Sophie Miles

Journal of Evidence-Based Social Work, 6:348–360, 2009

Copyright © Taylor & Francis Group, LLC

ISSN: 1543-3714 print/1543-3722 online

DOI: 10.1080/15433710903126778

Ethical Guidelines for Designing and Conducting Evaluations of Social

Work Practice

MICHAEL J. HOLOSKO University of Georgia School of Social Work, Athens, Georgia, USA

BRUCE A. THYER Florida State University School of Social Work, Tallahassee, Florida, USA

J. ELAINE HOWSE DANNER University of Georgia School of Social Work, Athens, Georgia, USA

We review selected aspects of current ethical guidelines pertaining

to the design and conduct of social work evaluation and research

studies. We contend that there are significant differences between

social science research and evaluation studies, and that the uncrit-

ical application of ethical guidelines suitable for regulating social

science research may hinder social workers undertaking clinical

and program evaluations. What is needed are ethical guidelines

that distinguish between retrospective and prospectively designed

studies, which enumerate when voluntary and informed consent

may not be necessary in order to use data obtained from clients,

and clearer standards pertaining to exempting evaluation studies

from oversight by Institutional Review Boards.

KEYWORDS Ethical guidelines, social work practice, evaluation

In the past 25 years or so in North America, increased demands for health and social services coupled with scarcer financial resources have resulted

Portions of this paper were previously presented at the annual conference of the Society for Social Work and Research in Charleston, South Carolina, USA, on January 29–31, 2000, and at the International Conference on Evaluation for Practice held at the University of Huddersfield in the United Kingdom on July 12–14, 2000.

Address correspondence to Michael J. Holosko, Ph.D., Pauline M. Berger, Professor of Family and Child Welfare, University of Georgia School of Social Work, Athens, GA 30602. E-mail: [email protected]

348

Ethical Guidelines for Evaluation 349

in a burgeoning of evaluation activities. Indeed, as stated by Cronbach and colleagues (1980), ‘‘evaluation has become the liveliest frontier of American social science’’ (pp. 12–13). Such activities may validate and provide legiti- macy to these human services programs, but they also may provide scientific knowledge about their participants and interventions and the interventions used with them (Holosko, 1996).

Ever since Harriet Bartlett (1958) included research as a component of social work in her highly influential working definition of practice, the Code of Ethics (COE) of the United States National Association of Social Workers (NASW) has promoted a position that professional social workers engaged in scientific inquiry should adhere to a set of ethical principles. Section 5 of the COE (NASW, 1999), Social Worker’s Ethical Responsibilities to the Social Work Profession, describes these principles. Subsection 5.02, Evaluation and Research, describes 16 ethical guidelines, including these pertinent ones:

(d) Social workers engaged in evaluation or research should carefully consider possible consequences and should follow guidelines developed for the protection of evaluation and research participants. Appropriate institutional review boards should be consulted.

(e) Social workers engaged in evaluation or research should obtain vol- untary and written informed consent from participants, when appropri- ate, without any implied or actual deprivation or penalty for refusal to participate; without undue inducement to participate; and with due regard for participants’ well-being, privacy, and dignity. Informed consent should include information about the nature, extent, and duration of the participation requested and disclosure of the risks and benefits of participation in the research.

(h) Social workers should inform participants of their right to withdraw from evaluation and research at any time without penalty.

(o) Social workers engaged in evaluation or research should be alert to and avoid conflicts of interest and dual relationships with participants, should inform participants when a real or potential conflict of interest arises, and should take steps to resolve the issue in a manner that makes participants’ interest primary. (NASW, 1999)

It has only been since 1990 that the amended COE has included both evaluation and research in its guidelines. Prior to that time, the term Scholar- ship and Research was used as the sub-heading for a less detailed subsection on research ethics, which included only six guidelines (NASW, 1980).

