Otima Z. Doyle, Ph.D.

Duke University Medical Center, Department of Psychiatry and Behavioral Sciences

Shari E. Miller, Ph.D.

University of Georgia, School of Social Work

Fatima Y. Mirza

University of Maryland, Baltimore, School of Social Work

Journal of Social Work Values and Ethics, Volume 6, Number 1 (2009)

Copyright © 2009, White Hat Communications

This text may be freely shared among individuals, but it may not be republished in any medium without express written consent from the authors and advanced notification of White Hat Communications.

The authors would like to acknowledge support from the University of Maryland, Baltimore, School of Social Work. The authors wish to thank Elizabeth Greeno, Ph.D., Donna Harrington, Ph.D., Anna Hayward, Anne K. Hughes, Kelli Larsen, Ph.D., JuYoung Park, & Karen L. Parker. Correspondence may be addressed to Otima Doyle, Ph.D., Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, Box 3527, 718 Rutherford St., Durham, NC 27710 or otima.doyle@duke.edu.

Abstract

Little systematic research appears to exist that explores the complex and essential process of ethical decision-making among social workers. This paper presents results of a study of NASW members that explored factors that relate to ethical decision-making, discrepancies in ethical decision-making, and rationales for courses of action chosen. Findings suggest that both personal and professional factors are related to ethical decision-making and predict the degree to which ethical decisions are discrepant.

Key words: Social Work Values, Ethics, Ethical Decision-Making, Diversity, Code of Ethics

1. Introduction

Questions of ethics and its relationship to human consciousness have been the focus of philosophical consideration for thousands of years and can be understood as encapsulating “traditions of belief that have evolved…in societies concerning right and wrong behavior” (Hopkins, 1997, p. 5). Modern professions incorporate the idea of ethics into practice by developing specialized codes of ethics to apply order and guide professional decision-making (Dolgoff, Loewenberg, & Harrington, 2005). In the United States, the National Association of Social Workers’ (NASW) Code of Ethics (1999) is accepted as the primary ethical standard for the profession. The Code of Ethics’ set of core values and the principles and standards developed on the basis of those values are designed to inform ethical decision-making. However, the Code of Ethics does not provide universal prescriptions for behavior, nor does it specify a hierarchy of values, ethical principles, or standards. This is likely due to the complex, context-bound nature of the process of ethical decision-making (NASW, 1999).

In America in particular, social work’s ethical considerations have evolved to include ethical standards (Dolgoff et al., 2005), questions of risk management and quality of service delivery, and the need to satisfy funding agencies and regulatory bodies with demonstrated competence in practice (Strom-Gottfried, 2000). As social workers increasingly navigate community-based treatment modalities and participate in interdisciplinary teams in which each member might operate from a different service or treatment paradigm or from a different set of professional values, the frequency and intricacy of ethical dilemmas will likely increase (Hoy & Feigenbaum, 2005). Given the complexity of this topic, the current article focuses on values and ethical decision-making among social workers in the United States, although it is important to note that some social work values may transcend international boundaries (Abbott, 1999).

The social work profession’s value base and its attendant ethical structure in the United States are built on a foundation characterized by mainstream cultural values (Abbott, 1988). These generally accepted “mainstream” values and moral traditions might serve to exclude culturally determined ethical standards of people from other backgrounds, whether they are social workers themselves or clients served by the profession. Respect for diversity, one of the profession’s explicit values, applies to both social work professionals and clients (CSWE, 2002). Social workers, regardless of their own background, must incorporate a comprehensive understanding of each client’s culture and values into their work (CSWE, 2002). They also need to be aware of the possible influence of their own beliefs and values on practice in ethically challenging situations (Csikai, 1999). Therefore, in addition to managing competing professional values social workers must also manage the influence of their own personal value systems while simultaneously considering their clients’ values when making ethical decisions.

Despite the importance of the topic, there appears to be a paucity of research that explores the relationships between the multiplicity of factors that influence social workers’ values and ethical decision-making. Many articles that document factors related to how and why practitioners make ethical decisions are based on theoretical aspects of ethical decision-making (Mattison, 2000; Walz & Ritchie, 2000), descriptions of the types of ethical violations reported to NASW (Strom-Gottfried, 2000), or recommendations for future practices (Dolgoff et al., 2005; Hoy & Feigenbaum, 2005; Mattison, 2000). Given the limited empirical base in this area, as well as the complexity and importance of ethical decision-making for social work practice, it is critical to develop a greater understanding of the patterns of ethical decision-making as they relate to the diverse personal and professional characteristics of social workers themselves. Further, it is important to understand the rationales that social workers use to make such decisions. Do social workers base their decisions on the standards within the Code of Ethics, or are their decisions influenced by other factors?

Almost all of the issues faced by social workers are based in ethical principals (Dolgoff et al, 2005). Becoming aware of, and being willing to acknowledge ethical dilemmas and their complexity, in practice, are important steps toward acquiring the skills to manage those ethical dilemmas (Dolgoff et al., 2005). However it is suggested that some social workers may not acknowledge the ethical dilemmas that arise in practice for a variety of reasons. For example, they may be uncomfortable making ethical decisions, they may think they know the “right” answer to the ethical dilemma, or they may be uneasy dealing with such issues because they do not feel they have the skills to manage ethical problems (Dolgoff et al., 2005). Given these possibilities, it is also critical to develop an understanding of the personal and professional factors that influence discrepancies in the ethical decision-making process.

This paper presents findings from an empirical study that explored patterns of ethical decision-making and rationales for making those decisions among a national sample of NASW members. Findings are also presented regarding personal and professional factors that predict discrepancies between what practicing social workers reported they would do versus what they felt they should do when confronted with ethical dilemmas. This exploratory study focused on three research questions: 1) What personal and professional factors are related to social workers’ courses of action when making ethical decisions?, 2) What rationales do social workers use to make ethical decisions?, and 3) What personal and professional factors are related to discrepancies in ethical decision-making among social workers? After an extensive review of the literature, it appears that no social work specific theoretical model describing what factors impact upon ethical decision-making exists. As such, a theoretical model from the business literature (Hopkins, 1997) was adapted, incorporating relevant factors supported by the social work literature and used as a guiding framework for this study.

