary P. Rowe, an ombuds practitioner for the past 19 years, is Special Assistant to the Presi- dent of the Massachusetts Institute of Technology, as well as Adjunct Professor at MIT’s Sloan School of Management. She co-founded the Corporate Ombudsman Association. Her mailing address is: MIT 10-213, Cambridge, Mass. 02139.

074W 526wI/IoM-0353$ 6 50/o G 1991 Plenum Negotiation JoumaI October 1991 353 reason and of persuasion.” Ombuds- managers have, especially when men thus have all the functions of any things are going wrong. complaint-handler except that of for- ● Consultation to help people help mal fact-finder, judge or arbitrator. ~ themselves. This can involve such Ombudsmen do not “deliver due practices as counseling with process” in the sense of a court sys- empIoyees and managers; invent- tem.4 They encoumge practices that ing new options; listing all possi- are fair and just and respectful. They ble options for the choice of the work to foster whatever responsible person(s) with a problem; consult- process is “due under the circum- ing and coaching on how a per- stances”; (in the ideal situation, this son or group may deal with the process is one chosen, or at least problem directly (’problem solving, agreed to, by the parties). role-playing, teaching Functions and Characteristics skills, anticipating possible out- of Dispute Resolution Systems comes, etc.); or helping review the strengths and weaknesses of previ- An effective dispute system includes all ous dispute resolution efforts. This of the following functions: is the function that helps to define ● Expressing respect for feelings what process is “due under the (especially rage, fear of retaliation circumstances.” and grief). HeIping people deal ● Shuttle diplomacy by a third with their feelings—so they will party. be able to make responsible deci- In this process, the third party “shuttles” back and forth sions and be able to deal effec- tively with their problems or among those with a problem, to resolve the matter at hand (some- complaints—may be the most cost-effective element of a dispute times called “” or resolution system. This is true in “caucusing’ ‘). Shuttle diplomaq part because providing respect and and both may include dealing with feelings cost very lit- offering advice as to what may happen if informal problem solv- tle. It is also because respect is a parent of productive work rela- ing fails, including advisory ; tions, and because humiliation is a parent of destructive behavior. In ● Mediation. The settlements of my experience, this is the function mediation may be formal or infor- most likely to fail in a dispute reso- mal, and on file or off-the-record. lution system; in the language of The key element of this function Total Quality Management (TQM), is a neutral third party who brings the system fails where ‘‘cus- together the people or groups tomers” have been ignored. with a problem, so they reach ● Giving and receiving informat- their own settlement or are helped ion on a one-to-one basis (mak- by the third party to reach their ing referrals, telling people how own settlement. the system works, receiving whist- ● Fact-finding or investigation, leblowing complaints, etc.). Many This may be done either formally people overestimate how much or informally. Reports may be made information disputants have. either with or without recommend- Nearly everyone overestimates ations from the fact-finder to one how much information top or more decision makers. ● Decision making arbitration or good faith; employees are not adjudication. h this process, a punished for mising questions or person or body with power and/or for responsible disputing. formal authority decides a dispute. ● i%e system provides options— This may be structured as part of and choice—for pwwing most a formal compIaint-and-appeals complaints. The system allows channel or formal grievance disputants to have as much choice procedure As Ury, Brett and GoId- as possible, rather than requiring berg (1988) have pointed out, it is that a given problem may by pur- often useful to consider a variety sued in only one way. This respect of mechanisms to provide rights- for the “customers,” now immor- bmed and power-based decisions.5 talized by TQM, is particularly ● Upu’ard feedback, problem likely to be ignored by people prevention, and systemschange. who think they know best what This can involve designing a disputants need. generic address to a problem, or ● Zbe system provides loopsback, a single complaint, or a pattern of from adjudicative options to dispute; fostering change in poli- problem-solving options, and also cies, procedures, or structures as loopsfomard, so that most peo- a result of inquiry, suggestion, ple with problems can at any stage complaint or dispute, or an evalu- choose investigation and adjudica- ation of the handling of a previ- tion of their complaints, so long ous dispute; or providing group as thq do so in good faith.6 training in dispute resolution skills. ● The system is available to every- An effective internal dispute resolu- one, managers and employees tion system also has the following alike, for every type of problem. chamcteristics: ● lhe system provides in-house., ● i%e s>stem is taken seriously. It designated neutraLs, who: help has strong support from top people deal with the system; legiti- management. It is widely publi- mate the asking of questions and cized. Managers and employees raising of concerns; minimize hear discussion and receive some retaliation against those who com- tmining in . The plain; provide consultation on system reports back aggregate options; review how conflicts have statistics about problems and dis- been handled in the past (espe- putes to top management and the cially patterns of conflict); are alert community, as an integral part of for new problems, as well as avail- the organizaion’s management able for bizarre, delicate, distastef- information system. ul or frightening problems; ● The system provides signz~icant provide individualized coaching evid~zce of change (including on negotiation skills, and (where reversal of some management appropriate) keep disputants decisions) as a result of complaints focused on interests and on cost- and disputes. effective modes of disputing.