Outside of the fact that such guidelines are just that—‘‘guidelines’’—they have been roundly criticized for having no ‘‘teeth’’ and for being ambiguous (Reamer, 1993). The larger problem is that the COE guidelines on Evaluation and Research fail to distinguish evaluation from research. As a result, social

350 M. J. Holosko et al.

workers involved in the three main forms of scientific inquiry: (a) basic social science research (research about social and behavioral phenomena), (b) practice research (research about social work practice), or (c) evaluation research, with its two sub-categories of practice evaluation (assessments of the efficacy of practice) and program evaluation (assessments of programs), tend to clump these together as ‘‘similar scientific endeavors,’’ all falling under the rubric of the COE for research involving human subjects.

Although it may appear to be appropriate to use the COE in this manner due to the relative similarities of these forms of inquiry, it has been demon- strated that research and evaluation are decidedly different in that they stem from differing assumptions (Gingerich, 1990) and have different purposes (Ashmore, 2005; Rubin & Babbie, 2005; Barlow, Hayes, & Nelson, 1984). Applying ethical guidelines (which come very close to being interpreted as mandates in some instances) appropriate for research can have an unduly constraining influence on the design and conduct of evaluation, constraints that interfere with ongoing efforts within the profession to encourage more social workers to conduct more and better empirical evaluations of their own practice using the methods of scientific inquiry. We believe this to be a serious problem. The last thing we need at this point in the maturation of the profession is the erection of unnecessary barriers to conducting evaluation studies. We will discuss some of these issues after reviewing some of the features that we believe distinguish social research efforts from practice evaluation studies.

SOCIAL RESEARCH VERSUS PRACTICE EVALUATION

Social research is defined as the systematic investigation of phenomena. All social research follows a process of inquiry, which minimally includes defining a research problem or question, reviewing literature and theory pertaining to the problem/question, collecting data, analyzing data, and concluding about the phenomena. The ultimate goal of social research con- ducted by social workers is to seek new generalizable knowledge about issues emanating from the practice worlds in which they work, namely: in- dividuals, small groups, families, human service organizations, communities, and the social environment. Some of the more commonly used quantita- tive methods of social research used by social workers include exploratory studies, quantitative-descriptive studies, quasi-experimental studies, descrip- tive studies, and meta-analyses. Frequently used qualitative methods in- clude field studies, phenomenological studies, case analyses, and ethnog- raphies.

Practice evaluations assess interventions to determine their potential benefits to clients or client systems (i.e., the systems or context in which clients find themselves, such as a family unit, school, treatment setting,

Ethical Guidelines for Evaluation 351

or community). Practice evaluations also follow a process of inquiry that minimally includes defining the client system problem, defining the inter- vention(s) used, collecting data about the intervention’s impact, analyzing data, and concluding about the intervention and its benefit to the client system. The goal of practice evaluation is to assess the potential bene- fit to client systems, and, while the development of generalizable scien- tific findings is desirable, it is of secondary importance (Bloom & Orme, 1993), if it occurs at all (and, in many instances, it does not). The primary scientific methods used by social workers to evaluate their practice out- comes involve single-system designs (see Bloom, Fischer, & Orme, 1999) and simple group research designs (see Royse, Thyer, Padgett, & Logan, 2006).

As one may surmise, the majority of social research and practice evaluations both conducted and published by social workers follows a hypothetico-deductive approach framed in a three-step research process: purpose, method, and findings. As a result, their differences are often obfuscated by the fact that they follow this same mode of empirical inquiry. Table 1 presents a set of criteria differentiating social research from practice evaluation. No published account of distinguishing features is presented in the literature, and, as such, this differentiation extends the groundswell argument offered by other authors for a distinct set of ethical guidelines for practice evaluation (c.f., Bloom & Orme, 1993; Grigsby & Roof, 1993; Millstein, Dare-Winters, & Sullivan, 1994).