2. Literature Review

Ethics are distinct from values, although the terms are often used interchangeably, and the concepts are inextricably linked. For the purposes of this paper, that distinction is a key consideration, and the relationship between the two, particularly the direct influence values have on the development of ethical standards (Boland, 2006; Csikai, 1999), is essential. The notion that personal and professional values are integral to the ethical decision-making process is strongly supported by the conceptual and theoretical literature on the topic (Abramson, 1996; Dolgoff et al., 2005; Freud & Krug, 2002; Hopkins, 1997; Mattison, 2000; Pike, 1996; Reamer, 1998). However there is a limited body of research that explores the relationship between values and ethical decision-making in social work. Some suggest that personal rather than professional codes for behavior are more likely to influence the resolution of ethical decisions (Canda & Furman, 1999; Haynes, 1999; Smith, McGuire, Abbott & Blau, 1991). Study findings suggest that the Code of Ethics does not serve as the primary basis for ethical decision-making but that practice wisdom, personal values (Dolgoff & Skolnik, 1996), and supervision (Landau, 1999b) are key resources that inform decision-making. In a study of ethical decision-making in a related field, researchers concluded that psychologists utilized formal rules and codes of ethics when considering what one should do when presented with an ethical dilemma. However, they relied on personal values and practical considerations in order to determine what they would actually do given the same situation (Smith et al., 1991).

From this perspective, questions regarding the source(s) of personal values become relevant considerations. It is commonly believed that people from different backgrounds will have different perspectives about how to behave in different situations. The influence of one’s culture on one’s personal values and ethics (Hopkins, 1997) has been supported by others in the social work field who note that values influence ethical decision-making (Boland, 2006; Csikai, 1999).

Because “there are as many different ethical…standards as there are different cultures” (Hopkins, 1997, p. 16), it is important to identify personal factors that may relate to differences in values, and in turn, discrepancies in ethical decision-making. Hopkins identifies several major demographic factors related to human diversity. These categories go beyond racial, cultural, and religious contexts to include age, gender, spirituality, language, disability, sexuality, and geography. There are a number of ways that each category can be explored to identify differences in people’s value systems. For example culture might be explored through the examination of race, ethnicity, nationality, or color. Several factors identified by Hopkins are also noted in the social work literature as having an influence on one’s values and ethics. They are: culture/ethnicity, gender, and religion/spirituality. The empirical literature in these areas is discussed below.

2.1 Personal Factors and Ethical Decision-Making

2.1.1 Culture/Ethnicity

The majority of articles included in this review largely focused on Caucasian respondents, and articles that addressed culture approached the topic from different perspectives (e.g., race, ethnicity). Empirical evidence exists within the social work literature suggesting that race has a bearing on the degree to which social work values can be considered universally accepted (Abbott, 1999). Examining issues specific to hospital social work, Csikai (1999) found that cultural beliefs tended to be negatively correlated with attitudes toward the legalization of euthanasia and assisted suicide. However, Perkins, Hudson, Gray, and Stewart’s (1998) study involving community mental health providers did not find significant relationships between ethnicity and the tendency towards making conservative decisions in ethically challenging situations.

2.1.2 Religion/Spirituality

Though some argue that religion is incompatible with the mission of social work (Miller, 2001) there has been increased interest in, and emphasis on, religion and spirituality in the social work field over approximately the last decade (Canda & Furman, 1999). Religion and/or spirituality are likely to present themselves in practice when dealing with a multitude of issues such as terminal illness, bereavement, aging, difficult family relations, foster parenting, domestic violence, natural disaster, mental illness, and poverty (Canda & Furman, 1999; Dudley & Helfgott, 1990; Hodge, 2005; Miller, 2001). Findings from empirical research demonstrate that many social workers value religion and spirituality in their own lives and are incorporating both in their practice (Canda & Furman, 1999; Miller, 2001; Sheridan, Bullis, Adcock, Berlin, & Miller, 1992). Further research has demonstrated that graduate students in an NASW sample were highly motivated by their own religious belief systems (Hodge, 2005). Csikai (1999) found that both personal (religious and cultural) and professional values impacted upon social workers’ attitudes toward morally and ethically charged issues. Specifically, religious beliefs among social workers have been found to be negatively associated with attitudes related to euthanasia and assisted suicide (Csikai, 1999). With regard to ethical judgment, Landau (1999a) found that those who perceived themselves as religious appeared to assign greater importance to moral considerations than those who perceived themselves to be non-religious.

2.1.3 Gender

A number of the articles included in this review did not explicitly explore differences related to gender in their studies. Haas et al. (1988) reported moderate gender effects on patterns found between courses of action respondents selected and the rationale used to support those decisions. However in their study of community mental health providers Perkins et al. (1998) did not find gender to be significantly related to conservative ethical decision-making.

2.2 Professional Factors and Ethical Decision-Making

Social workers are expected to critically examine ethical issues in order to come to a resolution that is consistent with social work values and ethical principles and to thereby minimize unethical behavior. In order to assist students and practitioners, various process models have been suggested to guide ethical decision-making (e.g., Abramson, 1996; Dolgoff et al., 2005; Linzer, 1999; Linzer, 2004; Manning, 1997; Mattison, 2000; Reisch, & Lowe, 2000; Spano & Koenig, 2003). These models are thought to assist in the application of social work values in the decision-making process. In addition, the social work literature focuses on several factors that are thought to minimize unethical decisions. These include the years of experience one has in the field, exposure to formal and informal ethics training, the degree to which one identifies with social work values, and one’s level of social work education.

2.2.1 Social work education

There is an implicit assumption that incorporating ethics content in the curriculum will affect the attitudes and behaviors of future social workers. However, few empirical studies have challenged this assumption (Dolgoff et al., 2005). Some posit that “students understand the professional in terms of the personal” (Haynes, 1999, p. 41), and others suggest that learning about values and ethics may not result in the incorporation of those ethics into decision-making (Dolgoff et al., 2005; Haynes, 1999).

Landau (1999a) explored the impact of professional socialization and a variety of demographic variables on both ethical judgment and ethical decision-making. Her findings suggest that although social work education plays a key role in the acquisition of social work values, professional socialization does not affect ethical judgment and does not appear to directly support the use of a process of ethical decision-making. Another study indicates that although BSW students and faculty can recognize ethical dilemmas and identify conflicting values, they are not proficient with regard to proposing resolutions to ethical dilemmas (Wesley, 2002). There is some support for the notion that one’s level of education relates to ethical attitudes. More specifically, master’s level social work students were reported as more likely to agree with the legalization of assisted suicide and euthanasia than bachelor’s level students (Csikai, 1999).