● Policies against retaliation are ● 7he systemprovides, z~possiblq taken seriously by all. Managem more than one available neutral, typically are not punished for so that people with problems have reve~als of decisions they made in a choice. Ideally, people should

Negotiation ]oumal October 1991 355 have the choice of dealing with an blow off steam, or fmd out a fact or impartial complaint-handler or two, or to learn how to help them- mediator of the same gender and selves. g In these common cases, the mce, m this makes it much more ombudsperson may be the only likely that different kinds of peo- complaint-handler, and also does not ple will feel welcomed and intervene. respectfully treated by the dispute Many workplaces also have other resolution system. Providing more offices where people may go to than one neutral or impartial per- express or sort out their feelings off son also helps in cases where the the record, give or receive information first such person is no longer on a contldential basis, or develop and appropriate or available, and choose effective options. These where there is a wide variety of include sensitive supervisors, employee disputes requiring a variety of assistance, equal opportunity officers, skills. human tesources penonnel, the appro- ● 7%e systemguarantees confiden- priate medical department, religious tiality to all who approach an in- counselors, student affairs deans, etc. housel designated neutml off the Ombudspeople quite regularly refer record (e.g., for consultation and visitors to such offices, and receive counseling), except in the rare referrals from these colleagues, as all case where there is a duty to pro- these practitioners seek to build an tect (i.e., a danger to self or othem). effective support network for those The pmctice and perception of who are raising concerns. near absolute cofildentiality is Ombudspeople may also intervene essential to building trust in a sys- as third parties. They are sometimes tem that is going to handle dehcate asked to pursue shuttle diplomacy and difficult disputes.’ between peers, and it is common for an ombuds practitioner to be asked to Where the Ombudsman Fits go back and forth between the person in a Dispute Resolution System with a concern and hi? or her super- An ombuds ol%ce may be seen by itself visor. .Many ombudspeople are medi- as a mini-system, since the internal ators. (Formal mediation is more ombuds pmctitioner has all the func- common betxveen peers than between tions of any complaint handler except supewisor and supervise within a that of formal decision maker, investi- workplace.) Here again, most work- gator or arbitrator. In addition, the places also have other people who ombuds practitioner typically works serve these functions: skilled supervi- closely with supemison and with sors, human resource managers, deans other dispute resolution structures in academic settings, and outside con- within an organization. sultants. Ombudspeople make and An internal ombudsperson is often accept referrals to and from these the first person approached for other helping resources. difficult problems within a given Informal investigation is a common workplace. In these cases, the ombuds function for an ombudsman. Fre- office may be the point of entry into quently the practitioner will get per- the system rather than the only per- mission from a visitor to look into and son of contact. However, many pursue a concern. This often entails an managers and employees who seek informal inquiry. Thereafter the out an ombudsperson come in just to ombudsman may make informal recommendations to a decision maker management consultants, miners, and and or lobby quite stubbornly for change agents. This may occur in change in policy. h is, however, rare for many ways, Sometimes the best way an ombudsman to be asked to do a to deal with a specitlc problem is formal investigation in a formal through a generic response, where the grievance process, and many ombuds- ombuds practitioner will be working people will not do so. (The common with the relmmt line manager or per- belief that ombudspeople are formal sonnel specialist.9 Sometimes the investigators applies more appropri- ombudsman will be called to conduct ately to classic public ombudsmen tmining programs on conflict manage- than to internal practitioners.) ment or negotiation skills, for people Informal and formal investigation or groups that will be working are functions also shared with labor- together. relations, human resource personnel, student affairs administrators, active The Ombudsman’s supervisors, and some other special- Sources of Power ized pe~onnel such as safety, equal Because ombudspeople have no line opportunity, security and audit profes- authority, people often presume that sionals. As noted earlier, it is common they “have no power.” This assertion, for referrals to come to the ombuds however, reflects a misunderstanding office and be made from the ombuds of the sources of power in negotiation. office to these colleagues. In particu- Following are some commonly recog- lar, an ombudsman who is the nized sources of power and the extent recipient of a whistleblowing report to which they are helpful to the will likely be working with line ombudsperson: managers and/or other staff offices to s Legitimate authority. Most inter- see the matter properly referred to nal ombudsmen do not have line appropriate persons. management power. Those few In some workplaces, ombudsmen who are empowetvd to make bind- are so much a symbol of “interest- ing decisions typically choose not based” dispute resolution that some to do so V~ Often, choosing mther people presume that these practi- to aff~m the responsibility and tioner function mainly as a loopback rights of line authority and of dis- process from adjudication to problem putants. (The rare ombudspeople solving, Looping back is, in fact, com- who occasionally do formal find- mon. However, most ombudsmen also ings of fact often do not provide facibte and support looping forward formal recommendations for (to rights-based, formal investigation future action for just this remon.) and adjudication) on important (if s Reulards. While internal ombuds- uncommon) occasions where this is people do not set mises or promo- the option responsibly chosen by a tions, their affirmations of good visitor. (Some conflicts need a win/lose management and productive response.) Ombudsmen also may serve behavior often serve to illuminate as nonvoting managms of a peer review excellence in the workplace. process and in other ways support for- Ombudspeople commend as well mal complaint and appeals channels. as criticize; commendations are Research indicates that internal often seen as (‘rewards,” and pro- ombudsmen typically spend a quamer vide considerable power as well as to a third of their time as internal entree.