TABLE 1 Distinguishing Features between Social Research (SR) and Practice Evaluation (PE)

Selected criteria Social research Practice evaluation

I. Purpose � To develop and test hypothesis

� To seek new knowledge

� To assess how practice benefits clients/client systems

II. Doctrine � Value free � Ideally altruistic

� Value laden � Deterministic

III. Theoretical Basis � Theoretical � Atheoretical IV. Intended Audiences � Other researchers

� Academics � Funding bodies

� Clients � Agency staff � Boards of directors � Funding bodies

V. Methodological Issues i. Sampling � Multiple

� n 9 10 � Singular � n D 1

ii. Instrumentation � Standardized � Crafted and standardized iii. Statistics � Descriptive and

inferential � Graphs, percents,

descriptive iv. Generalizability � High � Low v. Design � Much variability in rigor � Rigorous vi. Dissemination plan � Optional

� Academic conferences � Clients � Agency staff

352 M. J. Holosko et al.

ETHICAL CONSIDERATIONS

The topic of ethics involving research on human subjects has been ex- tensively studied. Less well known are how the issues of ethics apply to practice evaluation. Table 1 argues that social research (SR) and practice evaluation (PE) have a number of distinguishing features, and some of the main principles of research ethics may apply to PE. These are informing about the purpose of the study, informed consent, minimal risk, potential and harm from participation/non-participation, and confidentiality. Since 1980, the COE has also included one separate guideline for the evaluation of practice: subsection 5.02 (k), which states ‘‘Social workers engaged in the evaluation of services should discuss collected information only for profes- sional purposes and only with people professionally concerned with this information’’ (NASW, 1999).

Bloom and Orme (1993) offer 10 ethical principles for PEs that highlight the differences between research and practice evaluation ethics. These are presented in Table 2.

According to Gillespie (1987), the three major ethical risks involved in any research using human subjects include potential harm to subjects, participants, and the community and society. He suggested that the former potential harm to subjects receives much more discussion than the others. In an effort to offset this concern and address ethical issues from a participant perspective, Table 3 presents a list of prerequisite process questions that social work practitioners could preliminarily address prior to their conducting practice evaluations.

Table 3 is not presented as an all-inclusive set of questions but could be used as a guide for further discussion and consideration. We note that these questions could be used to develop a one-page letter of Intent to Evaluate

TABLE 2 Ethical Principles for the Evaluation of Practice

Ten ethical principles

1. Provide demonstrated help. 2. Demonstrate that no harm is done. 3. Involve the client in the evaluation coming to an agreement on the overall practice

relationship. 4. Involve the client in the identification of the specific problems and/or objectives and the

data collection process, as far as possible. 5. Demonstrate how the evaluation will not intrude on the intervention process. 6. Stop the evaluation whenever painful or harmful to the client, physically,

psychologically, or socially, without prejudice to the services being offered. 7. Maintain confidentiality with regard to the data resulting from the evaluation. 8. Balance the benefits of evaluating practice against the costs of not evaluating practice. 9. Use the Clients Bill of Rights as an intrinsic part of evaluating practice.

10. It should be recognized that any evaluation process reflects the values of the researcher.

Note. From Bloom and Orme, 1993, pp. 164–174.

Ethical Guidelines for Evaluation 353

TABLE 3 Preliminary Process Questions which may be Considered Prior to Conducting a Practice Evaluation

Process questions for practice evaluation

1. Do I have the necessary skills, education, and training to effectively evaluate my practice?

2. Why do I want to evaluate my practice at this time? 3. What are three benefits that can come from conducting this PE? 4. What are three consequences that can result from conducting this PE? 5. To what extent are my colleagues, supervisors, and/or administrators committed to this

PE? 6. Who will I seek as a mentor to assist with this PE? 7. How do I propose to evaluate my practice at this time? 8. How will I seek the client’s permission in this? 9. What will happen with this information after I collect it?

10. What steps will I take to ensure that ethical principles are dealt with in this PE? 11. Have I developed my Intent to Evaluate Practice Form and had my supervisor sign it. 12. Has the client signed the Consent to Treatment Form?

Practice that should be written and signed by the practitioner, signed by a supervisor, and eventually shared with the client during the next phase of this process. This letter then should be placed on file in the client’s case record. This suggestion is offered in the context of viewing evaluation as a form of social work research and, as such, falls under the purview of ethical guidelines pertaining to research.

