Introduction

Heterosexist discrimination continues to negatively affect lesbian, gay, and bisexual (LGB) individuals. In a comprehensive meta-analysis of LGB victimization studies between 1992 and 2009, Katz-Wise and Hyde (2012) found continued substantial rates of victimization among LGB individuals. Indeed, some areas of victimization increased over time (i.e., sexual assault from family, school victimization, and relational victimization) and no area decreased. Recognition of discrimination against lesbian, gay, bisexual, and transgender (LGBT) people, particularly LGBT youth (Wernick et al. 2013), and inclusive LGBT nondiscrimination policies are related to greater well-being for LGBT people (Riggle et al. 2010; Toomey et al. 2011). Although LGBT people are gaining ground in civil rights, they still experience exclusion or discrimination through social policy (Woodford et al. 2013a), and lapses in LGBT civil rights occur at national, state, and local levels. The field of social work in particular plays a significant role in addressing LGBT inequality, and addressing discrimination is an important part of social work preparation and practice (Chonody and Smith 2013). Social work graduates must be prepared to champion greater LGBT inclusion and policy protection as a routine part of practice (NASW 2008), and policy advocacy on behalf of LGBT people is still needed in a number of areas.

Important national progress was made in three recent US Supreme Court decisions that affirmed significant rights for LGBT people. This includes Obergefell v. Hodges (2015), in which the Court secured marriage equality for LGBT individuals, USA v. Windsor (2013), which made it illegal to deny married same-sex couples the same federal benefits afforded heterosexual couples, and Lawrence v. Texas (2003), which struck down sodomy laws and made same-sex activity legal across the country. This progress for marital equality is a major achievement with regard to the 1138 federal rights accorded legally married people in the USA (Human Rights Campaign 2015).

Despite these advancements, 39 % of Americans still oppose same-sex marriage (Pew Research Center, Religion and Public Life Project 2015). Further, national policies still do not adequately protect many LGBT civil rights. For example, the federal government has not yet passed an Employment Nondiscrimination Act (ENDA) outlawing work-related discrimination, even though such laws have been helpful in many states and municipalities (Klawitter 2011). In a sample composed of General Social Survey and Census data, Martell (2013) found that state ENDAs decreased the earning differential between behaviorally gay (single and cohabiting gay men) and heterosexual men by about 20 %. This reduction was explained by ENDA’s effect on discrimination overall; behaviorally gay men in states with ENDAs experienced less discrimination than in states without this policy protection. The authors recommended ENDAs as an effective way to combat discrimination against gay men. Green et al. (2011) analyzed workplace nondiscrimination policies relative to bisexual employees and found workplace policies more effective if their language also included gender identity and expression as well as sexual orientation. Riggle et al. (2010) found that state and local nondiscrimination policies improved LGB residents’ perception of a welcoming social environment, with fewer negative social messages, higher levels of sexual identity disclosure and support, and lower levels of internalized homophobia. In a 2011 national study, between 15 and 47 % of LGBT people reported experiencing some form of employment discrimination (Grant et al. 2011).

Similarly, though discrimination against same-sex couples in metropolitan online housing markets has been documented (Friedman et al. 2013), the federal government does not yet prohibit housing discrimination based on sexual orientation (US Department of Housing and Urban Development (HUD) nd). Advocacy for more inclusive LGBT policy is needed in the education arena as well. For example, while policy recommendations have been made with regard to sexuality education and support (i.e., reducing reliance on abstinence-only curriculum, including sexual orientation education), most schools have not yet implemented them (Rienzo et al. 2009). Since the “Don’t Ask, Don’t Tell” policy in the military was only repealed in 2010 (Don’t Ask, Don’t Tell Repeal Act of 2010, Pub. L. No 111-321, 124 Stat.3515, 2010), a potential increase in reactionary discrimination is possible, as well as continued stigma, prejudice, and need for care for almost one million LGBT veterans and service members (Ramirez et al. 2013).

Finally, 27 religious colleges and universities in 17 states have recently been granted waivers from Title IX that permit them to ban LGBT students from enrollment (Birkey 2015). At least six of these are accredited social work schools. Public schools continue to be sites of discriminatory policies and practices for LGBT youth (GLSEN 2013). All these areas in which LGBT policy is lagging point to the need for heightened policy advocacy and enforcement by social workers.

LGBT Discrimination, Policies, and Social Work

Recognition of the discrimination that LGBT people face, including prejudice against LGBT youth in educational settings (Wernick et al. 2013), is a critical component of professional fields such as law, psychology, and social work. For example, Courson (2012) noted implicit bias as a possible reason why LGBT people appear to be underrepresented among licensed attorneys. A study in Greece found psychology students scored higher on positive attitudes toward lesbians and gay men than did social work, medical, and nursing students (Papadaki et al. 2015). The American Psychological Association (APA) issued three anti-discrimination policies in 2004 affirming support for LGBT marriage rights (APA 2004a), adoption, child custody/visitation, foster care, and reproductive health care rights (APA 2004b), and equal opportunity for military service (APA 2004c).