;Vegotiation Journal October 1991 357 ● Sanctions. Ombudspeople obvi- FaZlback position or BATNA. 10 ously illuminate bad behavior as The BATNA of an ombudsman is well as good, raising the concern usually to turn over the dispute, or of sanctions from authorities. The let it devolve into, the next pos- fear of sanctions is a potent source sible mode of resolution: line of ombudsman influence. supervisors, formal grievance ● Force. The fact that other people mechanisms, the courts, letting the may use force (sabotage, violence, disputants quit the workplace, etc. work stoppages, etc.) provides This is often a very useful source power to alternative dispute of power since frequently dispu- resolvm, including ombudspeop[e. tants think that all alternatives are worse than dealing with the ● Moral autborit~ charisma. Obvi- ombudsman. ously the idea of an ombuds office is to affirm that which is just and Relationships. The professional fair; the office therefore has strong relationships of the ombudsman mod authority. In addition, most are typically an important source ombudspeople are chosen in part of power. In particular, most for charisma and/or reputation. ombudspeople work for the CEO or other very senior manager, and ● Commitment. Stubbornness, and many pmctitioners are old friends a resolve never to give up on a of senior managets These intan- problem until it appears to be gible points am widely considettd, resolved, are qualities much by ombudsmen themselves, to be needed by pmctitioners. These major sources of power for prac- qualities can be a major source of titioners. (In addition, it is perhaps power, in continuing to mise ques- easier to be an ombudsman than tions with recalcitrant managets, to be in other areas of senior in seeking systems change, and in management, in terms of not mak- “staying power” with disputants ing enemies. Although many peo- in mediation. ple think that it must be hard not ● Information and expmise. These to make enemies as an ombuds- classic sources of power are man, in fact most people in a usually available to an ombuds- given workplace appear to under- man, who typically has access to stand the peculiar charge given to every database in the organization, the pmctitioner. If people come to and who knows m well as anyone learn that the practitioner keeps “how to make something work” near absolute confidence, is in fact in the given workplace. neutral and personally disin- ● Elegant solutions. Since the terested, they are usually gracious ombuds practitioner is personally and respectful to their unusual col- disinterested, committed wherever league.) appropriate to integrative solu- tions, has information about Cost-Effectiveness interests on all sides of a dispute, Corporate Ombudsman Association has the luxury of concentrating on research, based on information dispute resolution, and is unlikely provided by practitioners, indicates to lose interest (or composure), he that internal ombudsman offices may or she can sometimes find a be quite cost-effective (Rowe and Per- reasonably elegant solution. neski, 1990). Preliminary estimates indicate cost-effectiveness ratios in cor- save money for the employer (in pomte workplaces between 1:2 and addition to the increases in 1:6. The cost savings estimated by productivity that one hopes take practitioners include such items as place when disputes are ttsolved). . providing alternatives to some litiga- “ Ombudsmen often fill in for parts tion (for exampIe wrongfd terminat- of a dispute resolution system that ion suits); averting or dealing are not functioning well, as fail- prompt[y with some harassment, safe, back-up, check and balance. fraud, theft, and other unethical Moreover, these practitione~ can behavior; preventing or dealing earIy focus precisely on the dispute with some threats, safety problems, resolution element that is failing. sabotage, and potential violence; reten- In particular, the ombuds practi- tion of some highly valued profes- tioner can sometimes alleviate the sionals who would otherwise leave. damage done when someone feeIs Here are some hypotheses as to why humiliated, enraged or afraid. 1z ombuds offices may be effective: The practitioner also may be in a “ Because the exkence of an ombuds position to provide a crucial bit of office Legitimates the idea that it is information, or infusion of prob- acceptable to raise questions (even lem-solving skills, to help dissolve small questions) and because there a dispute. In addition, the ombuds- is almost no costl 1to contacting an man sometimes fills in where an ombudsman, people with ques- established complaint procedure tions and problems often come in may not be helpful, as with union early, when most disputes are mom worker-to-worker problems. easily resolved. Ombuds offices ● Ombudsman offices can help dis- are especially useful with respect putants choose an option which is to whistleblowing. In the course “right” for them. My research of informaI problem solving, (Rowe, 1990a and forthcoming) ombudspeople almost never iden- indicates that people with difficult tify a visitor or caIler without per- problems often have very firmly mission. They are, therefore, often held—and disparate-ideas about in a position to act as a buffer for dispute resolution. Thus the a legitimate whistleblower, and can chance to choose or custom-tailor talk with that person to ask details an option is likely to be appealing useful to management in address- to a complainant. Moreover, an ing the complaint. Ombuds offices ombudsman may be able to help thus can be effective in surfacing fashion unusual remedies (even if unethical behavior (and in reassur- sometimes quite small remedies) ing callers whose concerns turn which exactly fit unusual circum- out not to be serious). stances and the~fore are relatively ● In addition, if it is appropriate to pleasing to one or more party. In support disputants to choose an my experience, finding the (‘right” interest-based or low-cost rights or option for a complainant makes it power-based approach, the less likely that a complainant will ombuds practitioner will usually reject a solution in a costly try to do SO. fmhion.