There are, however, some alternative perspectives we would like to raise. If many practice evaluation activities are aimed at producing evidence- based knowledge about the outcomes of practice with particular clients or obtained from specific human service programs, not necessarily general- izable knowledge primarily intended to contribute to the foundations of science, then it could be contended that evaluation is not research, and that ethical guidelines intended to govern the design and conduct of research projects may not be relevant. We certainly need ethical guidelines to assist with research, but we already have extensive guidelines for practice, and if evaluation efforts are construed as a part and parcel of everyday social work practice (and not research), then the existing ethical standards related to practice also cover our evaluation efforts, and no additional governing authority (e.g., ethical guidelines dealing with research) is necessary. For example, the NASW’s ethical guidelines related to evaluation and research virtually mandate obtaining informed consent:

Social workers engaged in evaluation or research should obtain voluntary and written informed consent from participants, when appropriate: : : : Social workers should never design or conduct evaluation or research that does not use consent procedures, such as certain forms of naturalistic observation and archival research, unless rigorous and responsible review of the research has found it to be justified. (NASW, 1999)

354 M. J. Holosko et al.

The suggestions in this article are consistent with these standards. There are several considerations that render difficult viewing program and clinical evaluation as forms of research requiring ethical oversight as described in the NASW COE. A few of these difficulties are reviewed below.

The Complicated Issue of Informed Consent. Informed consent is sim- ply not always possible, as in the case of retrospective program evalua- tions (Feely, 2007) based upon agency records—studies that may not have been planned until very recently, by practitioners who wish to make use of archived agency and client records dating back some years. Since the data were gathered for administrative and clinical purposes, there is no informed consent form completed by clients in the charts. Since many clients may have been discharged, moved, or died, as a practical and logistical matter, contacting these former clients to obtain their consent is simply not possible. If you limit your data to studying information only obtained from those clients who are subsequently contacted and give permission to use their data, then the representativeness of your data is open to question. Does this mean that archival data cannot be used for evaluation purposes? Such a requirement could bring many types of evaluation to a halt. A literalist interpretation would also hinder using data obtained from public records or acquired from commercially available large-scale databases. We believe that an absolutist standard mandating voluntary informed consent for all forms of evaluation is a serious barrier to social workers designing and conducting outcome studies of their own practice.

Some Client Groups Will Refuse Consent. Many agencies and social workers serve involuntary clients such as sex offenders, substance abusers, perpetrators of domestic violence or child abuse and neglect, and court- mandated criminals. Large proportions of such individuals can be expected to decline the opportunity to participate in evaluation studies, to refuse voluntary informed consent, even if provided assurances regarding data con- fidentiality or anonymity. Evaluation research with such client groups could also grind to a halt. Would feminists support less research on evaluating the outcomes of spousal battering? Would ‘‘Mothers Against Drunk Driving’’ be happy if alcohol abuse prevention programs could not be credibly evaluated? Would parents want to discontinue research on orienting pedophiles to nor- mal, adult, consensual sexual relationships? Mandating voluntary informed consent in order to conduct evaluation research with such client groups is another unrealistic barrier to social workers evaluating their own practice.

Who Owns the Data? Who owns information obtained from clients— the clients, the agency administration, or the social worker who gathers it? Take the social worker in clinical practice who wishes to use school atten- dance records to evaluate the outcomes of a truancy prevention program. Is the information about the child’s attendance the property of the child, the parents, or the school—or is it ‘‘public’’ information? Would it be ethical to

Ethical Guidelines for Evaluation 355

not use these data to evaluate the program, irrespective of the consent of the children or parents concerned?

It could be contended that the data gathered by social workers in the routine course of their assessment, delivery of care, and monitoring of outcomes belong to the professional social worker, who has the latitude to employ the data for evaluation purposes, for teaching, in supervision, and other scholarly purposes, up and including the publication of articles in pro- fessional journals, as long as individually identifiable data were not disclosed. Client permission should not be mandated as an absolute prerequisite for using clinical data in evaluation studies.

Do You Need IRB Review and Approval for Evaluation Studies? An- swering this question depends upon the all-important definition of research. American universities usually follow the definition set forth in the federal regulation on human subjects which reads: ‘‘Research means a systematic investigation, including research development, testing and evaluation, de- signed to develop or contribute to generalizable knowledge’’ (Protection of Human Subjects, 2005, p. 3).