Lapses in policy to protect and include LGBT people affect social work practice and education directly. For example, discriminatory social policies against LGBT people in housing, employment, education, and the military influence social work education and practice (Anastas 2013; Blackwell et al. 2004; Chonody et al. 2012). Further, even within higher education, some social work policies exclude and discriminate against LGBT people. One example of this is that there continues to be accredited Christian schools of social work with policies that exclude sexual orientation in their non-discrimination policies or that explicitly state that same-sex behavior is a breach of standard of conduct for employees (Abilene Christian Abilene Christian University 2011; Ross 2011). In one school, engaging in same sex sexual activity leads to disciplinary action (Abilene Christian University 2011). These policies violate the social work code of ethics (Dessel and Bolen 2014; Reamer 2014). A national study of social work program directors found that 134 programs (90 %) had sexual orientation non-discrimination policies, 12 programs (8 %) were reported as not having these policies, and three directors (2 %) did not know if they had policies (Martin et al. 2009).

The National Association of Social Workers practice standards, which use the social work code of ethics as a guide, indicate that social workers should practice without discrimination based on sexual orientation (NASW 2008). The Council on Social Work Education’s Council on Sexual Orientation and Gender Expression (CSOGE) (2008) has also set standards for practice with LGBT populations (Fredriksen-Goldsen and LaSala 2008; Messinger 2013). Yet, social workers’ cultural competence and practice skills with LGB people are still lacking in terms of the education that they receive (Martin et al. 2009; Messinger 2013). Full equality and civil rights for LGB people, particular in religious Christian schools of social work and practice settings, have yet to be achieved (Dessel and Bolen 2014). While NASW (2004) supports marriage equality, one study of 300 social work faculty found that 12 % opposed same-sex marriage, and an additional 10 % were “somewhat supportive” (Woodford et al. 2013b). In this study, faculty who had higher religiosity, as well as faculty with certain social attitudes about gender oppression, indicated less support for same-sex marriage.

Social work practice areas, such as the military, adoption, mental and physical health care, and other social services, intersect with policies relevant to LGBT people’s civil rights. Social workers who serve military personnel and veterans need to be equipped to manage the recently repealed Don’t Ask Don’t Tell policy and its implications for LGBT people (Johnson et al. 2013). Social service policies, such as those on adoption, discriminate against same-gender couples and limit their civil rights (ACLU 2014; Montero 2014). While LGB individuals can now adopt in most states, only 19 states and Washington, D.C., permit same-gender couples’ unrestricted adoption. Social workers continue to play a significant role in adoption policy and practice (Montero 2014).

Social policies also influence LGB people’s mental health, access to services, and freedom of expression (Wright et al. 2013). Legal bills and policies curtailing reparative therapy, for example, can protect LGBT youth from such harmful interventions (Anastas 2013). However, “conscience clauses” which may permit social workers to refer out LGB clients (Anastas 2013; Kaplan 2014) are complex. While it is not in clients’ best interest to receive services from social workers who are incompetent to provide them, who use religious liberties to justify denial of service, or who hold bias against LGBT clients, ideally, all social workers should be knowledgeable in working with LGBT people (Anastas 2013; Kaplan 2014).

Public school educational policies that include sexual orientation in their non-discrimination protected groups are critical to support and affirm LGBT youth and prevent bullying and harassment (MacGillivray 2003; Szalacha 2003). School social workers play a key role in advocating for anti-bullying policies (Kopels and Paceley 2012). Other areas of civil rights policies are relevant to social work practice. For example, supportive health policies are needed to address health disparities and provide the competent health care needed by LGB people, such as family medical leave and medical visitation (Anastas 2013; Conlon and Aldredge 2013). Social workers also interact with LGBT seniors, who may have experienced discrimination and stigmatization and who have specific social support needs (Orel 2004; Sullivan 2014).

In all of these social work areas, understanding the implications and effects of policy and discrimination on the lives of LGBT people is critical (Chonody et al. 2012). In order to better understand what impacts social worker attitudes about LGBT discrimination and policy, this paper examines the influence of demographics such as gender and race, religious attendance, political views, and beliefs about sexual orientation and gender identity choice, as well as variables of active thinking and cognitive empathy, on MSW student attitudes about LGBT social policies related to housing, civil rights, employment, education, the military and recognition of discrimination. Further, we examine whether these student attitudes and recognition of discrimination changed as a result of participation in a cultural competence course. This study extends earlier work by exploring the effects of a multicultural educational intervention on these attitudes. Further, this research examined cognitive processes such as active thinking and cognitive empathy that have been found to be critical components of undergraduate social justice education courses (Gurin et al. 2013) and have received limited attention for graduate social work populations (Gerdes et al. 2011a, b).