2.2.2 Formal and informal ethics training

Studies exploring ethical decision-making among psychology students have not found significant relationships between formal and informal ethics training and ethical decision-making (Haas et al., 1988; Perkins et al., 1998). However, findings from social work research suggest that whether or not one has had ethics training influences one’s ethical attitudes (Csikai, 1999) and ethical decision-making (Boland, 2006) in particular for hospital social workers. Specifically, Csikai found that attitudes toward the legalization of assisted suicide differed significantly in relationship to whether or not hospital social workers had received formal ethics training. Boland found that prior ethics training was identified as the only significant predictor of the use of a decision-making process when hospital social workers confronted ethical dilemmas.

2.2.3 Years of experience

Haas and colleagues (1988) found that respondents with fewer years of experience seemed to be more inclined to choose to report or confront an offending party in an ethical dilemma than were those with more years of experience. In addition, Perkins and colleagues (1998) found that those who had longer tenure at an agency tended to make fewer conservative decisions regarding boundary related ethical dilemmas. However, they did not find the same association between experience and conservative decision-making. Another study found that as social work experience increased, so too did the likelihood that social workers agreed to participate in either euthanasia or assisted suicide (Csikai, 1999). However, Boland (2006) did not find a significant relationship between years of experience, in either social work or hospital social work, and the use of a process for making ethical decisions.

2.2.4 Social work values

The social work literature is characterized by mixed findings with regard to social work students’ adherence to the professional values base. Some findings support the idea that students’ adherence to social work values increases over time (Abbott, 1988; Frans & Moran, 1993), whereas others either indicate no change over time, or change that was not in a desirable direction (Enoch, 1989; Manzo & Ross-Gordon, 1990; Wodarski, Pippin, & Daniels, 1988). However few studies were located that examine the relationship between social work values and ethical decision-making. Boland (2006) explored the degree to which health care social workers can identify an ethical dilemma, provide a rationale for a decision made, and follow a process to resolve the identified dilemma. No significant relationship was found between social work values and the ability to identify an ethical dilemma. Findings do suggest however that internalized social work values are related to the use of a higher order rationale process for identifying ethical dilemmas. Csikai (1999), in a study designed to examine the impact of personal and professional values on the ethical attitudes of hospital social workers found that professional values (self-determination and social justice) were positively correlated with willingness to participate in euthanasia or assisted suicide.

2.3 Summarizing the Literature: A Model for Exploration

Taken together, the literature provides support for the exploration of patterns in ethical decision-making among social workers, as well as identifies both personal and professional factors that may either minimize or increase discrepancies in ethical decision-making. The combination of personal and professional factors represents an adaptation of Hopkins’ model in two ways. First the model is modified to include only those personal factors available in our dataset that are both identified by Hopkins’ model and supported by the social work literature, and are thought to relate to differences in values and in turn, discrepancies in ethical decision-making. Second, the model is expanded to incorporate social work factors that are thought to minimize discrepancies in ethical decision-making. Specifically, the literature provides support based upon theory and prior research that the following variables are potentially related to ethical decision-making among social workers: ethnicity, religion, gender, social work education, ethics training, and social work values.

3. Method

3.1 Participants

Participant demographics are consistent with the general population of licensed social workers (NASW, 2006). The sample is comprised of primarily white (88.3%) women (80.5%) with a mean age of 49.5 years (SD = 14.39, range = 20 to 85 years) who have been practicing on average for 18.82 years (SD = 12.32 years, range from 0 to 53 years) and work primarily in direct practice (68.7%) in the areas of mental health (43%) and child welfare/family (10.3%). In addition, the majority of respondents hold a master’s degree (84.2%), are licensed social workers (85.8%), and have had formal ethics training (82.1%).

3.2 Procedure

As a part of a larger collaborative project, the research team developed a survey that included demographic questions, an adapted version of the Professional Opinion Scale (Abbott, 1988), and an adapted version of the vignette-based Ethical Choice Score Rating System (Smith et al., 1991). The survey was pilot tested with 30 social work practitioners and students. Modifications were made and the survey was distributed following procedures approved by the University Institutional Review Board. Five hundred and one individuals were randomly selected using a random number generator from a sample of 2005-2006 NASW members. This national sampling frame was selected because of its potential to capture a broad range of practitioners, both clinical and non-clinical in orientation.

Participants received four mailings between January and March 2006 following a modified Dillman (2000) approach to increase response rate. Materials sent to eight individuals were returned by the postal service marked “return to sender” with no forwarding address, leaving a final sample of 493. Two hundred thirty-four respondents returned a completed survey by June 30, 2006, and six declined to participate in the study, yielding a response rate of 47.5%.

3.3 Measures

The survey contained a series of demographic questions about personal and professional characteristics and two standardized measures. Questions regarding personal characteristics included demographic questions such as, age, gender, race/ethnicity, religious affiliation, and household income. Some of these demographic variables were converted to dummy variables based upon patterns in the overall sample, as well as the requirements of particular analyses. For example, the following variables were created for religion: Catholic, Jewish, and Protestant. Additionally, due to the limited racial/ethnic variability in the sample, the variable “minority” was created to incorporate all participants who identified themselves as racial or ethnic minorities (e.g., African American, Spanish, and so forth). Though this delineation does not capture the differences between the cultures in each category, it provides a starting point from which to explore the relationship between ethnicity and ethical decision-making. Questions related to professional characteristics included years of professional experience, educational degrees obtained, and whether or not participants had engaged in formal ethics training.

3.3.1 Social work values

The Professional Opinion Scale (POS) (Abbott, 1988) was utilized for this study because it assesses social workers’ commitment to social work values (Abbott, 1999). An adapted version of the 40-item POS (Abbott, 1988) was used (Alpha = .86) (for more information, see Greeno et. al, 2007). Questions are designed on a five-point Likert scale (“1” = strongly agree to “5” = strongly disagree) with higher scores corresponding to greater commitment to social work values and lower scores corresponding to  lesser commitment to social work values (Abbott, 1988). The scoring procedures for this study followed Abbott (2003), Boland (2006), and Greeno et al. (2007).

