The Development and Initial Validation of the Empathy Scale for Social Workers

Abstract

Empathy is a core principle essential to social work. Despite this emphasis, minimal empirical research of empathy has been undertaken by social work researchers. The purpose of this study was to develop and initially validate the Empathy Scale for Social Workers (ESSW). The ESSW is a 41-item self-report inventory designed to assess empathy in social work practitioners. The sample (N = 271) consisted of social workers who had attained the Master of Social Work (MSW) degree. Findings revealed promising psychometric properties for the ESSW, and exploratory factor analysis (EFA) demonstrated content, construct, and factorial validity. Results were encouraging and they lay the ground work for the continued development of the ESSW. This scale addresses a gap in social work knowledge regarding the empirical evaluation of empathy. Results have implications for social work as the scale may be used to assess student training needs and/or as a screening tool for social work supervisors and practitioners.

Keywords

empathy scale development social work

Introduction

Direct social work practice by its nature is an interactive, mutual, and dynamic process. Lacking probative technologies and a unifying theoretical authority, social workers rely on a tool kit of diffuse interpersonal skills and techniques to understand unique client needs, and choose appropriate and effective therapeutic interventions. A collaborative helping relationship that enlists inherent client strengths and environmental resources requires practitioners who can readily and accurately detect the nuanced perspectives and emotions of their clients. These core practice skills include reflective listening techniques and empathic inquiry. Despite its acceptance as an important underlying aspect of human personality, interpersonal relationships, and the helping process, empathy’s definition remains undifferentiated. Notably, the understanding and practice of empathy as an important theoretical construct and personal attribute in social work practice has been well documented (Freedberg, 2007; Hepworth & Larsen, 1982; Hollis & Woods, 1981; Lantz, 2001; Rothery & Tutty, 2001; Saulnier, 1996). Ironically, the empirical assessment and measurement of empathy has not been vigorously pursued by social work researchers despite the profession’s anchoring of the construct in education and practice. Gerdes and Segal (2009) provided a recent and notable exception in their theoretical contributions to the exploration of the important role of empathy in social work practice.

Most human behavior and development theories that inform social work place emphasize on the role of empathy in developing a productive relationship between worker and client. Psychodynamic theories such as self-psychology and object relations require an empathic therapist for their delivery (Martin, 2001). Cognitive-behavioral approaches and crisis-intervention models describe an explicit need for empathy skills on the part of the practitioner. Humanistic models such as client-centered therapy (Rogers, 1951; Rothery & Tutty, 2001) and existential social work (Lantz, 2001) also predicate the applications of their theoretical frameworks with empathy and its adjacent interpersonal attributes and abilities.

Social work practice texts are replete with references regarding the importance of empathy in the successful implementation of face-to-face practice with individuals, groups, families, and communities. Students often receive specific training in the use of empathy skills and rehearse these in the classroom and field placements prior to beginning their professional careers (Woods & Hollis, 2000).

The utility of the helping relationship in social work has been repeatedly demonstrated. Indeed, the process skills of the worker provide the potential for change, and have been identified by social work clients as the single most important factor in treatment success. Empathy and other similar constructs repeatedly appear in the literature examining the therapeutic relationship (TR; Coady & Marziali, 1994; Lejuez, Hopko, Levine, Gholkar, & Collins, 2006; Olio & Cornell, 1993). The field emphasizes empathy, tolerance, and genuine concern for the welfare of vulnerable populations in society and social work outcome research continues to support the empathic and intuitive nature of successful social work practice (Smith, Thomas, & Jackson, 2004; Tevithick, 2003).

Previous reviews of available empathy measures (Chlopan, McCain, Carbonell, & Hagen, 1985) revealed no instrument designed to measure empathy for social workers. Other available empathy instruments have limited utility for the measurement of empathy in social workers. The empathy quotient (Baron-Cohen & Wheelwright, 2004), while reliable (α = .92), the scale is specifically designed for clinical populations. The interpersonal reactivity index (Davis, 1983) has been repeatedly subjected to empirical testing (Pulos, Elison, & Lennon, 2004), but it was not designed to assess empathy in a helping relationship. The Jefferson Scale of Physician Empathy (Hojat, 2007) evaluates empathy for physicians and is based on concepts and terminology specific to the practice of medicine. These scales have informed the development of the current scale but were not considered adequate to evaluate empathy for social workers. Notably, the lack of an objective measure of empathy for social workers represents a knowledge gap in the social work practice and research literature given its role as a core and guiding principle in social work.

Statement of Purpose

The purpose of this research was to develop and initially validate a measure, the Empathy Scale for Social Workers (ESSW) to assess empathy among social work practitioners (see Appendix A). It is intended to extend social work research further into the empirical study of one of the essential process skills used by social workers.

The scale was designed to assess the complex structural nature of empathy among social work practitioners. Empathy was considered as a concept with cognitive, affective, and behavioral dimensions (Davis, 1983). This study presents the latent structure of empathy’s three dimensions as supported by six interlocking constructs. Based on an extensive review of current literature, these constructs are: (a) perspective taking (PT; Baron-Cohen & Wheelwright, 2004; Cliffordson, 2002; Davis, 1980), (b) altruism (Batson, Eklund, Chermok, Hoyt, & Ortiz, 2007; Jolliffe & Farrington, 2005; Koss-Chioino, 2006), (c) caring (Danby, 2004; Engster, 2005; Skovholt, 2005), (d) congruence (Freedberg, 2007; Rogers, 1951; Rothery & Tutty, 2001), (e) interpersonal sensitivity (IS; Carney & Harrigan, 2003; Hall & Mast, 2007; Snodgrass & Rosenthal, 1985), and (f) the TR (Rogers, 1951; Rothery & Tutty, 2001; Wickman & Campbell, 2003). They are proposed as expressions and theoretical evidence of empathy in social workers.