We first review literature on demographic and attitudinal variables that predict support for LGBT people and policies, as well as literature on skills of empathy and critical thinking. We also review previous research on cultural competence courses related to attitudes about LGBT policy. Transgender people were sometimes not included in the literature cited here. We thus refer to LGBT or LGB as the original authors used these terms. We continue with methods and results and conclude with discussion, limitations, and recommendations.

Literature Review

There are a number of variables, including demographics, religious attendance and political ideology, knowledge about LGBT people, critical thinking, and empathy, that may influence views about LGBT people. Demographic influences may be explained by theories of traditional gender norms that reinforce traditional constructions of masculinity and dominance (Herek and McLemore 2013). These theories state that transgressing traditional gender norms can lead to bias and exclusion (Fineran 2002). In communities of color, this holds true for theories of black masculinity as well, related to the historical oppression of African American men and the sexual oppression of slavery (Harris 2009; Lemelle and Battle 2004). Religious and political influences also may be explained by theories of religious fundamentalism and authoritarianism (Schulte and Battle 2004). Religious fundamentalism in US Christianity refers to following the literal Biblical text (Hood et al. 2005), and authoritarianism has been defined as unquestioning following of rules and authority (Stenner 2005). Socialization theory refers to the process by which we are socialized into accepting systems of oppression that exclude certain social groups as normal (Adams et al. 2007). Conversely, education and new knowledge can counteract the effects of negative socialization (Adams et al. 2007; Bassett and Day 2003). These attitudes and belief systems may then relate to the support of social policies regarding LGBT people (Brewer 2003; Pearte et al. 2013).

Gender

A number of studies found gender significant in predicting views about LGBT people, with men more homophobic and less supportive of sexual minority people than women (Brumbaugh et al. 2008; Herek and McLemore 2013; Swank and Raiz 2010a). One study of social work faculty found that women reported more support for sexual orientation and gender content in social work curriculum than did men (Fredriksen-Goldsen et al. 2011), and another study found that women more likely than men to sign a petition in support of LGBT rights (Swank et al. 2013).

Race

Studies examining the effects of racial identity on attitudes about LGBT people and inclusive policies have found varied results. Some national studies found African Americans less supportive than Whites of marriage equality (Brumbaugh et al. 2008; Pew Research Center, Religion and Public Life Project 2015). Research on student populations found that African American students hold less supportive attitudes toward LGBT people and have significantly higher homophobia and transphobia than White students (Logie et al. 2007; Swank and Raiz 2007; Swank et al. 2008; Woodford et al. 2013a). Studies have also found that social work faculty of color hold significantly less accepting attitudes toward LGBT people than White faculty (Einbinder et al. 2012; Woodford et al. 2013b).

Other research of college and social work student populations found no differences by race on attitudes toward LGBT people or on intention to vote in support of employment protection for LGBT people (Crisp 2006; Jayakumar 2009; Swank and Raiz 2010b; Swank et al. 2013). Lewis (2003) found that African Americans held more homophobic views than Whites but were more inclined to support gay civil liberties.

Religious Attendance

Religion is a well-researched predictor of attitudes about LGBT people, and religious fundamentalism and authoritarianism have been posed as theories that explain sexual prejudice (Herek and McLemore 2013). Many studies measured religious affiliation or religiosity. Several studies found that religious or conservative Christians, including religious Christian social work faculty, hold more negative attitudes toward LGBT people (Dessel et al. 2012; Logie et al. 2007; Pearte et al. 2013; Smith-Osborne and Rosenwald 2009; Swank and Raiz 2010a, b; Walls 2010; Woodford et al. 2013a, b).

Fewer studies have examined the unique effects of religious attendance as a predictor of attitudes toward LGBT people. Studies of college students found that attending religious services frequently predicted increased sexual prejudice and less support for the LGBT civil rights of marriage equality and employment protection (Jayakumar 2009; Woodford et al. 2013a). Other studies similarly found that frequent religious attendance predicted less support for same-sex relationships and marriage rights (Barth et al. 2009; Haider-Markel and Joslyn 2008; Swank and Raiz 2010b). Swank and Fahs (2014) found that religious attendance impeded social worker activism for gay rights.

Political Ideology

Political conservatism has been found to have a strong relationship to lack of support for LGB people and marriage equality (Brown and Henriquez 2008; Brumbaugh et al. 2008; Lewis and Gossett 2008; Pearte et al. 2013; Smith-Osborne and Rosenwald 2009; Woodford et al. 2012a). In one study, politically liberal college students had higher LGBT civil rights scores measuring marriage equality and employment protections (Woodford et al. 2013a). Swank et al. (2013) found that among college students, a liberal political identity predicted greater support for signing an employment protection petition for LGBT people. Political ideology and party affiliation have significantly predicted social work student attitudes toward lesbians and gay men, with Republican or conservative beliefs associated with more homophobia and negative attitudes toward sexual minority people in the military (Cluse-Tolar et al. 2004; Snively et al. 2004; Wallenberg et al. 2011). Another study of social workers found that conservative political ideology predicted less support for gay rights legislation (Smith-Osborne and Rosenwald 2009).