“ Many suggestions that come to an ● Ombuds offices are widely sought ombudsperson directly make or out. 13 As is the case with most

Negotiation Journal Cktober 1991 359 forms of mediation, many people managers and employees to who see an ombudsman appear to employee assistance or medical be reasonably satisfied by the help, for people who have not yet chance to get a problem exunined quite agreed to go to seek support or resolved, and to learn new and help. skills. These former “customers” “ Ombuds practitioner work in a send in new people. There is then low-key, usually evolutionary a widening pool of people who fashion, for steady systems change practice and teach others their to meet changing needs. (In fact, new negotiations and problem- a few ombudsmen deal solely solving skills. It is also quite com- with systems problems.) This ele- mon for people to seek consulta- ment of legitimating disputes and tion on problems (before a dispute problem prevention is hard to has taken place), after working evaluate in economic terms, but is with an ombudsman. thought by ombudsmen them- ● Ombudsmen provide low-cost selves to be an important element data collection, by tmcking their of effectiveness. caseloads and running surveys. A particularly importmt data collec- Conclusion tion function is that of identifying In the terms of Ury, Brett, and Gold- and reporting problems that are berg, ombuds practitioners can help to new to the o~-tion, for which provide “motivation, resources, and appropriate policies and proce- skills” for continuous problem solving dures do not yet exist. 14Another in times of change, within a dispute is the ability to collect and put resolution system. Ombuds offices together little pieces of dan from help to foster interest-based solutions. many sources, or complaints from They can help disputants to loop back- disparate areas, about the same ward or loop forward, where such person or service. actions are appropriate. In the lan- “ Ombudsmen help to deal with guage of Total Quality Management, peculiar, delicate questions, with ombudsmanry is focused on the needs people whom others find to be of the “customers” (that is, the pemons difficult, and with cross-cultural involved in dispute), in particular by issues,15Pmxitioners often become providing respect and by providing reasonably adept at understanding options. In human terms, ombuds and surfacing hidden agendas, offices appear to be widely used especially from “chronic com- where they have appeared, thus plainers.” Ombuds offices are one indicating some effectiveness of useful path for making appropri- response to the needs of people in ate referrals-for example, to get conflict.

