PoliticalIdeologyandPerceptionsofSocialJusticeAdvocacyAmongMembersoftheAmericanCounselingAssociation.pdf

ORIGINAL ARTICLE

Political Ideology and Perceptions of Social Justice
Advocacy Among Members of the American Counseling
Association

Janeé M. Steele & Gary H. Bischof & Stephen E. Craig

Published online: 17 September 2014
# Springer Science+Business Media New York 2014

Abstract This study explored perceptions of social justice advocacy among liberal, moderate,
and conservative members (N=214) of the American Counseling Association (ACA). Results
showed that conservative participants had somewhat less favorable perceptions of social
justice advocacy, but generally did not differ statistically from liberal and moderate partici-
pants. Statistically significant differences, however, were found among extremely liberal
participants. All participants generally supported the use of ACA resources for social activism.
Implications and limitations are discussed.

Keywords Social Justice Advocacy. Political Ideology. American Counseling Association

Introduction

Organizations in countries around the world embrace social justice advocacy as an important
aspect of counselor identity. The mission of the International Association for Counselling
(IAC), for example, is “To serve as an international leader and catalyst for counsellors and
counselling associations by advancing culturally relevant counselling practice, research and
policy to promote well being, respect, social justice and peace worldwide” (www.iac-irtac.org/
node/36). Similarly, the Canadian Counselling and Psychotherapy Association’s (CCPA)
Standards of Practice require counselors to “convey respect for human dignity, principles of
equity and social justice, and speak out or take other appropriate actions against practices,

Int J Adv Counselling (2014) 36:450–467
DOI 10.1007/s10447-014-9217-0

J. M. Steele (*): G. H. Bischof
Department of Counselor Education and Counseling Psychology, Western Michigan University,
Kalamazoo, MI 49008-5226, USA
e-mail: [email protected]

G. H. Bischof
e-mail: [email protected]

S. E. Craig
Family and Consumer Sciences, Western Michigan University, Kalamazoo, MI 49008-5322, USA
e-mail: [email protected]

www.iac-irtac.org/node/36

www.iac-irtac.org/node/36

policies, laws, and regulations that directly or indirectly bring harm to others or violate their
human rights” (www.ccpa-accp.ca/en /standardsofpractice/). Social justice advocacy is
considered one of five primary themes in the internationalization movement currently taking
place across the counseling profession today (Ng & Noonan 2012), and is gradually becoming
recognized as one of counseling’s core competencies.

Despite notable support throughout counseling’s international community, social justice
advocacy has become an increasingly controversial topic in the United States, especially
among some members of the American Counseling Association (ACA). Specifically, concerns
have been raised over what is perceived to be: (a) the liberal political agenda of social justice
advocates, (b) the marginalization of conservative counselors, and (c) the inappropriate use of
ACA resources for social activism (Canfield 2007, 2008a, b; Hunsaker 2008; King 2010).
Concerns of this nature suggest that contrary to what is most often stated by social justice
proponents, many counselors may question if efforts to address social and political issues are
appropriate tasks for counselors in their professional roles. They further imply that opinions
about social justice advocacy among counselors may be shaped by differences in liberal and
conservative political ideologies.

While various opinions have been expressed about the nature and purpose of social justice
advocacy in counseling (cf., Harrist & Richardson 2012; Hunsaker 2011; Smith, Reynolds, &
Rovnak 2009), there is little empirical evidence to document perceptions of social justice
advocacy among counseling professionals. Claims that conservative counselors are likely to
object to social justice advocacy are also largely unsubstantiated within the literature. The
purpose of this study, therefore, was to examine the relationship between perceptions of social
justice advocacy in counseling and political ideology among members of the American
Counseling Association (ACA). Specifically, this study examined: (a) perceptions of the
attitudes, behaviors, knowledge, skills, importance, and personal practices associated with
social justice advocacy, as well as differences in those perceptions among liberal, moderate,
and conservative counselors; (b) support for the use of ACA resources to advocate for social
issues; and (c) the relationship between perceptions of social justice advocacy and social and
political characteristics, including highest degree obtained, gender, age, race, sexual orienta-
tion, income, political party affiliation, and political involvement.