Beliefs About Sexual Orientation and Gender Identity Etiology

Beliefs about whether sexual orientation or gender identities are a choice have been increasingly studied as predictors of attitudes about LGBT people. A number of national surveys found a significant link between acknowledging sexual orientation as biologically based and support for LGB rights (Haider-Markel and Joslyn 2008; Lewis 2009). Related, another large national survey found that politically conservative people are more likely than liberal people to endorse choice explanations for LGBT orientation (Suhay and Jayaratne 2012). A study of student and community samples indicated that those who believed that lesbian or gay people choose this sexual orientation and also choose to violate certain values by living out their lives as lesbian or gay were more likely to oppose gay rights policies. Further, the study found that stereotypes about gay and lesbian people predicted negative beliefs about same-sex marriage, adoption, and open military service policies (Reyna et al. 2014). Other studies on college student civil rights attitudes about marriage equality and employment protection, and anti-gay bias, found that a belief that biology is the causative factor in sexual orientation predicts more support for civil rights and less bias (Eldridge et al. 2006; Rutledge et al. 2012; Woodford et al. 2013a). Studies of social work students specifically have found that believing sexual orientation is a choice predicted less support for same-sex relationships and marriage equality (Swank and Raiz 2010a, b). Etiology beliefs about sexual orientation and gender identity are an important variable to measure (Chonody and Smith 2013).

Critical Thinking and Empathy

Critical, analytical thinking and empathy are two different and important skills for social workers to develop with regard to their work with vulnerable and marginalized groups such as LGBT populations (Fredriksen-Goldsen et al. 2014; Gerdes et al. 2011a). Critical thinking involves attending to multiple perspectives, self-reflection, and recognition of personal bias and analysis of social power and inequality. These are necessary components of social work practice in order to promote justice for oppressed groups (Miller et al. 2011). Empathy, which refers to the understanding of the feelings, thoughts, and perspectives of another person, has been defined and measured in many different ways, most often differentiating between affective or emotional empathy and perspective taking or cognitive empathy (Gerdes et al. 2011a, b; Gurin et al. 2013). Some studies have found empathy to be related to prosocial behaviors and alliances (Stephan and Finlay 1999), while other lines of research have not found this correlation (Fingerhut 2011). Empathy has been understudied in social work practice and needs further attention (Gerdes et al. 2011a).

Cultural Competence Courses and Attitudes About LGBT People

Cultural competence courses are one approach to educating students, improving attitudes, and promoting political advocacy skills with regard to LGBT people (Swank and Fahs 2014; Van den Bergh and Crisp 2004). A number of different social work programs have used courses with LGBT content in this way. Results have indicated a reduction in homophobia and anti-gay attitudes, particularly for students with moderately (as compared to lower or higher) positive pre-test attitudes (Ben-Ari 1998; Bassett and Day 2003; Swank et al. 2008). Woodford et al. (2013a) found that having taken courses with LGBT content had a small but significant association with supporting LGBT civil rights such as marriage equality and employment rights. Related, cultural competence trainings have also found significant increases for participants in attitudes and skills to promote competence policies for LGBT people (Leyva et al. 2014). Overall, few studies have looked at the outcomes of cultural competence courses with regard to fostering support for policies that affect the rights of LGBT people.

IGD and Cultural Competence Education

This study examined a cultural competence course that included intergroup dialogue (IGD) as a pedagogical method. IGD is a face-to-face group work practice involving intentionally diverse participants from at least two social identity groups with a history of conflict (Dessel 2014). Co-facilitators represent the intentional diversity of group members. Dialogue occurs over stages that move from less to more challenging discussions of social identity experience and conflict and ends with both individual and collaborative social justice action planning (Dessel and Rodenborg in press). Participants examine their social identity beliefs, communications, actions, and conflicts through experiential activities and directed reading, writing, and reflection. IGD has been shown to be an effective social justice education method across many areas of higher education (Gurin et al. 2013), including social work (Dessel 2014), and specifically with regard to gender and sexual orientation social identity issues (Dessel et al. 2013b) (see Dessel and Rodenborg (in press) for a complete description of the IGD format used in this study).

The current study contributes to the literature by examining how social work students’ gender, race, religious attendance, political ideology, beliefs about sexual orientation and gender identity choice, and critical thinking and cognitive empathy affect attitudes about LGBT social policies. In particular, political ideology, critical thinking, and empathy have been less studied in the social work population and thus are important variables to examine in understanding LGBT bias (Chonody and Smith 2013; Chonody et al. 2014; Epstein 2011; Galambos 2009). The study also examines attitudes about LGBT military policy, which is understudied in social work, and looks at the effects of course participation on attitudes and recognition of discrimination. We asked the following research questions:

  1. RQ no. 1:

    Does race, gender, religious attendance, political ideology, belief about sexual orientation and gender identity choice, critical thinking, and cognitive empathy predict social work student attitudes about LGBT military policies, marital laws, and awareness of LGBT discrimination?