360 Ma?y P Rowe Tbe 0rnbu4man % Roie NOTES

1. There are many kinds of dispute resolution practitioners in North America who are called ombudsmen. These include the “pure” ombudsmen who are appointed and paid outside the arem over which they have oversigh~. In the classical czse, they are appointed by a legislative body to have oversight over actions of the executive branch of government. There are also many other kinds of “client” ombudsmen, for example, those who serve newspaper readers, hospital and nursing home patients, students in educational settings, defense department vendors, bank and insurance company clients. There are, in addition, some thousands of’ ‘internal” ombudsmen, who serve employees and managers within companies, universities, government agencies, foundations, etc. Ziegenfuss has written two books on ombudsmen, cast in somewhat different terms but along the same lines m this article (Ziegenfuss, 1985 and 1988), See also Anderson and Stockton, 1990 for the Adrninistaive Conference report recommending ombuds offices in the Federal Government. 2. There is no commonly accepted version of the word ombudsman, Many people say ombuds- person, ombud, ombuds practitioner, etc. 3. There is probably no rule about internal ombudspeople that is true for all such practi- tioners and this stxement is an example of a rule with exceptions. For instance, a few internal ombudspeople are empowered to undertake occasional formal investigations and/or make occa- sional management decisions if problem solving fails. 4. There are two common meanings for the concept of “due process,” The first is a set of elements of proper process in formal investigation and adjudication, such as rights to timeliness of procedures, to know and be able to respond to the charges made against oneself, to represen- tation by counsel, The other common meaning is simply ‘‘the process that is due under the cir- cumstances,” Many ombudsmen will work, if ~sked, to see that people get the rights that are due them in formal internal grievance processes; however, it is the second meaning of due process that better characterizes ombuds practitioners. 5. This list of functions includes several impommt points made by Ury, Brett and Goldberg (1988). Among them is the importance of providing low-cost alternatives to strikes, court action, sabotage and the like. Ury et al. suggest alternatives most relevant to collective bargaining situa- tions. I would add to their list, for nonunion adjudication, peer review and other similar mechan- isms. Three, fine, recent books that discuss such internal grievance procedures are: Ewing, 1989; McCabe, 1988; and Westin and Feliu, 1988. 6. Ury, Brett and Goldberg (1988) have given the name loopback to the process whereby a dispute can be taken from a rights-based, adjudicative, “distributive” process, to an interests- based, problem-solving, more ‘‘integmtive” process. My own research indicates that a small propor- tion of the population is only comfortable with and satisfied by adjudicative processes, espe- cially for problems like harassment and discrimination (see Rowe 1990a). I therefore argue that loopsfonuard are also an important chamcteristic of art effective internal dispute resolution sys- tem, and that people with problems should not necessarily be required to go through all the steps of a grievance procedure for every type of problem, 7. My research over the pa-w 19 years indicates that an employer must choose between: (1) guaranteeing near-absolute confidentiality [and the choice of the complainant about whether and how to pursue a complaint], which will produce a relatively high reporting rate of complaints and concerns; and (2) no effective confidentiality [and therefore no reliable choice for complainants about what will hap- pen], and a much lower rate of reports. Tttis is especially true for very costly and dit%cult problems such as safety, ethics, harassment, misconduct, etc. (Rowe, 1990a). Whether an ombudsman can be subpoenaed and forced to testify, and thus break corttidentiality, is a topic now being tested in various ways, but there is an emerging professional consensus that ombuds practitioners must not break con- fidentiality. A few courts have upheld this principle for ombudsmen as they have for other kinds of mediators (Rowe, Simon and Bensinger, 1990), and the Admirtistrative Dispute Resolution Act of 1990 also provides strong protection for neutrals in federal agencies. 8. Please see Rowe, 1990b for a discussion on helping people help themselves as an- ADR technique. 9. AS an e=mple, if a person who complains of racial or sexual harassment does not want to come forward personally but asks that the alleged offender be trained and/or warned about harassment, an ombudsperson may go to a department head or personnel manager to arrange for generic responses to the complaint (for example, a training program in the department and a letter from the head to every member of the department).