3.3.2 Ethical decision-making

The Ethical Choice Score Rating System (ECSRS) is a modified form of the 10-question vignette measure first developed by Haas et al. (1988) and then modified by Smith et al. (1991). Further modifications were made to Smith and colleagues’ (1991) measure for the purposes of this research in an attempt to better capture the broad range of social work practice. This resulted in an adapted measure consisting of six vignettes addressing both macro and clinical issues. The measure asked respondents to read each vignette containing an ethical dilemma and then to answer four questions using the response choices provided: (a) what the respondent would do in the situation described, (b) the associated rationale choice, (c) what the respondent thinks he/she should do in that situation, and d) the associated rationale choice. Rationale choices included: (a) upholding the law, (b) upholding the Code of Ethics, (c) unable to identify a specific reason/it feels right (intuition), (d) upholding personal moral values/standards, (e) financial need, (f) fear of reprisal (e.g. malpractice suit), (g) fear of verbal/social reprisal from supervisor, colleague, or client, and (h) protection of personal/professional reputation (adapted from Smith et al., 1991).

Each research question captured specific aspects of this measure for its corresponding analysis. The first research question focused on the courses of action that respondents indicated they “would” and “should” take given the scenarios in the vignettes. For the second research question, the relationship between these answers and rationale choices was explored. The original eight rationale choices were condensed into two categories representing “codified” or rule-based options (e.g., upholding the law or Code of Ethics) and “non-codified” options (e.g., “it just feels right”) following the model used by Haas and colleagues (1988) (see Appendix 1). For the third question, a difference score was calculated according to the number of times that the respondent selected a different answer for their “would” and “should” response across the six vignettes, thus representing a discrepancy between what the respondents reported they would do in the given situation versus what they felt they should do in the same situation.

3.4 Data Analysis

All survey responses were entered into a database in SPSS® version 11.0.1 (SPSS Inc., 2001). Descriptive statistics were used to gather sample demographics and to determine frequencies, means, and standard deviations where relevant for each of the study variables. To answer the first question, “What personal and professional factors are related to social workers’ courses of action when making ethical decisions?,” chi square analyses and a MANOVA were conducted to examine the relationships between social workers’ demographic characteristics and their selected courses of action for both would and should responses. In order to answer the second question, “What rationales do social workers use to make ethical decisions?” chi-square analyses and a MANOVA were conducted to examine the relationships between the respondents’ courses of action and their rationale choices. In order to answer the third question, “What personal and professional factors are related to discrepancies in ethical decision-making among social workers?” multiple regression was utilized to examine which factors were related to higher or lower discrepancy scores on the ECSRS. Predictors included formal training, years of experience, highest degree, minority status (0 = non-minority, 1 = minority), Catholic (0 = other religions, 1 = Catholic), Protestant (0 = other religions, 1 = Protestant), Jewish (0 = other religions, 1 = Jewish), spirituality, and POS score. Following Stevens (2002), the sample size: predictor ratio was within reasonable limits (n = 201 with 10 predictors) for the regression analysis. Because of the exploratory nature of this research question, an alpha level of 0.10 was used (Cohen, Cohen, West & Aiken, 2003). Assumptions were checked and adequately met.

4. Results

The vignettes were examined in order to identify patterns in decision-making and whether those patterns were related to codified or non-codified responses. Due to the complexity of the data, and the relatedness of research questions one and two, results for both of these questions are integrated in this section of the manuscript. The vignettes are presented sequentially and organized in the following manner: (a) a brief summary of the vignette, (b) a discussion of significant relationships between the demographics and the corresponding courses of action for both “would” and “should” responses, (c) a table depicting the significant findings reported in “b” above (where applicable), (d) a discussion of any significant relationships between the courses of action and the rationale choices for both “would” and “should” responses, and (e) a table depicting the significant findings reported in “d” above (where applicable). In some instances, near significant findings of interest are also presented.

4.1 Vignette 1 (Referral/Do Not Respect Coworker)

Vignette 1 presented the following scenario: as a therapist you are asked by the Clinical Director to refer a client to a therapist whose ability you do not respect. The three courses of action offered to the respondent were: a) refer the patient, b) refer the patient and indicate your reservations, and c) refuse to refer the patient.

4.1.1 Demographics and course of action

Only commitment to social work values was related to how individuals felt they would respond (F = 3.274, p = 0.04). Though the relationship only approached significance (p = 0.054), it appeared that those with greater commitment to social work values selected option “C” (refuse to refer the patient) (M = 4.065) versus “B” (refer the patient and indicate your reservations) (M = 3.942).

Both ethnicity (ChiSq = 10.744, p = 0.005) and commitment to social work values (F = 3.244, p = 0.041) were related to what individuals felt they should do. Minority respondents selected option “B” (refer the patient and indicate your reservations), whereas non-minority respondents selected option “C” (refuse to refer the patient). As before, commitment to social work values showed a relationship that only approached significance (p = 0.071): those with lesser commitment to social work values (M = 3.942) selected option “B” (refer the patient and indicate your reservations) and those with greater commitment to social work values (M = 4.069) selected option “C” (refuse to refer the patient).

4.1.2 Course of action and rationales

No significant relationships were found between response choices and codified or non-codified rationale choices for either “would” or “should” questions.

4.2 Vignette 2 (Sexual Misconduct)

Vignette 2 presented the following scenario: a client tells you that a previous therapist made sexual advances toward her. This is the third client from whom you have heard such allegations. The four courses of action offered to the respondent were: a) call ethics committee/state licensing board, b) tell patient she has right to contact ethics committee or state licensing board, c) call the previous therapist about the violation, and d) discuss the patient’s anger but not the issue of professional standards.

4.2.1 Demographics and course of action

None of the demographics were significantly related to how individuals felt they would respond to the scenario. Being Jewish (ChiSq = 11.670, p = 0.009) was related to what individuals felt they should do. Jewish individuals tended to select option “A” (call ethics committee/state licensing board) whereas non-Jewish individuals selected option “B” (tell patient she has right to contact ethics committee or state licensing board).

4.2.2 Course of action and rationales

For the “would” response patterns in this vignette, those respondents who chose option “A” (call ethics committee/state licensing board) tended to do so based upon codified rationale choices, and those who chose options “B” (tell patient she has right to contact ethics committee or state licensing board) and “D” (discuss the patient’s anger but not the issue of professional standards) tended to do so based on non-codified rationale choices (ChiSq = 14.186, p = 0.003).