  2. RQ no. 2:

    Does taking a cultural competence course influence social work student attitudes about LGBT military policies, LGBT marital laws, and awareness of LGBT discrimination?

Methods

Sample

This sample included three cohorts of 149 MSW students in a Midwest department of social work from 2008 to 2011 who were enrolled in a two-semester cultural competence course that included seven intergroup dialogue (IGD) sessions. The course assessment examined learning outcomes for students in cultural competence and IGD educational experiences. A total of 149 students enrolled in second-year MSW classes between 2008 and 2011 and were invited to participate in the study. After deleting students who did not complete both the pre- and post-course survey, the sample consisted of 134 students, which is a 90 % response rate. This study was IRB approved.

Course Description

In brief, the course consisted of a traditional fall semester diversity course in which macro students studied inclusive organizational practice, affirmative action, and other macro-level content, while clinical students focused on advanced cultural competence skills in direct practice (see Dessel and Rodenborg (in press) for a complete description of both the cultural competence course and its IGD pedagogy). The second semester included 14 h of facilitated small-group IGD led by intentionally diverse two-person teams of trained facilitators. Between five and seven groups of seven to ten students participated each year, with group assignments made to maximize social identity diversity. IGD followed a four-stage group work model adapted from the work of the University of Michigan Program on Intergroup Relations (Gurin et al. 2013) that included ground rules, sharing of experiences related to social identity, discussion of “hot-button” issues, and planning for social justice action.

Important topics introduced in the fall semester were also reinforced during dialogue in the second semester, such as implicit bias, microaggressions, aversive racism, segregation, and other theories and concepts. Required reading was also assigned, as was a brief individual written reflection after each dialogue session. Finally, students wrote a summary paper reflecting on their experience throughout the two-semester course.

Measures

Dependent Variables

We used three dependent variables in this analysis that were derived from a larger survey of cultural competence learning in an MSW course (Dessel and Rodenborg in press).

  • Attitudes about LGBT military policy. This was measured with one item at pre-test and post-test that stated, “The military policy of ‘don’t ask’ is a sound policy.” Responses were 1 = strongly disagree to 7 = strongly agree. Higher scores indicated less affirming attitudes.

  • Attitudes about LGBT marriage equality. This was measured with one item at pre-test and post-test that stated, “Same sex couples should not have the right to legal marital status.” Responses were 1 = strongly disagree to 7 = strongly agree. Higher scores indicated less affirming attitudes.

  • Recognition of LGBT discrimination. This scale consisted of two items measured at pre-test and post-test that stated, “People who identify as GLBTQ face discrimination in the USA in areas such as housing, civil rights, and employment,” and “Prejudice and discrimination in the educational systems limit success of GLBTQ youth.” The Cronbach’s alpha for this scale was 0.80. Responses were 1 = strongly disagree to 7 = strongly agree. Higher scores indicated more affirming attitudes.

Independent Variables

Sociodemographics

Gender was measured with three choices, male, female, or transgender or intersex and was recoded (0 = female, 1 = male), as there were no respondents who identified as transgender or intersex. Race was measured by asking, “What is your racial/ethnic identification” with eight choices, including “other.” Due to the small number of students of color, race was recoded into 0 = “White” and 1 = “person of color.” Frequency of religious attendance was measured with six categorical choices. We originally entered this as a continuous variable and found that it had no significance. Based on previous research (Wright 2014), we then dummy coded this variable using categories of 0 = 2–4× a year, 1× year, or never and 1 = multiple times a week, once a week, or at least once a month. A higher number indicated more frequent attendance.

Covariates

Political Ideology

Political ideology was measured with an item that asked, “Concerning your political views, where would you place yourself on this scale that ranges from extremely liberal to extremely conservative? (Select only one).” Responses ranged from 1 = extremely liberal to 7 = extremely conservative, so that higher numbers indicated more a more conservative political ideology.

Beliefs About Choice

Belief about sexual orientation and gender identity choice was measured with an item that asked how much students agreed or disagreed with the following statement: “Being gay, lesbian, or transgender is a choice people make.” The continuous scale responses ranged from 1 = strongly agree to 7 = strongly disagree (higher number indicated believing it is not a choice).

Active Thinking (Gurin et al. 2013)

The eight-item Active Thinking scale (five reverse-coded items), previously used in a large national study of intergroup dialogue courses, measures the capacity for critical and analytical thinking (Gurin et al. 2013) and asks “statements concerning thinking about people, society, and the world, and how much these statements describe you.” Responses ranged from 1 (not at all like me) to 7 (very much like me). Higher scores indicated more active thinking. Reliability for this scale was good, with pre Cronbach’s alpha = 0.81. Item examples included “I really enjoy analyzing the reasons or causes for people’s behavior” and “I am fascinated by the complexity of the social institutions that affect people’s lives.”