Overview

Social justice advocacy, defined as “professional practice, research, or scholarship intended to
identify and intervene in social policies and practices that have a negative impact on the mental
health of clients who are marginalized on the basis of their social status” (Steele 2008, pp. 75–
76), has been infused into nearly all aspects of counseling, including counselor training, ethics,
and professional development in the United States and other parts of the world. The Council
for the Accreditation of Counseling and Related Educational Programs Standards (CACREP)
(2009), for example, states that all accredited counselor education programs must provide
training in social justice and advocacy processes (Section II Standard G.1.i.; Standard G.2.e.).
Similarly, the ACA Code of Ethics (2005) indicates that counselors have an ethical obligation
to advocate for clients at individual, group, institutional, or societal levels when appropriate
(A.6.a). Attitudes, knowledge, and skills necessary for advocacy at these levels are defined
through ACA endorsed competencies (Lewis, Arnold, House, & Toporek 2003). Numerous
books, articles, and special journal issues published within the last decade further define how
counselors can work within their professional roles to eliminate societal barriers that hinder
mental health and human development (e.g., Chang, Crethar, & Ratts 2010; Crethar &
Winterowd 2012; Goodman 2009; Holcomb-McCoy 2007; Lee 2007).

Int J Adv Counselling (2014) 36:450–467 451

www.ccpa-accp.ca/en

In spite of what appears to be widespread support for social justice advocacy in counseling,
recent literature highlights significant concerns regarding its role in the profession, especially
as it relates to its political aspects (cf., Harrist & Richardson 2012; Hunsaker 2011; Smith et al.
2009). Special opinion pieces and letters to the editor of Counseling Today, ACA’s monthly
newsletter, illustrate these concerns among practitioners, students, and even those within
counseling’s leadership ranks. For example, in articles written for Counseling Today, Brian
Canfield (2008a, b), a former ACA President, argued that the social justice movement in
counseling is based primarily on liberal values that are not reflective of a consensus within the
profession. According to Canfield, because counselors come from both liberal and conserva-
tive backgrounds, social justice activists should stop trying to utilize the resources of profes-
sional organizations like ACA in to advocate for specific social positions, as to use such
resources would be unfair to members who would advocate for opposing positions. Similarly,
Robert Hunsaker (2008), a practicing counselor, and students Fred Lockhard and Christopher
Stack (2008) also utilized Counseling Today to speak out against what they considered a
growing liberal and political agenda in counseling. Hunsaker (2008) stated:

…it is rather nonsensical to say that social justice is “highly political,” when, in
fact, it is entirely political. What else does one call activism on behalf of
minority issues at the group level? …social justice can be practiced only by
those on the political far-left. Consider for example how incongruent it would be
for Republican, objectivist, pastoral, independent or perhaps even moderate
Democrat counselors to advocate for gay marriage or a variety of other group-
level minority issues (p. 21)…it makes more sense when you consider that
social justice is largely the product of academics…several studies show an
extreme liberal bias in universities. (p. 43)

Whereas, Lockhard and Stack (2008) wrote:

It was our extreme displeasure to see a clearly political, anti-war, anti-service member
letter to the editor posted in the November 2007 issue of Counseling Today. We have our
own, differing political views and opinions, yet we would never even think of asking
ACA as a body to support our political agenda. Is that objective? Is that non-biased? If
so, perhaps the professor will join us in asking ACA to speak out against gun control! Is
this type of political rhetoric in the best interest of counseling as a profession? We do not
think so. (p. 4)

Research in the political science discipline has shown that differences in liberal and
conservative ideologies do account for preferences in many areas outside of formal politics.
In particular, liberal and conservative political ideologies have been found to account for
preferential differences in issues closely related to social justice, including social change,
equality, progress, and flexibility (Jost, Federico, & Napier 2009).

Political ideology in this article is defined as a broad system of ideas, beliefs, attitudes, and
values that give meaning to social concerns and guide ideas concerning how society should
operate (Adams 2001; Freeden 1996; Kerlinger 1984). Liberalism is defined as an ideology
characterized by concern for change, equality, and positive government action to improve
human welfare, while conservatism is defined as concern for stability, tradition, and limited
government involvement in domestic social interests (Dunn & Woodard 2003; Jost et al. 2009;
Kerlinger 1984).