This definition lends itself to some gray areas. For example, single- subject designs used to evaluate outcomes with one or a very small number of individuals are not aimed at producing generalizable knowledge, nor do smaller-scale group designs, such as the posttest-only (X-O) or pretest- posttest (O-X-O) design, using non-probability samples. If the intent is to produce non-generalizable knowledge pertinent only to specific persons or agency programs, such evaluation studies may not fall under this federal definition of research, and thus may be exempt from a university’s institu- tional review board’s (IRB) oversight. Could such activities be called ‘‘quality assurance studies’’ or ‘‘program/practice evaluation studies,’’ and thereby avoid the entanglements of IRBs?

In fact, federal regulations already recognize this type of need and provide exemption for educators to evaluate their practice on an ongoing basis. Specifically exempted from human subjects policies are: ‘‘Research on regular and special education instructional strategies or the effectiveness of or the comparison among instructional techniques, curricula, or classroom management methods’’ (Protection of Human Subjects, 2005, p. 2). If federal regulations recognize the need for these evaluation activities to be conducted without the oversight of IRBs, then shouldn’t the same hold true for similar activities undertaken by social practitioners?

It is worth noting that the IRB review and approval process is not without its own costs. The paperwork activities are merely the beginning. Sometimes personal appearances are required to explain content in your research proposal to IRB committee members. (This can chew up half a day, easily!) IRBs often operate according to university timetables and slow down or halt operation during holidays or inter-term breaks. Gaining IRB approval

356 M. J. Holosko et al.

may take weeks, a month, or longer—precious time during which windows of opportunities for data collection may close.

There is also a measure of hypocrisy in the IRB process. While it is piously touted as a method of protecting the rights of human subjects, the reality is that it primarily functions as a federally mandated mechanism necessary to keep the flow of federal research dollars flowing to the entire university—dollars containing perhaps a 40% or more institutional overhead figure.

Recently, in the United States, a spate of universities have had their entire research programs shut down by the federal government because of violations of human subjects protection regulations by one small study (Oakes, 2002; Brainard, 2000). This heavy-handed approach to research oversight is not consistent with free scientific inquiry (Feely, 2007; Dingwall, 2007). Why should you be prohibited from conducting your study, which is appropriately designed and possesses suitable client protections because some individual in another department violated an IRB policy?

We would like to note that the authors always follow their respective university IRB policies and procedures and require their students to do so as well. Given the present punitive contingencies for failing to follow IRB policies (Oakes, 2002), we do not believe it is justifiable to place our students or department colleagues at risk for our not complying with them. But, we question the present conflation of social science research with evaluation, and the blanket application of research ethical guidelines to evaluation studies.

What about Practitioners without IRB Oversight? In this discussion, it is important to note that social workers work at all practice levels. While those working in macro or mezzo settings, like the government, academia, or large agencies with federal funding, have clear-cut accountability to federal regu- lations and IRB requirements, many working on the micro or mezzo level, such as in private practice or in smaller agencies without federal funding, do not have clear oversight. Although exempt from federal requirements, these practitioners are bound by ethical standards to protect their subjects. The most important way this is done is through informed consent. Ethical practitioners in most settings already routinely obtain informed consent at the beginning of work with clients. However, this type of general consent is not considered by the authors to be sufficient to meet ethical obligations when conducting research and/or evaluation studies. It is necessary to delineate treatment from study largely because clients should not be obligated to be human subjects (Oakes, 2002). Therefore, separate informed consent specific to the study must be obtained, which should minimally include explanation of (a) the purpose, duration, and procedures of the study; (b) foreseeable risks and benefits; (c) alternate procedures or options; (d) confidentiality of records; (e) who to contact for questions; and (f ) voluntary participation with no penalty for non-participation (Protection of Human Subjects, 2005;

Ethical Guidelines for Evaluation 357

NASW, 1999). Reamer (2001) offers a sample informed consent form that meets these criteria and could easily be modified as needed.