Cognitive Empathy (Davis 1983)

This scale of five items (two reverse-coded items), also previously used in a large national study of intergroup dialogue courses, captures the capacity to express an understanding of the thoughts, experiences, or perspectives of another person or group (Gurin et al. 2013). This particular type of empathy involves the ability to consider multiple perspectives (Davis 1983). The Cronbach’s alpha was fair at 0.64. Responses ranged from 1 (not at all like me) to 7 (very much like me). Higher score indicated more cognitive empathy. Item examples included “I strive to see issues from many points of view” and “I believe that there are many sides to every issue and try to look at most of them.”

Data Analysis and Results

Descriptive statistics were conducted for all study variables using SPSS version 22. Data were examined, and no problems were found with normality or with multicollinearity as indicated by the variance inflation scores (range 1.063–1.290). We used correlations, linear regression, and a paired sample t test to answer our research questions. Descriptive statistics are reported in Table 1. Demographics indicated that the sample was 91.8 % female and 80.5 % White. The majority of respondents were raised Christian (91.2 %), thus indicating that religious attendance may have referred to church attendance or other types of church functions. The mean for political ideology was 2.63 (range of 1–7), with higher score indicating more conservative political ideology; thus, this was a fairly liberal sample. The mean for beliefs about sexual orientation and gender identity choice was 5.78 (range of 1–7) with higher scores indicating believing that sexual orientation and gender identity are not a choice. This sample largely but not completely held this belief. The mean for active thinking was 5.62 (pre-test) and for cognitive empathy was 5.58 (pre-test), range of 1–7, with higher scores indicating a higher level of active critical thinking and cognitive empathy skills.

Table 1 Descriptive statistics for sample demographics and independent variables
Full size table

Bivariate and Multivariate Statistics

Tables 1 and 2 show descriptive statistics and the correlations among the variables of interest. We conducted linear regressions to examine our first question of what variables predicted student attitudes about LGBT policy and discrimination. The first model explained 30 % of the variance in the dependent variable’s score of pre-test military policy attitudes (Table 3). Regression results indicated that only gender and political views significantly predicted the pre-test score for attitudes about military policy. Men and more conservative students had less affirming attitudes than women and less conservative students. The second model explained 22 % of the variance in the dependent variable score of pre-test attitudes about LGBT marital policy (Table 4). Results indicated that gender, political views, active thinking, and cognitive empathy significantly predicted the pre-test score for attitudes about marital policy. Men, students with more conservative political ideology, and students who reported less empathy and who reported more active thinking held less affirming attitudes about marital policy. Finally, the third model explained 22 % of the variance in the dependent variable’s score of recognition of LGBT discrimination (Table 5). In this model, only beliefs about sexual orientation and gender identity etiology significantly predicted students’ recognition of LGBT discrimination. Those students who believed that it is not a choice were significantly more likely to report recognition of discrimination.

Table 2 Pearson correlations among all continuous and dichotomous study variables
Full size table
Table 3 Multivariable linear regression on the effects of gender, race, religious attendance, political views, beliefs about choice, active thinking, and cognitive empathy on pre-test attitudes about LGBT military policy
Full size table
Table 4 Multivariable linear regression on the effects of gender, race, religious attendance, political views, beliefs about choice, active thinking, and cognitive empathy on pre-test attitudes about LGBT marital policy
Full size table
Table 5 Multivariable linear regression on the effects of gender, race, religious attendance, political views, beliefs about choice, active thinking, and cognitive empathy on pre-test recognition of LGBT discrimination
Full size table

In order to answer our second research question of whether taking a cultural competence and IGD course influenced attitudes about LGBT military policies, LGBT marital equality, and recognition of LGBT discrimination, we used a paired t test (Table 6). Results indicated a significant positive change pre-test to post-test on the measure of recognition of LGBT discrimination and no significant change on the measures of attitudes about military policy or marital equality.

Table 6 Pre- and post-test mean difference for attitudes about LGBT policies and discrimination (n = 130)
Full size table

Discussion and Implications for Social Policy

This study built on previous research suggesting that religion, political ideology, and belief about whether sexual orientation and gender identity are a choice are associated with social work student attitudes about LGBT people and related policies. Additionally, we sought to explore the influence of student’s critical thinking skills and empathy on attitudes about LGBT discrimination and policies, given that these are important areas for social work education (Fredriksen-Goldsen et al. 2014; Gerdes et al. 2011a). We also examined the influence of a cultural competence course on student attitudes about LGBT policies.

With regard to research question 1, we found that gender and political views predicted post-test attitudes about military policy, with men and more conservative students holding less affirming attitudes than women and less conservative students. Very few studies of social work populations have examined attitudes about LGBT military populations (Anastas 2013; Blackwell et al. 2004; Wooten 2015). Thus, this analysis contributes to the body of literature. Our findings indicated that race, religious attendance, beliefs about choice, and active thinking and empathy did not predict student attitudes about LGBT military policies. This finding may have been influenced by the characteristics of our particular sample, which was primarily Christian (91.2 %) and leaned liberal (mean = 2.62 in a 1–7 scale, with 1 being most liberal). About half attended a religious institution frequently (47.7 %), which, in this case, was likely a Christian church.