Theoretical Framework

Interest in empathy led researchers to study its underlying properties and attempt to quantify its dimensions (Davis, 1980; Dymond, 1949; Hogan, 1969; Hojat, 2007; Mehrabian & Epstein, 1972). Empathy is normally defined as having two primary dimensions, affective and cognitive (Cliffordson, 2002). The framework described in Figure 1 illustrates a model of empathy extending the primary cognitive and affective dimensions by incorporating its behavioral manifestations. Recently, theoretical inquiry into the nature of empathy in social work practice has considered the essential role of clinical expressions of the central concept (Gerdes & Segal, 2009). In Figure 1, these three dimensions are operationalized into six underlying constructs that provide a proposed definition of empathy in this study. This process required an initial comprehensive theoretical scaffolding to begin to document evidence for the construct validity of this measure. This requisite step in objective scale development has been outlined by researchers (Cronbach & Meehl, 1955; Nunnally, 1967; Rubin & Babbie, 2005; Spector, 1992). The creation of a comprehensive theoretical framework permits test items to be drawn from a universe of concepts related to the latent structure under scrutiny. An exhaustive approach to scale item creation and selection enhanced the likelihood that as many aspects of empathy as possible would be incorporated into the design of the ESSW.

Figure 1.

Conceptual framework of empathy in social work practice.

The six constructs depicted in Figure 1 are (a) caring, (b) congruence, (c) IS, (d) PT, (e) altruism, and (f) the TR. They were taken from the literature in social work and related disciplines and provide both theoretical and empirical evidence of the existence and importance of empathy in social work practice. Just as the affective, cognitive, and behavioral dimensions of empathy share components in their manifestations, so do these six specific nonmutually exclusive constructs.

It is difficult to imagine affective, cognitive, and behavioral components of empathy ever being “stand alone” concepts. Empathy, by its interpersonal and dynamic nature, is something we think, feel, and do. In clinical practice, the use of empathy is indeed a highly complex and dynamic process and its use in social work often requires the practitioner to have all three dimensions “turned on” at once.

Each of the constructs outlined in this theoretical framework has affective, cognitive, and behavioral components and manifestations. They have been determined, for the purposes of this study, to be primarily affective, cognitive, or behavioral. For example, in this study, IS is grouped within the cognitive dimension. The thoughtful attunement to nonverbal communications from clients (a part of IS) will often identify illicit intense emotional material within the clinical practice relationship, and reflection or inquiry into this material by the therapist then by definition becomes a behavioral extension of IS. As such, the use of IS is primarily cognitive in that the nonverbal cues of the client are compared to an internal and established clinical knowledge base. Limited eye contact may communicate deference, anger, or psychosis depending on the social context and interpersonal and intrapsychic dynamics of the interaction. The IS in a helping relationship depends on the perceptiveness and content knowledge base of the clinician. Likewise, altruism is behavior rooted in emotional as well as cognitive motivations, and while these affective and intellectual components are crucial to our understanding of altruism, it remains primarily a behavior.

The six constructs underlying a definition of empathy proposed here have varied meanings and connotations to wide and disparate audiences. Terms like congruence (Rogers, 1951) and IS (Snodgrass, Hecht, & Ploutz-Snyder, 1998) have been redefined by theorists and researchers, with unique meanings for a specific field or discipline. This process further complicates the definitive categorization of these concepts. The exploratory aspects of this study aim to shed light on the degrees of difference and similarity between these terms and constructs.

The Affective Dimension

The affective dimension of empathy is an interactive process of emotional connection and concern for others. It involves emotions defined by how a person feels in the context of an interpersonal experience. Perception of the emotional world of another and an emotionally empathic approach to helping are manifested by two supporting constructs, caring and congruence.

Caring

Caring as a theoretical construct has been developed and refined in part, within the disciples of nursing, philosophy, and counseling. It has been defined as behavior directed at meeting the immediate needs of another by the use of a discipline-specific skill set (Lee-Hsieh, Kuo, Tseng, & Tuton, 2004; Skovholt, 2005). Alternative theories of this notion stress the importance of caring as a social construct with implications in moral and ethical decision-making processes (Danby, 2004).

Caring as both a goal-oriented interactional behavior and theoretical construct is notably absent in the empathy scale development literature (Davis, 1980; De Kemp, Overbeek, de Wied, Engels, & Scholte, 2007; Hashimoto & Shiomi, 2002; Hojat, 2007; Mehrabian & Epstein, 1972). By contrast, other related constructs such as empathic concern, PT, and compassion repeatedly appear in the literature on caring (Danby, 2004; Engster, 2005; Skovholt, 2005). Clearly, caring is an emotional and interactive process that taps the affective components of helping relationships and a relational connection to the feelings of another lays the groundwork for the experience and expression of caring (Brenner & Wrubel, 1989; Engster, 2005).

Congruence

Congruence is defined as an ability to be open, nonjudgmental, and honest within helping relationships (Tudor & Worrall, 1994). Congruence was proposed by Rogers (1951) to be a “core condition” of an empathic and productive therapist–client relationship. Congruence is a therapeutic and emotional connection associated with positive outcomes in psychotherapy (Marziali & Alexander, 1991). It requires a therapist to communicate verbally as well as nonverbally that he or she is a learner anticipating understanding the client’s unique situation and perspective (Wickman & Campbell, 2003). Social work theorists and practitioners have continued to embrace congruence and empathy as essential change facilitating skills (Freedberg, 2007; Rothery & Tutty, 2001).

Research regarding the measurement of empathy has captured elements of congruence without explicating it as a unique construct in its own right (Davis, 1980; Hogan, 1969; Hojat, 2007; Mehrabian & Epstein, 1972). Congruence and empathy are cooccurring conditions with behavioral similarities and have been described as overlapping conditions of an effective TR (Allen-Meares & Burman, 1999; Houston, 1990; Lambert & Barley, 2001; Rogers, 1951; Tudor & Worrall, 1994; Wickman & Campbell, 2003). Congruence is both an interactive and emotional process, it is a helper’s resonance with the feelings of another, and it supports a relationship from which healing can emerge (Freedberg, 2007; Rogers, 1951).

The Cognitive Dimension

Cognitive dimensions of empathy involve interpersonal perception, intellectual flexibility, and openness to understanding the experiences of another in the service of helping. It includes a group of conceptual processing and thinking skills that utilize a level of objectivity and distance from the emotional content evident in a client’s presentation and a careful assessment of the contextual cues therein.