Negotiation Journal October 1991 361 10. BATNA an acronym for Best Alternative to a Negotiated Agreement, is a concept ihmi- nated by Fisher and Ury (1981). 11.As mentioned slier, most ombudsmen now refuse to testify in formal grievance processes, thus underscoring the near-absolute cofildentiality of the office. 12. In addition to alleviating some great emotional anguish, in my opinion this is the func- tion of an ombudsman [hat is most likely to reduce the costs of lawsui~l sabotage, public attacks, etc. 13. Many full-time ombudsmen have contact with hundreds or even thousancb of peop[e a year 14. Examples from the past include dealing with fear of AIDS; the need for policies on harass- ment, fraud and misconduct, dependent care; unusual safety problems. Current examples include dealing with threats, genetic testing, intra-minority group harassment, discrimination on the basis of sexual orientation. 15. Examples of delicate issues include the disputes of family members in family-owned bus- inesses, people who smell bad, or beh~ve bizarrely, Common examples of difficult people me those who sczre others through temper tirades. Cross-cultuml misunderstandings and tensions are becoming much more common in recent years. An ombudsman of the same background as the complainant may be helpful. Moreover, employem are increasingly providing the option of two or more designated neutrals who cm work together with sometimes especially helpful where each neutral is similar in gender and ethnic background to eoch disputant.

REFERENCES

Anderson, D. R. and Stockton, D. M. (1990). Ombud$men in federal agencies: 7he tbeop and the practice. Wmhington: Administrative Conference of the C’nited Stmes. Ewing, D. (1989). Justice on the job, Boston: Hannrd Business School PreSS. Fisher, R. and Ury, W. L. (1981). Getting to YES, Boston: Houghton }Iifflin. McCabq D. (1988). Co@orate non-union cornplaintprvcedures and systetns. Xew lbrk: Pmeger. Rowe, M. P. (1990a). “People who feel harassed need I compllint system with both form~l ~nd informal options,” ,Vegotiation Journal 6 (2): 161-172. ———. (1990b). “Helping people help themselves: An ADR option for interpersonzd conflic[.” Negotiation Journal 6 (3): 239-248. ———. (forthcoming), “Options and choice for conflict resolution in the workplace.” In Chang- ing tactics, edited by Lavinia Hall. Washington: Nmional Institute for Dispute Resolution. Rowe, M. P. and Perneski, T. (1990), ‘‘Cost-effect i\-eness of ombudsmm offices.” Corporate Ombudsman ,Veu’sletteq Mq-. Rowe, M. P. , Sirnon, M. and Bensinger, A. t1990). “Ombudmm dilemmas: confiden[ialit~, neutrality, testifying, record-keeping.” In Proceedings, Annual Conference of the Society of Professionals in Dispute Resolution, edited by C, Cutrona, Washington: SPIDR. Ury, W. L., Brett, J. M. and Goldberg, S. B. (1988), Getting disputes resofL’ed: Designing sys- tems to cut the costs of confi’ict. San Francisco: Jossey-Bass. Westin, A. F. and Feliu, A. G. (1988). Resolving employment disputes without [itigation. Yew York: BNA Books. Ziegenfuss, J. T. (1985). Patient/client/employee comp[aintprograms: An organizational sys- tems model. Springfield, 111.:Thomm. ———. (1988). Oqanizational troubleshooters: Resolving prob[ems uqitb customem and employees, San Francisco: Jossey-Bass.