Although nearly half the sample (48 %) reported frequent church attendance, the overall mean for this group of frequent church attenders (6.12) was fairly affirming in their pre-test attitudes about LGBT policies and discrimination. Social workers are expected to be policy advocates for disadvantaged groups (NASW 2008), and this sample may have brought this ideal with them as they entered their educational experience. Social work educators should build upon this by continuing to include LGBT policy advocacy in the curriculum (Chonody et al. 2012; NASW 2010). It is also possible that these students may attend churches that affirm LGBT people (Levy 2014). Given the role of churches in social justice work (Todd and Rufa 2013), this information, as well as other role models of Christian LGBT affirmation (Brice 2014; Dessel and Bolen 2014; Drumm et al. 2014), can be highlighted in courses on social policy and social change. Finally, syncretism, or the degree to which individuals’ beliefs match their church doctrine (Woodford et al. 2012), may have played a role in the non-significant effect of religious attendance. Despite the potentially negative teachings of some churches (Levy 2014) that these students attend, the students may hold their own, different views that are more affirming (Woodford et al. 2012b). Curriculum that includes attention to critical consciousness around policies that impact LGBT oppression and civil rights will assist students in practicing their social work values (Saltzburg 2008).

Results indicated that gender and political views also predicted attitudes about marital equality, such that men and students who held a more conservative political ideology were less likely to endorse LGBT marital equality. This influence of gender is line with previous research (Swank and Raiz 2010a), as is the influence of political ideology (Smith-Osborne and Rosenwald 2009; Swank et al. 2013) and points to the importance of attending to gender dynamics in educational settings and helping social workers critically examine their belief systems with regard to political issues and LGBT civil rights.

Further, active thinking and empathy were also predictive of attitudes about marital equality. Students with less empathy had less affirming attitudes. Interestingly, students who reported higher active critical thinking also indicated less affirming attitudes. The active thinking finding is not what would be expected. However, it is possible that students’ ratings of their critical thinking skills were over inflated at the start of the course, given that previous research found that students’ scores on this scale significantly decreased after taking this course (Dessel and Rodenborg in press).

Given that politically conservative students may feel silenced in the classroom (Flaherty et al. 2013) and are less likely to support LGBT equality (Smith-Osborne and Rosenwald 2009), teaching approaches are needed that help students critically analyze social policy with regard to LGBT rights (Dessel and Bolen 2014; Galambos 2009; Walls and Seelman 2014). Rosenwald and colleagues (2012) discuss the complicated process of engaging students across the political spectrum in hot topic issues such as abortion or health care reform and offer numerous approaches for faculty to engage students. This engagement in critical thinking about LGBT rights is necessary, as social workers may hold a wide diversity of political opinions (Rosenwald 2006; Bolen and Dessel 2013) and politics affects social policy (Epstein 2011; Swank et al. 2013). Teachers must be intentional in establishing inclusive classroom that include all students regardless of political views.

Lewis (2009) questions whether attitudes about LGBT policies are shaped by beliefs about choice. Instead, people’s beliefs about choice may be shaped by political ideology and religious values, which may then influence beliefs about policy (Lewis 2009). Suhay and Jayaratne (2012) confirm the complexity of these relationships. Reyna and colleagues (2014) concur and found that beliefs that lesbian and gay people choose to violate certain perceived normative societal values were predictors of bias accounted for the relationship between attribution of choice and attitudes about LGBT policies. In the Reyna et al. (2014) study, stereotypes about LGB people were strongly associated with attributions of lesbian and gay people violating values. This indicates that in order to promote support for equitable policies, stereotypes and value violations need to be more carefully examined and explored in social work educational settings in order to highlight the common values held by people of all sexual orientations (Reyna et al. 2014).

Results for recognition of LGBT discrimination indicated that only beliefs about etiology of sexual orientation and gender identity were predictive. Most previous research has measured this predictor with regard to attitudes about LGBT people and policies (Haider-Markel and Joslyn 2008; Lewis 2009; Swank and Raiz 2010a, b). Our results indicate that etiology beliefs also predict recognition of discrimination, and thus, the connection between recognition of discrimination and attitudes needs further study.

With regard to research question 2, while students did not change significantly pre-test to post-test on measures of attitudes about military policy or marital equality, they did change significantly in recognition of LGBT discrimination. The effect size of this change is 0.29, so this is a small effect size. However, the clinical significance of this is notable in that the recognition moved in the direction from “agree somewhat” to “agree strongly.” Given the evidence of discrimination faced by LGBT people, it is critical that all social workers move toward “agreeing strongly” that LGBT people face discrimination in housing, civil rights and employment and that prejudice and discrimination in the educational system limit success for LGBT youth.