In terms of “should” response patterns, those who chose option “A” (call ethics committee/state licensing board) tended to do so based upon codified rationale choices, and those who chose “D” (discuss the patient’s anger but not the issue of professional standards) tended to do so based upon non-codified rationale choices (ChiSq = 9.296, p = 0.026).

4.3 Vignette 3 (Referral/Funding Cut)

Vignette 3 presented the following scenario: funding has been cut for a drug treatment center and, as executive director, you have been asked to decide which clients will be served. The four courses of action offered were: a) discharge/refer based on own judgment, b) advocate for additional funding, c) take no action, and d) hold staff meeting to discuss discharges and/or referrals.

4.3.1 Demographics and course of action

Education level (ChiSq = 18.757, p = 0.027), commitment to social work values (F = 8.826, p < 0.0005), and being Jewish (ChiSq = 8.263, p = 0.041) were related to how individuals felt they would respond to the scenario. Having a Bachelor’s degree, a greater commitment to social work values (M = 4.122), and being non-Jewish were related to choosing option “B” (advocate for additional funding). Having a Master’s degree, a lesser commitment to social work values (M = 3.933), and being Jewish were related to choosing option “D” (hold staff meeting to discuss discharges and/or referrals). Being non-Jewish was also related to choosing option “A” (discharge/refer based on own judgment).

Both education level (ChiSq = 18.973, p = 0.025) and having had formal ethics training (ChiSq = 8.068, p = 0.045) were related to what individuals felt they should do. Those with Bachelor’s degrees tended to select “B” (advocate for additional funding) whereas those with Master’s and Ph.D. degrees tended to select “D” (hold staff meeting to discuss discharges and/or referrals). Those with formal training tended to select “B” (advocate for additional funding) and those without tended to select “D” (hold staff meeting to discuss discharges and/or referrals).

4.3.2 Course of action and rationales

For the “would” response patterns, those who chose options “A” (discharge/refer based on own judgment) and “B” (advocate for additional funding) tended to do so based on codified rationales, and those who chose “C” (take no action) and “D” (hold staff meeting to discuss discharges and/or referrals) tended to do so based upon non-codified rationales. (ChiSq = 17.785, p < 0.0005).

In terms of the “should” response patterns, those who chose “A” (discharge/refer based on own judgment) and “B” (advocate for additional funding) also made those selections on the basis of codified rationales and those who chose “D” (hold staff meeting to discuss discharges and/or referrals) based their decisions upon non-codified rationales (ChiSq = 11.482, p = 0.009).

4.4 Vignette 4 (Immigration Status and Petty Crime)

In sum Vignette 4 presented the following scenario: a refugee resettlement center providing emergency services has been turning a blind eye to the immigration status of clients. The case worker comes to you (Program Manager) regarding an illegal immigrant who may have committed a petty crime. The four courses of action offered were: a) take no action, b) inform the Executive Director of the situation, c) contact appropriate officials, and d) direct the case worker to address the issue.

4.4.1 Demographics and course of action

None of the demographics were significantly related to how individuals felt they would respond to the scenario. Education level (ChiSq = 17.521, p = 0.041) was related to what individuals felt they should do. Those with Bachelor’s degrees tended to select “B” (inform the executive director of the situation), those with Master’s degrees tended to select “C” (contact appropriate officials), and those with Ph.D.’s tended to select “D” (direct the case worker to address the issue).

4.4.2 Course of action and rationale

For the “would” response patterns, those who chose option “B” and “C” tended to do so based upon codified rationales, and those who chose options “A” (take no action) and “D” (direct the case worker to address the issue) tended to do so based upon non-codified rationales (ChiSq = 21.142, p < 0.0005). In terms of “should” response patterns, those who chose option “C” (contact appropriate officials) tended to do so based upon codified rationales and those who chose “A,” (take no action) “B” (inform the executive director of the situation), and “D” (direct the case worker to address the issue) tended to do so based upon non-codified rationales (ChiSq = 16.694, p = 0.001).

4.5 Vignette 5 (Duty to Warn)

Vignette 5 presented the following scenario: you are a therapist for a Vietnam veteran with a history of impulsive antisocial actions who discloses that he is planning to kill his current girlfriend because she is dating another man. No significant relationships were found between demographics and course of action for option “A” (discuss this further at the next session). The other two response options for this vignette are described in the findings below.

4.5.1 Demographics and course of action

Gender (ChiSq = 6.919, p =0.031) was significantly related to how respondents felt they would respond to the scenario. Males were more likely to select “B” (contact his girlfriend and/or the police without informing him) and females were more likely to select “C” (inform him that you must warn his girlfriend and/or the police).

Gender (ChiSq = 7.517, p = 0.023), ethnicity (ChiSq = 10.580, p = 0.005), and being Jewish (ChiSq = 10.409, p = 0.005) were related to how individuals felt they should respond to the scenario. Males, minorities, and non-Jewish respondents were more likely to select choice “B” (contact his girlfriend and/or the police without informing him). Females, non-minority respondents and Jewish respondents were more likely to select “C” (inform him that you must warn his girlfriend and/or the police).

4.5.2 Course of action and rationales

No relationship was found between response choice and codified or non-codified rationale choices for either “would” or “should” responses.

4.6 Vignette 6 (Diagnosis)

Vignette 6 presented the following scenario: you are a worker in an emergency room of a community mental health center about to admit a man best diagnosed as paranoid schizophrenic – you are weighing risks of diagnosing him as schizophrenic, including his potential resistance to hospitalization. The three courses of action offered were: a) do not inform him of risks; give a “milder” diagnosis, b) do not inform him of risks; diagnose as indicated, and c) inform him of risks; diagnose as indicated.

4.6.1 Demographics and course of action

Formal training (ChiSq = 13.664, p = 0.001) was significantly related to how individuals felt they would respond to the scenario. Those with formal training tended to select “C” (inform him of risks; diagnose as indicated) and those without tended to select “B” (do not inform him of risks; diagnose as indicated).

Formal training (ChiSq = 18.397, p < 0.005) and being Jewish (ChiSq = 5.996, p = 0.05) was related to what individuals felt they should do. Those with formal training and those who were non-Jewish tended to select “C” (inform him of risks; diagnose as indicated). Those without formal training and those who were Jewish tended to select “B” (do not inform him of risks; diagnose as indicated).