The Problem of Overlapping Governing Authorities. In North America a social worker may be licensed to practice social work in the state or province in which s/he resides. S/he may also belong to a major professional association such as the Canadian Association of Social Workers, the Na- tional Association of Social Workers, or the Clinical Social Work Federation. S/he may also belong to a specialty membership organization, such as the American Association for Marital and Family Therapy, the Association for Behavior Analysis, or the Association for Advancement of Behavior Therapy. S/he may also be a member of an evaluation organization, such as the American Evaluation Association. We note that each and every one of the above groups provides its own unique and separate codes of ethics! A social worker belonging to all of these is subject to the ethical guidelines governing the practice of all members of each group, as well as being subject to the disciplinary authority constituted to deal with violations of ethical guide- lines. And there also remain civil legal remedies (e.g., lawsuits) to provide corrective actions against social workers who harm clients. Another problem is that sometimes the codes of ethics conflict is in their recommendations. A social worker who conducts an evaluation study consistent with one (less restrictive) code may find him/herself the focus of an ethics inquiry undertaken by the professional association they belong to, which has a more restrictive code.

Recently, there has been discussion among many of our professional organizations toward unification of the social work profession and its pro- fessional organizations (Fortune, 2007), which has some potential to simplify some of these issues of conflicting ethical guidelines and complicated layers of accountability.

Do We Really Need New or Additional Ethical Guidelines Regulating

Evaluation of Practice? Perhaps not. The existing codes of ethics promul- gated by the professional associations, those associated with licensure, and the legal system provide overlapping and redundant protection of clients’ rights. These layers of oversight and accountability are sufficient to ensure that social workers ethically evaluate their practice and can be reasonably contended to be overly restrictive in some matters.

IMPLICATIONS FOR PRACTICE

We believe that additional work needs to be undertaken so that not only ethical codes but also federal regulations and IRB policies more clearly discriminate between social science research activities and evaluations of social work practice. More generous recognition needs to be articulated that informed and voluntary consent is neither always practical (retrospectively

358 M. J. Holosko et al.

designed studies), and that for many types of evaluation (involuntary clients), it is simply not possible to obtain. In other studies, such as those involving publicly observable behavior in which no personally identifiable informa- tion is gathered, or which used public records, census data, or large scale databases, consent (and IRB review) may simply not be necessary at all.

Social workers need to more vigorously assert their ownership of client data collected in the context of their own practice and of their rights to use this information for evaluation purposes without informing clients or obtaining their consent, provided no individually identifiable data are pre- sented to the public (e.g., published). We need to more vigorously assert our rights to freely prepare scholarly articles describing evaluation studies we have conducted. Sometimes agency administrators inform practitioners that they may not prepare such studies or seek to publish them without agency approval. This, too, mitigates against open scholarly inquiry.

We need to work more closely with institutional IRBs and get them to clarify when IRB review and approval is mandatory and when it is not. At present, the conservative tendency is for IRBs to assert that such approval is required whenever one gathers data from human subjects for research purposes with the intent to produce generalizable knowledge and with the intent to publish. This standard is full of gray areas and presents social workers with the unnecessary burden of complying with IRB review require- ments, even though it is clear that the study will be found to be exempt. Sometimes one undertakes evaluation work with no intent to publish the results, and it is only after the study is conducted that the potential for publishing an evaluation project becomes evident. Can one escape IRB oversight by claiming no initial intent to publish, then change your intention when the study is complete and go ahead and publish the work? We suspect that IRBs would not be happy with such behavior. Can we never publish evaluation data whose value became evident in retrospect?

Does the student conducting a single case evaluation of his/her work with one client seen in practicum require IRB review and approval? A literal interpretation of current policies at most universities suggests that this is so. But, the time constraints of the semester schedule may preclude prospec- tively obtaining such reviews. Does a faculty member providing consultation with an agency, helping them to select an assessment measure to be used with clients when they begin and terminate treatment, need IRB review and approval to assist in the conduct of this O-X-O study? What about a faculty member contacted by an agency to help them analyze evaluation data the agency has independently collected and to help them write it up for publication? The faculty member had no role in collecting data from human subjects. But, is his/her career in jeopardy if their name appears as the co-author of a publication of this nature, one which did not receive IRB review and approval by the university that employs the social work faculty member?