Interpersonal sensitivity

The IS is a communicative process between individuals based on their understanding of one another’s body language and facial expressions (Snodgrass & Rosenthal, 1985). This skill varies considerably between individuals and is influenced by both social context and gender role expectations. The IS contributes to intimate relationship success and effective helping relationships (Snodgrass et al., 1998).

The IS is considered a necessary but insufficient condition for empathy. One cannot be empathic without being interpersonally sensitive, but sensitivity does not guarantee empathy (Carney & Harrigan, 2003). The IS can aid in the assessment of a client’s nonverbal behaviors, and it creates an opportunity for exploration of issues not initially presented by the client (Snodgrass & Rosenthal, 1985). Hall and Mast (2007) identified nonverbal communication and IS as confirmatory and complementary processes in one’s development of accurate empathy.

The IS is a cognitive exploration that injects the helping process with a level of objectivity in understanding the contextual but often unspoken nature of a client’s concerns (Carney & Harrigan, 2003; Underwood & Moore, 1982).

Perspective taking. PT is defined as the ability to accurately perceive another’s point of view (Davis, 1980). It involves the internal and cognitive interpretation and understanding of another’s mental, emotional, and situational status. It becomes necessary to suspend one’s own perspective or worldview and seek to understand the environmental issues and factors contributing to the thoughts and feelings of someone else (Baron-Cohen & Wheelwright, 2004; Cliffordson, 2002; Davis, 1980; Hojat, 2007; Johnson, Cheek, & Smither, 1983). Dymond (1949) defined empathy as “the imaginative transposing of oneself into the thinking, feeling, and acting of another, and so structuring the world as he does” (p. 127).

PT has been described as an essential part of the development and expression of empathy. Research has demonstrated the complex processes and relationship between PT and empathy (Oswald, 1996). Factor analytic studies of empathy measures suggest PT to be an underlying function of empathy as well (Cliffordson, 2002; Davis, 1980; Hojat, 2007; Johnson et al., 1983). Theoretical and empirical study has consistently identified PT as an important facet of empathy in professional helping relationships (Hojat, 2007). Flexible and objective attunement to the details of a client’s perspective is required to be of help. By taking a client’s perspective, a clinician enhances the chances of employing successful assessment and intervention strategies (Davis, 1980; Saulnier, 1996).

The Behavioral Dimension

Intuitively one recognizes that behavioral manifestations of empathy involve interpersonal actions and motivations. These are other-directed and outwardly observable expressions of empathy and demonstrate functional aspects of the concept and its concrete applications within helping relationships. The two constructs (in Figure 1) related to this dimension are altruism and the TR.

Altruism

Altruism has been described as a prosocial behavior intended to help or assist another individual (Bierhoff & Rohmann, 2004). Altruism can take the form of efforts to relieve distress, such as helping someone to stand after a fall, or goal-directed behavior such as opening a door for another. Altruistic behavior is distinct from collaboration in that all expect to benefit from a collaborative effort. Altruism has been identified as a behavioral indicator of empathy (Batson et al., 1991; Underwood & Moore, 1982). The role of motivation in altruistic behavior is essential to understanding the concept and the empathy–altruism hypothesis posits that altruistic behavior is motivated by an individual’s concern for the welfare of another as an outgrowth of empathic expression (Batson et al., 1991, 2007). Additionally, this hypothesis suggests that altruistic behavior is not driven by alternative motivations such as guilt within the helper or the anticipation of the rewarding experience of seeing change in the recipient (Batson et al., 1991).

While altruism serves as a behavioral indicator of empathy, most empathy measurement research does not attend to the relationship between altruism and empathy (Davis, 1980; Dymond, 1949; Hogan, 1969; Hojat, 2007; Mehrabian & Epstein, 1972). Empirical research studies on empathy have repeatedly found an inverse relationship between antisocial behavior and empathy (De Kemp et al., 2007; Jolliffe & Farrington, 2005). As the relationship between antisocial behavior and empathy helps define what empathy is not, the prosocial behavior of altruism should be a part of the theoretical network defining what empathy is. The empirical evidence of a relationship between the concepts helps to explain the complex theoretical overlap between them. Given the centrality of altruism to the character of social work, its role in the assessment of empathy in practitioners is warranted (Bulke, 1994; Holosko, 2006; Pins, 1963).

The therapeutic relationship

The TR has been identified as a significant agent of change and growth in a variety of helping relationships and clinical settings (Allen-Meares & Burman, 1999; Lambert & Barley, 2001; Olio & Cornell, 1993; Stewart, 1984). This consists in part, of a sense of trust and a bond between the therapist and client (Dykeman, Nelson, & Appleton, 1995). A TR is one in which a worker is accepting, nonjudgmental, supportive, and empathic. Other identified defining characteristics of TR are effective affirmation skills, caring, and respect (Allen-Meares & Burman, 1999; Lambert & Barley, 2001).

The importance of TR is a central construct in most theories of human behavior and therapeutic change. Psychodynamic theorists and researchers describe this alliance as a foundation for explaining therapeutic benefits of clinical intervention (Kradin, 2005). Theorists from existential, feminist, behavioral, and family systems schools of thought often center their models around an empathic TR (Lantz, 2001; Lejuez et al., 2006; Minuchin & Fishman, 1981; Saulnier, 2001). Empathy is a distinct process component of the worker–client relationship that is embraced by most theoretical therapeutic models of change. Freedberg (2007) stated, “I am suggesting that an enhanced feeling of power grows out of a healthy interaction with empathically attuned others, contributing to the capacity to act in the environment with a sense of self-efficacy and purposefulness” (p. 256).

Empathy and its ultimate behavioral and theoretical expressions appear in studies of TR. Accordingly, measures of the helping relationship have been successfully used in the evaluation of therapeutic outcomes (Coady & Marziali, 1994; Marziali & Alexander, 1991; Smith et al., 2004). While a reflective and thoughtful process, TR is an outwardly focused behavior initiated by the helper and it requires motivation and direct action by the clinician to provide a relational and behavioral vehicle for change.

In sum here, six constructs as noted above are associated with the concept of empathy (King, 2009). All have empirical and theoretical justification in the literature and all were included in the development and initial validation of the ESSW to assess empathy among social worker practitioners who use this skill in their day-to-day practice.