4.6.2 Course of action and rationales

For “would” response patterns, those who chose option “C” tended to do so based upon codified rationales and those who chose “A” (do not inform him of risks; give a “milder” diagnosis) and “B” (do not inform him of risks; diagnose as indicated) tended to do so based upon non-codified rationales (ChiSq = 6.949, p = 0.031). For “should” response patterns, those who chose option “C” (inform him of risks; diagnose as indicated) tended to do so based upon codified rationales and those who chose “A” (do not inform him of risks; give a “milder” diagnosis) and “B” (do not inform him of risks; diagnose as indicated) tended to do so based upon non-codified rationales (ChiSq = 45.191, p < 0.0005).

4.7 Predictors of Discrepancy Between Would and Should Courses of Action

The overall multiple regression model was significant (F = 1.726, p = .077) and accounted for 8.3% of the variance of “would/should” discrepancy scores. Years of experience, highest degree, Catholic, Protestant, Jewish, gender, and spirituality were not significant predictors of discrepancies in ethical decision-making. However formal training (B = -0.224, p = 0.094), ethnicity (B = 0.225, p = 0.1), and commitment to social work values (B = -0.286, p = 0.045) were significant predictors (see Table 11). The mean score on the POS was 4.03 on a five-point scale, with higher scores representing greater commitment to social work values. Minority status was not related to whether or not one had received formal training (p = 0.827), nor was it related to the respondents’ scores on the POS (p = 0.186). These results indicate that having formal ethics training and a greater commitment to social work values are associated with fewer discrepancies between respondents’ would and should choices, whereas minority status is associated with a greater number of differences.

5. Discussion

The first question this exploratory study sought to answer was, “What personal and professional factors are related to social workers’ courses of action when making ethical decisions?” The analyses of the data suggest that both personal and professional factors are related to ethical decision-making among social workers. Personal demographic factors were related to course of action for would and/or should responses in most vignettes, including vignettes 1) referral/do not respect co-worker dilemma), 2) sexual misconduct dilemma, 3) referral/funding cut dilemma, 5) duty to warn dilemma, and 6) diagnosis dilemma. Personal factors that were related to these vignettes included ethnicity, religion (being Jewish or non-Jewish), and gender. Vignette 5 (duty to warn dilemma) was the only vignette that showed significant relationships between all three of the above mentioned demographic factors and the courses of action selected. Differences in courses of action seemed to relate to whether or not the responding social workers would inform the patient of their duty to warn the girlfriend/police prior to the notification.

Professional factors were also related to course of action for would and/or should responses in most vignettes including, 1) referral/do not respect co-worker dilemma, 3) referral/funding cut dilemma, 4) immigration status and petty crime dilemma, and 6) diagnosis dilemma. Professional factors that were related to these vignettes included commitment to social work values, education, and whether or not respondents had received formal ethics training. Vignette 3 (referral/funding cut dilemma) was the only vignette that showed significant relationships between all three of the above mentioned demographic factors and the courses of action selected. No one personal or professional demographic factor was related to course of action across all vignettes for either would or should responses. In addition an interesting finding in Vignette 3 (Referral/Funding cut) points to the possibility of an inverse relationship between level of social work education and having had formal education and its relationship to how social workers’ feel they should manage caseloads in the face of budget cuts.

The second research question was, “What rationales do social workers use to make ethical decisions?” Data analyses indicated that regardless of demographic influences. social workers tend to make ethical decisions that are sometimes based upon rules and/or codes (codified), and at other times they tend to make decisions based upon other factors rather than rules and/or codes (e.g.. intuition). Unlike in the findings of Smith and colleagues (1991), no clear pattern emerged across all vignettes between the respondents’ course of action and their rationale for taking that particular course of action. Interestingly, in a number of instances, respondents chose more than one course of action in the same vignette and justified each action based upon the use of a codified rationale. For example, in Vignette 3 (referral/funding cut dilemma), those who selected options “A” (discharge/refer based on own judgment) and “B” (advocate for additional funding) did so based upon codified rationales. Conversely in Vignette 1 (referral/do not respect co-worker dilemma) and Vignette 5 (duty to warn dilemma), no relationships were found between social workers’ courses of action and their rationale for choosing that course of action.

The third research question was, “What personal and professional factors are related to discrepancies in ethical decision-making among social workers?” The findings suggest that fewer discrepancies in would/should choices are related to having had formal training, and to greater commitment to social work values, whereas more discrepancies are associated with being a minority rather than a non-minority.

Taken together, these findings indicate that, although there is no clear pattern of ethical decision-making among NASW members in either what they report they would do or in what they think they should do if faced with an ethical dilemma, differences in social workers’ courses of action do exist and appear to relate to both personal and professional demographic factors. The findings support previous literature suggesting that both personal and professional factors should be considered in regard to ethical decision-making (e.g. Csikai, 1999; Smith et al., 1991).

Social workers use both codified and non-codified rationales for dealing with ethical dilemmas. Significant relationships between demographics and courses of action indicate that multiple courses of action are utilized by social workers, and course of action bears some relationship to demographic factors (ethnicity, being Jewish, and gender). However only minorities report a discrepancy between what they would do versus what they think they should do in the face of some ethical dilemmas.

The fact that a greater commitment to social work values and having had formal ethics training are related to fewer discrepancies in ethical decision-making may be indicative of the idea that the more one is adherent to the set of social work values, the more solid a frame of reference one has to draw from in making ethical decisions in practice. However, there may be alternate considerations. Given the relatively inchoate state of research in this area, the reader is cautioned about reaching such a conclusion. As Dolgoff and colleagues (2005) note, some social workers may not acknowledge the ethical dilemmas that arise in practice because they are uncomfortable making ethical decisions, think they already know the answer to the ethical dilemma, or are uneasy dealing with such issues because they do not feel they have the skills to manage ethical problems (Dolgoff et al., 2005). Given Dolgoff and colleagues’ conception, the reasons social workers’ report minimal discrepancies in ethical decision-making remain unclear. Although ideally one would like to think that the social worker has learned to effectively resolve ethical dilemmas, this may not actually be the case. These findings reflect the complexity of ethical decisions given the difficulty in establishing universal prescriptions for behavior. Considering that there is no one right or wrong answer to address any particular dilemma much more information is needed to understand the process and outcomes of ethical decision-making among social workers.