Ethical Guidelines for Evaluation 359

Social workers conducting evaluation studies do need ethical guidelines. But, they need ethical guidelines that are appropriate to evaluation and not the unmodified application of ethical guidelines suitable for either practice or social science research. It is important that, when revisions to professional Codes of Ethics are undertaken, social workers with experience in evaluation research be appointed to work on the development of such guidelines more appropriate to this specialized form of inquiry.

As the profession continues its emphasis on evidence-based practice, practitioners at all practice levels need to be involved in evaluation of prac- tice. Rather than reject the legitimacy of IRBs, we suggest a refinement of the process, specifically clarifying the distinction between research and evaluation. This would go a long way toward creating conditions under which evaluation of practice can be conducted ethically and efficiently.

REFERENCES

Ashmore, R. (2005). A study on how mental health practitioners address ethical issues in clinical audit. Journal of Psychiatric and Mental Health Nursing, 12, 112–120.

Barlow, D. H., Hayes, S. C., & Nelson, R. O. (1984). The scientist practitioner. Elmsford, NY: Pergamon.

Bartlett, H. (1958). Toward clarification and improvement of social work. Social Work, 3, 3–9.

Bloom, M., Fischer, J., & Orme, J. (1999). Evaluating practice (3rd ed.). Boston: Allyn & Bacon.

Bloom, M., & Orme, J. (1993). Ethics and the single system design. Journal of Social Service Research, 18, 161–180.

Brainard, J. (2000, January 28). U.S. suspends human-subject research at Virginia Commonwealth University. The Chronicle of Higher Education, A33.

Cronbach, L. J., Ambron, S. R., Dornbusch, S. M., Hess, R. D., Hornick, R. C., Phillips, D. C., Walker, D. F., & Weiner, S. S. (1980). Toward reform of program evaluation. San Francisco: Jossey Bass.

Dingwall, R. (2007). Turn off the oxygen: : : : Law and Society Review, 41, 787–796. Feely, M. M. (2007). Legality, social research, and the challenges of institutional

review boards. Law and Society Review, 41, 757–776. Fortune, R. (2007). Unification of the social work profession. Retrieved January 21,

2008, from http://www.sswr.org/unification.php Gillespie, D. (1987). Ethical issues in research. In A. Minahan (Ed.), Encyclopedia of

social work (18th ed., pp. 503–512). Silver Spring, MD: NASW Press. Gingerich, W. J. (1990). Rethinking single-case evaluation. In L. Videka-Sherman &

W. J. Reid (Eds.), Advances in clinical social work research (pp. 11–24). Silver Spring, MD: NASW Press.

Grigsby, R. K., & Roof, H. (1993). Federal policy for the protection of human subjects: Applications to research on social work practice. Research in Social Work Practice, 3, 448–461.

360 M. J. Holosko et al.

Holosko, M. J. (1996). Obstacles to conducting program evaluations: Implications for public health evaluators. Journal of Health and Social Policy, 8, 91–101.

Millstein, K. H., Dare-Winters, K., & Sullivan, S. (1994). The power of silence: Ethical dilemmas of informed consent in practice evaluation. Clinical Social Work Journal, 22, 317–329.

National Association of Social Workers (1980). Code of ethics. Washington, DC: NASW Press.

National Association of Social Workers (1999). Code of ethics. Washington, DC: NASW Press. Retrieved January 14, 2008, from http://www.socialworkers.org/ pubs/code/code.asp

Oakes, J. M. (2002). Risks and wrongs in social science research: An evaluator’s guide to the IRB. Evaluation Review, 26, 443–479.

Protection of Human Subjects, 45 C.F.R. § 46 (2005). Reamer, F. G. (1993). Ethical dilemmas in social service (2nd ed.). New York:

Columbia University Press. Retrieved from www.hhs.gov/ohrp/humansubjects/ guidance/45cfr46.htm

Reamer, F. G. (2001). Ethical issues. In B. A. Thyer (Ed.), The handbook of social work research methods. Thousand Oaks, CA: Sage Publications.

Royse, D., Thyer, B. A., Padgett, D., & Logan, T. K. (2006). Program evaluation: An introduction (4th ed.). Pacific Grove, CA: Wadsworth.

Rubin, A., & Babbie, E. (2005). Research methods for social work (5th ed.). Belmont, CA: Thomson Brooks/Cole.