Method

Scale Construction

The initial items on the ESSW were developed using scale construction methods through three steps. First, a review of the literature from prior research on empathy and its related constructs was conducted. This provided descriptions of empathy that formed a large pool of terms and phrases from which scale item stems were developed. Second, items from existing general empathy scales were reviewed, and phraseology related to the concept of empathy was incorporated into the construction of complete social work practice-related scale items. Each of the six latent constructs of empathy (in Figure 1) was allotted 6–7 items each, resulting in a 41-item scale for initial testing. Third, to establish a level of content validity for the ESSW, expert reviewers were recruited to review the scale. All reviewers were social work practitioners with advanced clinical licensure (licensed clinical social worker [LCSW]) in the state of Georgia (N = 20). These initial pilot data were used to refine the scale items but were not included in the final data analysis.

The reviewers evaluated the ESSW using the following three criteria-based questions, how accurate is this scale in describing empathy in social work practice? Are the scale items worded coherently? Does the scale appear to be useful for social workers? Reviewers were asked to rate the scale according to these criteria using a 10-point scale, “1 = not at all” to “10 = very much.” Participants were also asked to identify items on the scale that they thought were worded poorly or confusing. Amendments were then made to the pretested scale.

Sample

The study used a nonprobability, purposive sampling technique, an approach common in the development of measurement instruments that seek to assess a specific group’s attitudes and underlying beliefs (Rubin & Babbie, 2005). The sample consisted of social work practitioners with at least 2 years of direct social work practice experience, and a Master of Social Work (MSW) degree. This inclusion criterion sought to capture a sample of social workers who had enough clinical experience to have had sufficient opportunities for exposure to the use of empathy in practice. Study participants were sought from a list of MSW graduates from the University of Georgia (UGA), School of Social Work between the years 1975 and 2005. Study participants were also recruited from the Georgia Society for Clinical Social Work. All participants’ response data were pooled for the final data analysis. The final sample size of the study was 271 (N = 271). As the purpose of the study was to evaluate empathy among social workers; the sample was discipline specific and relatively homogeneous.

Study Design

This study is a quantitative-descriptive research design. This design is appropriate, when the research goal is the development of an objective rating scale (Holosko, 2006). Specific objectives designed to further the purpose of the study were to describe and quantify the relationships between various study variables. In this case, a single, relatively homogenous, group of participants was sought for this study. The scale was electronically formatted by the Survey Research Center (SRC) at the UGA.

Data Collection

After study approval by the Institutional Review Board at UGA, the ESSW was made available electronically to potential participants, via a web link through the SRC at UGA. No specific incentives were offered to study respondents. Participants then accessed the link and were presented with an introductory cover letter informing them of their rights as research subjects including the lack of assured anonymity, when participating in an online survey. To ensure the confidentiality of participants no personal identifying information was requested by the researcher, and all participants were asked to provide general demographic information only in order to better describe the overall sample. Demographic data were gender, age, ethnicity, years of practice experience, and current practice setting. Participants then moved on to completing the ESSW and the additional validation scales. Once the survey became available via the Internet and participants were solicited, data collection began and continued for a period of 8 weeks from April 1, 2009, until June 1, 2009.

Instruments of Study

The ESSW contains 41 items describing thoughts, feelings, and actions involved in the use of empathy in social work practice. Participants are asked to respond to the items on a Likert-type frequency scale. Theoretically, higher scores reflect higher levels of empathy. The response format for the ESSW is 1 = never, 2 = rarely, 3 = sometimes, 4 = often, and 5 = always. The scale contains four reverse scored items. These are Items 25, 27, 29, 32, and 40 (King, 2009).

Participants also completed The Jefferson Scale of Empathy (JSE; clinical social work version). This scale was adapted (with the author’s permission) by the research team from The Jefferson Scale of Physician Empathy (JSPE; Hojat, 2007). The JSPE is a 20-item self-report questionnaire designed to assess empathy among medical doctors and has strong psychometric properties, α = .92 (Hojat, 2007). The scale was adapted, with feedback from the original author, using language more consistent with the direct practice of social work. The JSE (clinical social work version) uses a response set ranging from 1 = strongly disagree to 7 = strongly agree. This scale was used as a convergent validity check on the ESSW.

The Self-Report Altruism scale (SRA) is designed to assess altruistic behavior in adults. The SRA is a 20-item questionnaire with high reliability, α = .80, and acceptable validity correlations with related measures and was used as an additional construct validity measure for the ESSW (Rushton, Chrisjohn, & Fekken, 1981).

Finally here, study participants completed the Verbal Aggression Scale (VAS). This scale measured the degree to which someone uses verbal attacks and manipulation to influence others. High scores on the VAS reflect an interactive style in significant contrast to the constructs and concepts used in this study to define empathy. The VAS is a 20-item questionnaire with strong reliability, α = .80, and moderate validity correlations with related measures. It was used as a discriminant validity check on the ESSW. The response sets for this scale are 1 = almost never, 2 = rarely, 3 = sometimes, 4 = often, 5 = almost always (Infante & Wigley, 1986).

Data Analysis Procedures

Data were analyzed using the Statistical Program for the Social Sciences, 16.0 (SPSS). The data were examined for missing data entries for each respondent. Data from respondents who did not complete all sections of the survey were not included in the final analysis. This resulted in a final sample size of 271. Frequency and descriptive statistics for the demographic data were computed and the internal consistency of the ESSW was determined using Cronbach’s coefficient alpha (α). An item analysis was conducted to examine the distribution of specifically problematic items. The correlation matrix was analyzed using exploratory factor analysis (EFA) techniques. Variable coefficients (factor loadings) were computed and revealed the extent to which specific items grouped together to form factors. Due to the latent and undifferentiated nature of the construct being evaluated, and the theoretical overlap between items, an oblique rotation of factors was used to clarify the factor structure of the ESSW as is preferred when evaluating the factors of latent human capacities that are expected to correlate (Gorsuch, 1983; Kline, 1994).