5.1 Strengths

Findings from this exploratory study begin to build a foundation for future inquiry into the multi-faceted process of ethical decision-making. This study’s conceptual base is an important strength particularly given the absence of guiding social work models. Further, the study yielded an adequate response rate from a random, national sample of licensed social workers and as such may be generalizable. In addition, the measure used to assess commitment to social work values (POS) appears to be the most frequently used measure of this construct (Abbott, 2003). Utilization of this measure contributes to the knowledge base in this area.

5.2 Limitations

The limitations of the current inquiry relate to the lack of variability within the sample itself, the simplicity of measures, and the inability to compare vignette rationale choices to right or wrong answers. Although the study population matches that of NASW members (Whitaker, Weismiller, & Clark, 2006), it is not necessarily reflective of all social workers. For example, social workers with religious affiliations other than those identified in the dataset are not represented in this study. In fact members of NASW represented only approximately 38% of all self-identified social workers in the United States in 2004 (Whitaker et al., 2006). The available database limited the ability to examine particular constructs in their full complexity. The following constructs were explored using simple measures: culture/ethnicity (minority status), religion (Catholic, Protestant, Jewish), and training (received training or not). In addition, reliability and validity of the adapted version of the ECSRS remain unknown. And, it is also important to note “would/should” responses on the ECSRS were not compared to any version of right and wrong answers for this study.

5.3 Implications

5.3.1 Education

It is possible that cultural, ethnic, and/or gender differences might impact upon ethical decision-making among social workers. It is critical to consider the impacts of socio-cultural context on educating students regarding ethical decision-making. Given the potential impacts of ethnic/racial background on ethical decision-making, it is essential to create classroom environments that are inclusive, promote ethical self-awareness (Abramson, 1996), promote cultural awareness, and provide space for a diverse body of students to engage with this complex content.

5.3.2 Research

The development of adequate means of measuring the dimensions of ethical decision-making for social workers will be key to increasing our understanding of how social workers’ decisions are impacted upon by their continuing education, practice experiences, cultural backgrounds, and values, both personal and professional. Because ethical decision-making is a complex process rather than a rigidly defined construct (Dolgoff et al., 2005; Mattison, 2000; NASW, 1999), measurement of ethical decision-making is extremely challenging. Developing measures that capture all dimensions of the ethical decision-making process runs the risk of oversimplifying and/or misrepresenting ethical complexity (Walden, Wolock, & Demone, 1990). Attempting to establish “right” answers for the purposes of measurement does not account for ethical theory that suggests that there is usually no “one” right answer to an ethical dilemma.

Social work needs to begin to establish means of measuring ethical decision-making that take into account the complexity of ethical issues and are directly relevant to the circumstances practitioners might face. Vignette measures that focus on scenarios that are particularly relevant to social work practice at all levels might provide a useful starting point. In addition, more sophisticated measures that capture in greater detail the type, amount, quality, and the timing of ethical training will be essential to gaining greater understanding of how training relates to ethical decision-making.

Future research can also build upon this study by utilizing sampling frames that are more diverse in their racial, cultural, religious/spiritual, and gender make-up (e.g., state licensing boards), and that may be more representative of social workers in the United States. Such studies should include samples with larger numbers of cultural and religious minorities, and men in order to make comparisons both within and across groups. Further, the religion variable can be expanded to explore types and degrees (conservative, liberal, orthodox, etc.) of religious or spiritual affiliation. Finally, further research is needed to better understand the reasons why social workers may or may not report discrepancies when responding to ethical dilemmas. The complexity of the content and the scarcity of research suggest a strong need for qualitative research in this area.

5.3.3 Theory

Significant findings related to ethnicity in terms of both course of action and discrepancy scores provide support for Hopkins’ (1997) model. In adapting Hopkins’ model for the purposes of this research, elements that are particularly relevant to the culture of social work, are supported by the existing literature, and are available in the pre-existing dataset, were added. Findings from this study lend support to the need to include ethical training and social work values in the model. A further proposed adaptation would be the inclusion of geography and the nature of one’s practice setting. Continued refinement of the model can provide a foundation for the development of culturally sensitive, social work-specific ways of understanding ethical decision-making.

5.3.4 Practice

Particularly relevant is the fact that a large number of social workers are aging and nearing retirement (NASW, 2006; Whitaker et al., 2006). Accordingly, it will be necessary to recruit new social workers, and it has been argued that recruitment procedures should result in the inclusion of social workers that represent the demographic composition of the United Sates (Whitaker et al., 2006). Considering that minorities tended to reflect more discrepancies in “would/should” responses in this sample, it is imperative that more research in this area is conducted to understand what these differences mean and the impact they might have on the profession. In this vein, it will be important to be open to understanding diversity from the perspectives of minorities. This type of openness to diversity, with a particular focus on values and ethical decision-making, may serve as an asset to social work’s growth and development locally and globally as clients tend to be more ethnically diverse than social workers themselves (Whitaker et al., 2006).

This study adds to the body of social work literature by utilizing an adapted theoretical framework to ground the exploration of patterns in ethical decision-making among social workers. Findings from this exploratory study point to the possibility of new and essential areas of inquiry when considering ethics and its practical application for social work and social work education. The more we as a profession can shed light on the factors that play a role in the complex process of ethical decision-making, the better able we will be to educate our practitioners, and in turn, to serve our clients.

6. References

Abbott, A.A. (1988). Professional choices: Values at work. Silver Spring, MD: NASW.

Abbott, A. A. (1999). Measuring social work values: A cross-cultural challenge for global practice. International Social Work, 42(4), 455-470.

Abbott, A. A. (2003). A confirmatory factor analysis of the Professional Opinion Scale: A values assessment instrument. Research on Social Work Practice, 13(5), 641-666.

Abramson, M. (1996). Reflections on knowing oneself ethically: Toward a working framework for social work practice. Families in Society, 77(4), 195-201.

Boland, K. (2006). Ethical decision-making among hospital social workers. Journal of Social Work Values and Ethics, 3. Retrieved April 2, 2006, from http://www.socialworker.com/jswve

Canda, E. R., & Furman, L. D. (1999). Spiritual diversity in social work practice: The heart of helping. NY: The Free Press.

Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation for the behavioral sciences (3rd ed.). Mahwah, New Jersey: Lawrence Erlbaum Associates.

Csikai, E. L. (1999). The role of values and experience in determining social workers’ attitudes toward euthanasia and assisted suicide. Social Work in Health Care, 30(1), 75-95.