This suggests that participation in a cultural competence course that included IGD helped students better understand the discrimination and exclusion faced by LGBT people in social policy arenas; their attitudes in favor of inclusive policy improved. This finding adds to growing interdisciplinary evidence on IGD pedagogy as a way to help students learn about social identity complexity and difference within a social justice framework, including evidence from higher education (e.g., Alimo et al. 2002; Clark 2005; Dessel 2010), political science (Walsh 2006), conflict resolution (e.g., Dessel and Rogge 2008), and communication study (e.g., DeTurk 2006), among other areas. Within social work, authors have documented IGD’s usefulness in both classroom and community contexts (e.g., Boulden 2007; Dessel et al. 2006; Glass 2012; Grodofsky and Soffer 2011; Lopez-Humphreys 2012; Lopez-Humphreys and Dawson 2014; Rodenborg and Huynh 2006; Rozas 2007; Tauriac et al. 2013). A handful of authors has explored IGD in the context of sexual orientation in particular. For example, in a multicultural psychology course, Miles et al. (2014) found that IGD contributed to increased multicultural competence in knowledge, skills, and awareness. Dessel et al. (2013a) and Dessel et al. (2013b) found that IGD helped students better understand the oppression faced by LGB people and to change their own stereotypic views related to LGD identity. These studies add to the body of social work evidence supporting the effectiveness and importance of IGD pedagogy.

Limitations and Future Research

There were a number of limitations to this study. First, the sample was relatively small. Second, this was a convenience sample and not representative; thus, there was a large selection bias, and the results cannot be generalized beyond the students in this course, are specific to social work education, and do not necessarily reflect other US college students or US citizens. The geographic location, mission, values, and institutional history of the college in which these students were enrolled influenced study outcomes in undetermined ways. Further, as this sample was largely White, women, and Christian, having a sample of students with different demographics might result in different findings, and people of color and men were notably underrepresented.

We did not collect data on current religious beliefs, and this variable would be important to examine along with the data about religious background and religious attendance. Also, we did not collect data on sexual orientation. This is a significant limitation, as support for LGBT rights is notably different for sexual minority and heterosexual populations (Chonody and Smith 2013), and thus, we do not know the effects of students’ sexual orientation identity on this study’s outcomes. Future research should include sexual orientation identity as a control variable. The item that measured beliefs about both sexual orientation and gender identity being a choice also could have been separated out into two items, as differences in views about these two identities have been found in previous research (Chonody and Smith 2013). Future research should measure these beliefs with two separate items.

No control or comparison group was used, so the direct effects of the course cannot be determined. It would be important to directly examine the extent to which IGD works as a method to move students toward greater comfort with LGBT policy advocacy. Having a comparison group of students who did not take the course would provide a more substantial analysis to measure the effects of the course. It is possible that maturation effects rather than the course content contributed to the significant changes found. Finally, a different dependent variable might have captured other aspects of attitudes about LGBT policies, including intention or actual action around support of LGBT policy. These are limitations that we hope will be addressed in further research on how IGD cultural competence education affects students’ attitudes about LGBT policies in social work and other disciplines.

These limitations point to areas for further research. Future studies should collect data on sexual orientation as well as current religiosity. Also, it is not clear what explains the unexpected finding of no significance with regard to religious attendance. To advance LGBT policy, we should focus on what churches and other religious institutions are teaching (Levy 2014; Woodford et al. 2012b) about policy and activism. While some research has explored church and other religious institutions’ teaching about policy and activism (Levy 2014; Woodford et al. 2012a), the effect of these teachings on parishioners’ attitudes toward LGBT affirming policy should be addressed.

Research is needed with larger samples. This sample was small and the majority of people in the sample held similar political attitudes (liberal-leaning) and attitudes about LGBT policy (positive-leaning). A larger sample with more political ideology and attitudinal difference may have different outcomes. It is difficult to determine actual beliefs among students since they are likely to report responses that they think are socially acceptable. Many students may have felt that it was unacceptable to hold attitudes that they worried might suggest prejudice against LGBT people. Qualitative research may shed more light on these topics. Finally, future research is needed to address the intersection of race and LGBT discrimination and attitudes toward LGBT affirming social policy, in addition to exploring other social identity intersections.

Conclusion

LGBT rights are universal liberty claims (Kollman and Waites 2009; Yoshino 2006). Social work students have a professional code of ethics that promotes social justice and change in all areas of practice and policy (NASW 2008). This means that LGBT rights should be supported by social workers. In order to support social workers in promoting LGBT rights with regard to social policies, educators need to help students analyze their political views and the complicated nature of human sexuality as well as the effect of race. However, reflection and analysis must lead to skill development in policy advocacy areas. Social work is an action profession, and social work students need more help in practicing their advocacy skills (Schneider and Lester 2001). IGD is a useful pedagogy designed to help students move toward the social action that is necessary for LGBT civil rights (Lopez-Humphreys and Dawson 2014).