Results

Sample Demographics

The study sample (N = 271) was 80.9% female, 85.3% White, 7.5% African American, 2.3% Hispanic, 1% Asian, and 3% other. Participants ranged in age from 24 to 77 (M = 44.5), with a median age of 45. In terms of length of experience, 55% had 10 or more years, 25% had 5–10 years, and 20% had 2–5 years of experience. Participant areas of social work practice included: medical social work, 17%; social work education, 12.5%; community mental health, 11%; private practice, 9%; school social work, 8%; administration, 6.5%; child welfare, 4.5%; community organization, 4%; research, 4%; and other, 22%.

Psychometric Properties of the ESSW

The highest possible score is 210. The total scores of respondents on the scale ranged from 134 to 188 (M = 159, SD = 9.23). Each scale item on the ESSW had a possible high score of 5 and descriptive analyses of all items revealed a range of (M = 3.0–4.6, SD = .514–1.11). The internal consistency of the ESSW, assessed using Cronbach’s α was .83. The most highly endorsed items are presented in Table 1 to provide examples of scale items with promising yet preliminary distribution patterns, and to identify items in need of additional evaluation. Consequently, two of these items (25 and 41) were included in the item analysis below.

Table 1.

The 15 Most Highly Endorsed Items on the ESSW: Means and Standard Deviations (N = 271)

	M	SD
1. I try to let my clients know I am concerned for their welfare	4.64	.538
2. It is important for my clients to be able to trust me	4.62	.571
3. I try to give my clients a warm greeting when meeting them	4.62	.571
4. Having an intimate relationship with a client is appropriate (reverse scored)	4.61	1.110
5. If a client cannot afford treatment I try to find a way for them to receive the help they need	4.57	.572
6. I am kind to my clients	4.54	.514
7. My clients tell me I can be insensitive (reverse scored)	4.51	.570
8. I try to take a client’s cultural context into account when working with them	4.50	.577
9. I try to understand a client’s viewpoint before making suggestions	4.43	.572
10. I enjoy helping people	4.40	.561
11. It is important for my clients to know that I care about them	4.38	.649
12. I can disagree with a client and still appreciate their position	4.37	.572
13. I am attentive to my clients’ nonverbal cues	4.35	.536
14. An unbiased approach is helpful to clients	4.34	.652
15. I pay close attention when a client’s tone of voice changes	4.31	.604

Note. ESSW = Empathy Scale for Social Workers.

Responses: 1 = never; 2 = rarely; 3 = sometimes; 4 = often; 5 = always.

Item Analysis

An item analysis was conducted to evaluate specific scale items for future retention or deletion from the ESSW. Item frequency and distribution values were computed and evaluated to empirically inform this decision-making process. The skewness and kurtosis values of items are indicators of the normality of its distribution. Skewness and kurtosis values outside of the 2.0 to −2.0 range are generally considered problematic and suggest a closer examination of an item for evidence of its appropriateness for retention. This process revealed three items meeting these criteria.

Each of these items (20, 25, 26, and 41) were subjected to further analysis to determine how they were affecting the overall coefficient α of the ESSW. Examination of item 20, “I can put aside my feelings and listen attentively to a client” revealed kurtosis = 3.38 with 67.3% of participants responding “always” and 24.6% endorsing “often.” When this item was deleted the Cronbach’s α improved to .839 (King, 2009).

Item 25, “I am careless when working with clients,” had a kurtosis value of 2.77 and 60.3% of participants responded “rarely” and 32.7% responded “never.” When this item was deleted from the scale, the coefficient α improved to .835 (King, 2009).

Item 41, “Having an intimate relationship with a client is appropriate,” had a kurtosis value of 6.10 and 93% of respondents endorsed “never.” Eliminating this item improved the overall α of the scale to .840 (King, 2009).

In addition to the distribution concerns for Items 25 and 41, they are intuitively problematic. The wording of these items makes them subject to potential social desirability bias as it is unlikely practicing social workers would favorably endorse either item due to the phraseology of behavior detrimental to client welfare and a productive helping relationship. These findings certainly warrant the removal of these two items from the scale.

Correlations With Additional Measures

The following additional measures were utilized to assess the significance and direction of the scales’ relationships with each other. This procedure was meant to evaluate the construct and discriminant validity of the ESSW. Pearson correlation coefficients were computed for these scales’ relationships with one another and the level of significance in each.

It was hypothesized that the ESSW would correlate with both the JSE and the SRA in a positively direction and significantly. The construct validity of the ESSW is strengthened by its significant relationship with the JSE. This provides evidence that the ESSW was measuring a closely related but different theoretical construct. The comparison of the ESSW against the VAS was hypothesized to correlate in a negative direction with the ESSW to a significant degree as well. Table 2 displays the results of these analyses.

Table 2.

Correlation Coefficients Establishing Convergent and Discriminant Validity of the ESSW (N = 271)

Scales	ESSW	JSE	SAS	VAS
Empathy Scale For Social Work (ESSW)	1.00
Jefferson Scale of Empathy (JSE)	.335^**	1.00
Self-Report Altruism Scale (SAS)	.120^*	.067	1.00
Verbal Aggressiveness Scale (VAS)	−.204^**	−.238^**	−.033	1.00

^*Correlation is significant at the .05 level (two-tailed).

^**Correlation is significant at the .01 level (two-tailed).

Factor Analyses

The initial EFA and subsequent examination of factor eigenvalues and the Scree plot suggested that a three-factor solution could be the best fit for the data. A second EFA was conducted including an Oblimin rotation to achieve a factor pattern matrix (N = 271). The item criteria for retention on a factor included interpretation based on theory, interpretability of the factor, a minimum of three items loading (>.35) on each factor, and the examination of the Scree plot. As items grouped into factors, three distinct clusters of items were identified. These factors reflected the underlying elements of the six constructs posited to comprise empathy in social work practice.

The first factor was named “a compassionate contextual assessment” (CCA), and it describes a framework for understanding the experience of receiving and delivering social work services. The second factor identified in the analyses was “an accepting and attentive collaborative inquiry” (ACI), and it described the relationship style and quality inherent in direct social work practice. The final factor was named “intrinsic helping and emotional support” (IHS), and reflected behavioral expressions of caring and altruism in an empathic helping experience (Table 3 ).

Table 3.