Council on Social Work Education. (2002). Educational policy and accreditation standards. Alexandria: VA: Author.

Dillman, D. A. (2000). Mail and internet surveys: The tailored design method (2nd ed.). New York: John Wiley & Sons.

Dolgoff, R., Loewenberg, F. M., & Harrington, D. (2005). Ethical decisions for social work practice (7th ed.). Belmont, CA: Brooks/Cole.

Dolgoff, R., & Skolnik, L. (1996). Ethical decision-making in social work with groups: An empirical study. Social Work with Groups, 19(2), 49-65.

Dudley, J. R., & Helfgott, C. (1990). Exploring a place for spirituality in social curriculum. Journal of Social Work Education, 26(3), 287-294.

Enoch, Y. (1989). Change of values during socialization for a profession: An application of the marginal man theory. Human Relations, 42(3), 219-239.

Frans, D., & Moran, J. R. (1993). Social work education’s impact on students’ humanistic values and personal empowerment. Arete, 18(1), 1-11.

Freud, S., & Krug, S. (2002). Beyond the code of ethics, Part I: Complexities of ethical decision-making in social work practice. Families in Society, 83(5), 474-482.

Greeno, E. J., Hughes, E. K., Hayward, R. A., & Parker, K. L. (2007). A confirmatory factor analysis of the Professional Opinion Scale. Research on Social Work Practice, 17(4), 482-493.

Haas, L. J., Malouf, J. L., & Mayerson, N.H. (1988). Personal and professional characteristics as factors in psychologists’ ethical decision-making. Professional Psychology: Research and Practice, 19(1), 35-42.

Haynes, D. T. (1999). A theoretical integrative framework for teaching professional social work values. Journal of Social Work Education, 35(1), 39-50.

Hodge, D. R. (2005). Spirituality in social work education: A development and discussion of goals that flow from the profession’s ethical mandates. Social Work Education, 24(1), 37-55.

Hopkins, W. E. (1997). Ethical dimensions of diversity. London: SAGE Publications.

Hoy, J., & Feigenbaum, E. (2005). Ethics in community care: Making the case for ethics consultants in community mental health centers.  Community Mental Health Journal, 41(3), 235-250.

Landau, R. (1999a). Professional socialization, ethical judgment and decision-making orientation in social work. Journal of Social Service Research, 25(4), 57-75.

Landau, R. (1999b). Ethical judgment, code of ethics and supervision in ethical decision-making in social work: Findings from an Israeli sample. Journal of Applied Social Sciences, 23(2), 21-29.

Linzer, N. (1999). Resolving ethical dilemmas in social work practice. Needham Heights, MA: Allyn and Bacon.

Linzer, N. (2004). An ethical dilemma in elder abuse. Journal of Gerontological Social Work, 43(2/3), 165-173.

Manning, S. S. (1997). The social worker as moral citizen: Ethics in action. Social Work, 42(3), 223-230.

Manzo, D. A., & Ross-Gordon, J. (1990). Socialization outcomes of part time graduate professional social work education: A comparison of adult students in career transition to social work with returning adult students who have undergraduate training and practical experience in social work. (ERIC Document Reproduction Service No. ED 372 724).

Mattison, M. (2000). Ethical decision-making: The person in the process. Social Work, 45(3), 201-212.

Miller, D. W. (2001). Programs in social work embrace the teaching of spirituality. Retrieved April 18, 2006, from http://chronicle.com/weekly/v47/i36/36a01201.htm.

NASW. (1999). Code of Ethics of the National Association of Social Workers. Retrieved October 4, 2004, from http://www.socialworkers.org/pubs/code/code.asp

NASW. (2006). Licensed social workers in the U.S., 2004. Retrieved April 20, 2006, from http://workforce.socialworkers.org/studies/natstudy.asp

Perkins, D. V., Hudson, B. L., Gray, D. M., & Stewart, M. (1998). Decisions and justifications by community mental health providers about hypothetical ethical dilemmas. Psychiatric Services, 49(10), 1317-1322.

Pike, C. K. (1996). Development and initial validation of the Social Work Values Inventory. Research on Social Work Practice, 6(3), 337-352.

Reamer, F. G. (1998). The evolution of social work ethics. Social Work, 43(6), 488-500.

Reisch, M., & Lowe, J. I. (2000). “Of means and ends” revisited: Teaching ethical community organizing in an unethical society. Journal of Community Practice, 7(1), 19-38.

Sheridan, M. J., Bullis, R. K., Adcock, C. R., Berlin, S. D., & Miller, P. C. (1992). Practitioners’ personal and professional attitudes and behaviors toward religion and spirituality: Issues for education and practice. Journal of Social Work Education, 28(2), 190-204.

Smith, T. S., McGuire, J. M., Abbot, D. W. & Blau, B. I. (1991). Clinical ethical decision-making: An investigation of the rationales used to justify doing less than one believes one should. Professional Psychology: Research and Practice, 22(3), 235-239.

Spano, R. N., & Koenig, T. L. (2003). Moral dialogue: An interactional approach to ethical decision-making. Social Thought, 22(1), 91-103.

SPSS Inc. (2001). SPSS 11.0.1. Chicago: SPSS Inc.

Stevens, J. P. (2002). Applied multivariate statistics for the social sciences (4th ed.). Mahwah, New Jersey: Lawrence Erlbaum Associates.

Strom-Gottfried, K. (2000). Ensuring ethical practice: An examination of NASW code violations, 1986-97. Social Work, 45(3), 251-261.

Walden, T., Wolock, I., & Demone, H. W., Jr. (1990). Ethical decision-making in human services: A comparative study. Families in Society, 71(2), 67-75.

Walz, T. & Ritchie, H. (2000). Gandhian principles in social work practice: Ethics revisited. Social Work, 45(3), 213-222.

Wesley, S. C. (2002). Proficiency in ethical decision-making. Arete, 26(1), 21-31.

Whitaker, T., Weismiller, T., & Clark, E. (2006). Assuring the sufficiency of a frontline workforce: A national study of licensed social workers. Executive summary. Washington, DC: National Association of Social Workers.

Wodarski, J., Pippin, J., & Daniels, M. (1988). The effects of graduate social work education on personality, values and interpersonal skills. Journal of Social Work Education, 23(3), 266-277.