ESSW Factors and Salient Loadings (>.35; N = 271)

ESSW Factors and Items	CCA	ACI	IHS
Compassionate contextual assessment (CCA)
I can disagree with a client and still appreciate their position	.372
I put aside my own feelings to listen attentively to a client	.443
I pay close attention when a client’s tone of voice changes	.466
I am careless when working with clients (reverse scored)	.372
I am able to put aside my own feelings to be in accordance with a client’s emotions	.400
I try to take a client’s cultural context into account when working with them	.607
I try to understand a client’s viewpoint before making suggestions	.538
I am attentive to my clients’ nonverbal cues	.489
I carefully consider the ways that social gender role expectations affect my clients	.448
If a client cannot afford treatment I try to find a way for them to receive the help they need	.439
Accepting and attentive collaborative inquiry (ACI)
Knowing a client’s personal situation is important if I am really going to help them		.375
My relationship with a client can help them overcome their problems		.632
It is important for my clients to be able to trust me		.376
It can be helpful for clients to use our relationship to practice new interpersonal skills		.633
It is important for my clients to know that I care about them		.514
The personal dynamics of my relationship with a client are beneficial to the treatment process		.613
My relationship with a client can be therapeutic in and of itself		.586
I try to help clients even if they have not sought treatment voluntarily		.403
Intrinsic helping and emotional support (IHS)
I enjoy helping people			.639
I enjoy helping people even when I am not at work			.628
Helping clients is rewarding in and of itself			.600
I am kind to my clients			.420

Note. ESSW = Empathy Scale for Social Workers.

Conclusions

Empathy appears to be a complex, interactive, and dynamic grouping of affective, cognitive, and behavioral components that manifest themselves in social work practice. The ESSW is an important first step in the objective evaluation of empathy among social workers. Direct social work practitioners identified the concepts introduced here as relevant and influential in their use of empathy in practice.

The psychometric properties of the ESSW lend empirical support to the study’s theoretical claim that empathy consists of a variety of discreet but overlapping constructs and expressions similar to other latent human abilities and attributes. The internal consistency of the instrument was strong, indicating the measure is indeed tapping a commonly understood concept by study participants labeled as empathy in this study.

Factor analyses reinforced the idea that empathy in social work practice consisted of various expressions, rather than a singular all-inclusive concept. As social work is an applied profession, and this study involved the evaluation of social work practitioners and the factors were named in a way to reflect empathy as an active and dynamic process rather than a static and isolated concept. As is the case in this study, during the initial validation stages of scale development, naming factors is an inexact process requiring replication studies to define factors with increasing specificity (Gorsuch, 1983; Kline, 1994).

The first factor was named “a CCA,” and it described a framework for understanding the experience of receiving and delivering social work services. It contained items drawn from the theoretical construct PT. For example, Item 8 “I can disagree with a client and still appreciate their position,” and Item 31 “I try to understand a client’s point of view before making suggestions,” represent the contextual assessment aspects of this factor. Item 28, “I lose track of what a client is telling me,” (reverse scored) is an example of items representing the concept of congruence and Item 25, “I am careless when working with clients,” (reverse scored) is an example drawn from the construct of caring. These items influenced including the term compassionate, when naming the factor. The terms assessment and compassionate “bookend” the term contextual in Factor 1. Assessment describes what action a practitioner is taking, and compassionate connotes the interpersonal approach of how services are rendered in the practice of social work. The term contextual represents the perspective or position from which the practitioner considers the client’s individual circumstances. Social work theory emphasizes human behavior in the context of the social environment and makes this identification of this first factor appropriate and understandable.

The second factor identified was “accepting and attentive collaborative inquiry.” This also described a method of social work intervention, marked by the emphasis on the concept of a TR with clients. Item 13, “My relationship with a client can help them overcome their problems,” and Item 19,” The personal dynamics of my relationship with a client are beneficial to the treatment process,” are both examples of the TR construct and reflect a “collaborative inquiry.” The terms “accepting and attentive” are representative of the construct of caring in part by Item 14, “It is important for my clients to be able to trust me,” and Item 17, “It is important for my clients to know that I care about them.” Item 2, “Unconditional acceptance helps clients,” represents the concept of congruence.

The third factor was “intrinsic helping and emotional support.” It consisted of items representing caring and altruism constructs. For example, Item 4, “I enjoy helping people,” and Item 11, “Helping clients is rewarding in and of itself,” theoretically represent the concept of altruism. The role that altruism plays as a consistent personality marker and behavioral expression of the social work profession has been explored and empirically documented.

Study Limitations

The diffuse nature of empathy in social work practice makes it difficult to measure with specificity and is an inherent limitation of this research. The six constructs identified to underlie empathy are not unique to social work and take on differential meanings in the general population and other helping professions.

The choice of an electronic survey as a tool for data collection can be problematic as well. Survey research is susceptible to a level of bias as findings may be influenced by the responses of most motivated participants. The study survey was lengthy with four different scales with approximately 100 items. This could result in an exhaustion or capitulation effect among respondents. As noted before the sample was a highly homogenous group and would be inclined to answer items regarding empathy and social work practice in similar ways. Given the theoretical and educational emphases on empathy in clinical practice, a level of social desirability bias could have influenced these study findings.

Implications for Social Work

As practitioners determine which area of practice is the “best fit” for them as people and professionals, the ESSW may provide objective information in this decision-making process. This opportunity for self-evaluation may assist practitioners as a tool for practice and career choice decisions, as well as continuing education and supervision needs in the field. Frequently, social workers find themselves in supervisory positions of social service agencies and direct care staff. Many social work practitioners are well suited for these positions, due in part to the inclusive communication and contextual orientation and skills that they bring to clinical practice settings. Often, the skills of practitioners (such as empathy) can enhance the effectiveness and overall success of an agency supervisor, and the agency overall. The ESSW has utility as a screening and self-evaluation tool for potential social work supervisors, and their ability to balance and differentiate the types of empathy needed while supervising clinicians and staff.

As noted previously, empathy has been defined as essential for effective social work practice by its virtually universal emphasis in social work education. The ESSW can function as one of a group of self-evaluation and objective screening devices for social work students and practitioners. If empathy is central to social work practice, then social work students and practitioners lacking the capacity for empathy in general, or regarding specific client groups may be identified and the information gained by the use of the ESSW can inform schools of social work admission procedures, as well as student field placement assignments. Social work practice supervisors would be able to make more empirically informed decisions about clinical assignments and practitioner staffing patterns in agency settings with the use of the ESSW.

The ESSW may provide social work academics and researchers with information that brings the nature of social work into greater focus. It provides empirical research data regarding the constructs that underlie empathy. Understanding the processes by which empathy functions in social work practice brings the concept to the forefront, in explaining how and why practitioners deliver effective services. Through repeated testing of the instrument, the ESSW can become a mechanism for the generation of an informed and improved social work theory and its application in direct practice settings.

Social researchers from a variety of disciplines would benefit from the data generated by future administrations of the ESSW. The ESSW could function as an effective construct and divergent validity tool in the development of scales designed to measure similar or differentiated concepts. Researchers interested in the complexities and dynamics of human cognition, communication, and relationships could benefit from the continuing refinement and development of the ESSW for use in other helping professions. Adaptations of the ESSW would provide insight into the similarities and differences in the constructs underlying empathy in clinical populations as varied as the developmentally disabled, juvenile offenders, substance abusers, sex offenders, and child abuse victims.

The implications for social work practice outlined above represent the potential the ESSW has to impact the profession of social work and influence future research efforts. The use of varied research methodologies could be helpful in contributing to a deeper understanding of the nature of empathy in social work. Empirical outcome studies in the teaching of empathy skills to different groups of students and practitioners could extend further the clinical and research utility and capacities of the ESSW. Research inquiries that facilitate the development of an even more detailed and comprehensive picture of the components of empathy could provide important data for inclusion in the ongoing development of the ESSW. Finally, this process would further an understanding of how and why empathy figures as prominently as it does, not only in social work practice but also as a unique and varied human ability and attribute.

Footnotes

A full copy of the ESSW and its scoring protocol may be obtained by contacting the first author.

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

The authors received no financial support for the research, authorship, and/or publication of this article.

Appendix A

References

Allen-Meares

Burman

(1999). Cross-cultural therapeutic relationships: Entering the world of African Americans. Journal of Social Work Practice, 13, 49–57.

Baron-Cohen

Wheelwright

(2004). The empathy quotient: An investigation of adults with Asperger’s syndrome or high functioning autism, and normal sex differences. Journal of Autism and Developmental Disorders, 34, 163–175.

Batson

D. C.

Batson

J. G.

Slingsby

J. K.

Harrell

K. L.

Peekna

H. M.

Todd

M. R.

(1991). Empathic joy and the empathy-altruism hypothesis. Journal of Personality and Social Psychology, 61, 413–426.

Batson

D. C.

Eklund

J. H.

Chermok

V. L.

Hoyt

J. L.

Ortiz

B. G.

(2007). An additional antecedent of empathic concern: Valuing the welfare of the person in need. Journal of Personality and Social Psychology, 93, 65–74.

Bierhoff

Rohmann

(2004). Altruistic personality in the context of the empathy-altruism hypothesis. European Journal of Personality, 18, 351–365.

Brenner

Wrubel

(1989). The primacy of caring. Melno Park, CA. Addison-Wesley Publishing

Bulke

G. M.

(1994). An exploration of factors that motivate career choice in social work among a sample of undergraduate social work students (Unpublished doctoral dissertation). University of Windsor, Windsor, ON

Carney

D. R.

Harrigan

J. I.

(2003). It takes one to know one: Interpersonal sensitivity is related to accurate assessments of others’ interpersonal sensitivity. Emotion, 3, 194–300.

Chlopan

B. E.

McCain

M. L.

Carbonell

J. L.

Hagen

R. L.

(1985). Empathy: Review of available measures. Journal of Personality and Social Psychology, 48, 635–653.

10.

Cliffordson

(2002). The hierarchical structure of empathy: Dimensional organization and relations to social functioning. Scandinavian Journal of Psychology, 43, 49–59.

11.

Coady

N. F.

Marziali

(1994). The association between global and specific measures of the therapeutic relationship. Psychotherapy, 31, 17–27.

12.

Cronbach

L. J.

Meehl

P. E.

(1955). Construct validity in psychological tests. Psychological Bulletin, 4, 281–300.

13.

Danby

(2004). Lupita’s dress: Care in time. Hypatia, 19, 23–48.

14.

Davis

M. H.

(1980). A multidimensional approach to individual differences in empathy. JSAS Catalog of Selected Documents in Psychology, 10, 10–15.

15.

Davis

M. H.

(1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology, 44, 113–126.

16.

De Kemp

R. A. T.

Overbeek

de Wied

Engels

R. C. M. E.

Scholte

R. H. J.

(2007). Early adolescent empathy, parental support, and antisocial behavior. The Journal of Genetic Psychology, 168, 5–18.

17.

Dykeman

Nelson

R. J.

Appleton

(1995). Building strong working alliances with American Indian families. Social Work in Education, 17, 148–158.

18.

Dymond

R. F.

(1949). A scale for the measurement of empathic ability. Journal of Consulting Psychology, 13, 127–133.

19.

Engster

(2005). Rethinking care theory: The practice of caring and the obligation to care. Hypatia, 20, 50–74.

20.

Freedberg

(2007). Re-examining empathy: A relational-feminist point of view. Social Work, 52, 251–259.

21.

Gerdes

K. E.

Segal

E. A.

(2009). A social work model of empathy. Advances in Social Work, 10, 114–127.

22.

Gorsuch

R. L.

(1983). Factor analysis. (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

23.

Hall

J. A.

Mast

M. S.

(2007). Sources of accuracy in the empathic accuracy paradigm. Emotion, 7, 438–446.

24.

Hashimoto

Shiomi

(2002). The structure of empathy in Japanese adolescents: Construction and examination of an empathy scale. Social Behavior and Personality, 30, 593–602.

25.

Hepworth

D. H.

Larsen

J. A.

(1982). Direct Social Work Practice. Belmont, CA: Wadsworth.

26.

Hogan

(1969). Development of an empathy scale. Journal of Consulting and Clinical Psychology, 3, 307–313.

27.

Hojat

(2007). Empathy in patient care. New York, NY: Springer.

28.

Hollis

Woods

M. E.

(1981). Case work: A psychosocial therapy. New York, NY: Random House.

29.

Holosko

M. J.

(2006). Primer for critiquing social research. Belmont, CA: Thomson Cole/Brooks.

30.

Houston

D. A.

(1990). Empathy and the self: Cognitive and emotional influences on the evaluation of negative affect in others. Journal of Personality and Social Psychology, 59, 859–868.

31.

Infante

D. A.

Wigley

C. J.

(1986). Verbal aggressiveness: An interpersonal model and measure. Communication Monographs, 53, 61–69.

32.

Johnson

J. A.

Cheek

J. M.

Smither

(1983). The structure of empathy. Journal of Personality and Social Psychology, 45, 1299–1312.

33.

Jolliffe

Farrington

D. P.

(2005). Development and validation of the basic empathy scale. Journal of Adolescence, 29, 589–611.

34.

King

Jr. (2009). The development and initial validation of the empathy scale for social workers (Unpublished doctoral dissertation). University of Georgia, Athens, GA

35.

Kline

(1994). An easy guide to factor analysis. London: Routledge.

36.

Koss-Chioino

J. D.

(2006). Spiritual transformation, ritual healing, and altruism. Zygon, 41, 877–892.

37.

Kradin

(2005). The roots of empathy and aggression in analysis. Journal of Analytical Psychology, 50, 431–449.

38.

Lambert

M. J.

Barley

D. E.

(2001). Research summary on the therapeutic relationship and psychotherapy outcome. Psychotherapy, 38, 357–361.

39.

Lantz

(2001). Existential theory. In Lehmann

Coady

(Eds.), Theoretical perspectives for direct social work practice: A generalist-eclectic approach (pp. 240–254). New York: Springer.

40.

Lee-Hsieh

Kuo

C. L.

Tseng

Tuton

(2004). Development of an instrument to measure caring behaviors in nursing students in Taiwan. International Journal of Nursing Studies, 42, 579–586.

41.

Lejuez

C. W.

Hopko

D. R.

Levine

Gholkar

Collins

L. M.

(2006). The therapeutic alliance in behavior therapy. Psychotherapy: Theory, Research, Practice, Training, 43, 456–468.

42.

Martin

J. L.

(2001). Self psychology theory. In Lehmann

Coady

(Eds.), Theoretical perspectives for direct social work practice: A generalist-eclectic approach (pp. 145–164). New York: Springer.

43.

Marziali

Alexander

(1991). The power of the therapeutic relationship. American Journal of Orthopsychiatry, 61, 383–391.

44.

Mehrabian

Epstein

(1972). A measure of emotional empathy. Journal of Personality, 40, 525–543.

45.

Minuchin

Fishman

H. C.

(1981). Family therapy techniques. Cambridge, MA: Harvard University Press.

46.

Nunnally

J. C.

(1967). Psychometric theory. New York: McGraw-Hill.

47.

Olio

K. A.

Cornell

W. F.

(1993). The therapeutic relationship as the foundation for treatment with adult survivors of sexual abuse. Psychotherapy, 30, 512–522.

48.

Oswald

P. A.

(1996). The effects of cognitive and affective perspective taking on empathic concern and altruistic helping. The Journal of Social Psychology, 136, 613–623.

49.

Pins

A. M.

(1963). Who chooses social work, when and why? Council on Social Work Education, 63, 13.

50.

Pulos

Elison

Lennon

(2004). The hierarchical structure of the interpersonal reactivity index. Social Behavior and Personality, 32, 355–360.

51.

Rogers

C. R.

(1951). Client-centered therapy. Boston, MA: Houghton Mifflin.

52.

Rothery

Tutty

(2001). Client-centered theory. In Lehmann

Coady

(Eds.), Theoretical perspectives for direct social work practice: A generalist-eclectic approach (pp. 223–239). New York, NY: Springer.

53.

Rubin

Babbie

(2005). Research methods for social work. Belmont, CA: Thomson Brooks/Cole.

54.

Rushton

J. P.

Chrisjohn

R. D.

Fekken

G. C.

(1981). The altruistic personality and the self-report altruism scale. Personality and Individual Differences, 50, 1192–1196.

55.

Saulnier

C. F.

(1996). Feminist theories and social work. New York, NY: Haworth Press.

56.

Saulnier

C. F.

(2001). Feminist theories. In Lehmann

Coady

(Eds.), Theoretical perspectives for direct social work practice: A generalist-eclectic approach (pp. 255–278). New York, NY: Springer.

57.

Skovholt

(2005). The cycle of caring: A model of expertise in the helping professions. Journal of Mental Health Counseling, 27, 82–93.

58.

Smith

S. A.

Thomas

S. A.

Jackson

A. C.

(2004). An exploration of the therapeutic relationship and counseling outcomes in a problem gambling counseling service. Journal of Social Work Practice, 18, 99–112.

59.

Snodgrass

S. E.

Hecht

M. A.

Ploutz-Snyder

(1998). Interpersonal sensitivity: Expressivity or perceptivity? Journal of Personality and Social Psychology, 74, 238–249.

60.

Snodgrass

S. E.

Rosenthal

(1985). Interpersonal sensitivity and skills in decoding nonverbal channels: The value of face value. Basic and Applied Social Psychology, 6, 243–255.

61.

Spector

P. E.

(1992). Summated rating scale construction: An introduction. Sage University Paper series on quantitative applications in the social sciences, series no. 07-082. Newbury Park, CA: SAGE

62.

Stewart

R. P.

(1984). Building an alliance between the family and the institution. Social Work, 29, 386–390.

63.

Tevithick

(2003). Effective relationship-based practice: A theoretical exploration. Journal of Social Work Practice, 17, 163–176.

64.

Tudor

Worrall

(1994). Congruence reconsidered. British Journal of Guidance and Counseling, 22, 197–206.

65.

Underwood

Moore

(1982). Perspective-taking and altruism. Psychological Bulletin, 91, 143–173.

66.

Wickman

S. A.

Campbell

(2003). An analysis of how Carl Rogers enacted client-centered conversation with Gloria. Journal of Counseling and Development, 81, 178–181.

67.

Woods

M. E.

Hollis

(2000). Casework: A psychosocial therapy. Boston, MA: McGraw-Hill Press.