As previously mentioned, research concerning the role of political ideology in counseling,
specifically as it relates to perceptions of social justice advocacy, is limited. In a study of
American School Counselor Association (ASCA) members, Parikh, Post, and Flowers (2011)

452 Int J Adv Counselling (2014) 36:450–467

found that school counselors with liberal political ideologies were more likely to engage in
social justice advocacy behaviors and have positive attitudes toward social justice advocacy
than other school counselors. Similarly, in a study of clinical, counseling, and school psychol-
ogy graduate students in APA-accredited programs, Linnemeyer (2009) found that participants
who identified their political ideology as far left had significantly higher social justice
advocacy attitudes and behaviors than those who identified as liberal, middle of the road, or
conservative.

Overall, while the above-mentioned studies support claims of a relationship be-
tween political ideology and social justice advocacy in counseling, they fail to provide
evidence of objections to social justice advocacy among conservative counselors, and
provide little insight into opinions about the use of resources obtained through
professional organizations like ACA for social activism. These studies also fail to
provide empirical support for assertions that social justice advocacy is largely the
product of academicians and is unlikely to be supported among counselors outside of
the realm of higher education (Hunsaker 2011). The current study adds to the
literature by examining differences in perceptions of social justice advocacy among
liberal, moderate, and conservative counselor members of ACA, while also examining
differences between members who work primarily in higher education as counselor
educators, and those who work primarily in clinical settings.

Method

Procedure

Participants were selected from a randomized sample of 999 individuals from the
Counselor (n=500) and Counselor Educator (n=499) membership categories of the
American Counseling Association (ACA). After obtaining university institutional re-
view board approval, participants were invited to an online administration of the
study’s survey by postcards sent via the United States Postal Service, followed by
two email invitations. Of the 999 email invitations, 42 were returned as undeliverable.
Two hundred and sixty three (263) initiated the survey, representing an overall
response rate of 27.48 %. Five of the respondents indicated that they did not consent
to participate in the study, and 44 were excluded from analysis due to missing data.
The final number of respondents who were included in the study, therefore, was 214.
The survey took approximately 20 min to complete, and participants were offered a
chance to win one of four $50 VISA gift cards as an incentive.

Participants

Participants ranged from 27 to 75 years of age (M=51.30, SD=11.31). The majority of
participants reported their race/ethnic group as Caucasian (78.04 %, n=167). African
American was the second most highly represented race/ethnic group (10.28 %, n=22),
followed by Hispanic/Latino (2.80 %, n=6); Other (2.34 %, n=5); Asian/Pacific Islander
(1.87 %, n=4); Multiracial (1.87 %, n=4); and Native American (.47 %, n=1). Approximately
67 % of the participants were Female (n=143), 31.31 % (n=67) Male, and .47 % (n=1)
Transgendered. In regard to sexual orientation, 89.72 % (n=192) identified as Straight, 4.21 %
(n=9) identified as Lesbian, 2.80 % (n=6) identified as Gay, 1.40 % (n=3) identified as Other,
.93 % (n=2) identified as Bisexual, and .47 % (n=1) identified as Questioning. The largest

Int J Adv Counselling (2014) 36:450–467 453

(modal) group of participants had an annual family income of more than $100,000 (27.10 %,
n=58). There was nearly an equal amount of participants with Master’s (49.07 %,
n=105) and Doctoral degrees (47.20 %, n=101). Seven participants (3.27 %) held a
Specialist degree, which in the United States is a type of post-master’s degree,
typically from the field of education. Approximately 47 % (n=101) of the participants
indicated that they were Counselor Educators, while 52.34 % (n=112) were
Counselors.

Measures

Perceptions of social justice advocacy

Perceptions of social justice advocacy were measured using the Advocacy
Characteristics Scales (Paylo 2007). These scales include five separate self-report
measures of advocacy attributes, attitudes, behaviors, skills, and knowledge, as well
as three measures that assess respondents’ views regarding the importance of advo-
cacy, their actual advocacy practices, and the levels at which they advocate. The
Attributes, Attitudes, Behaviors, Skills, and Knowledge of Advocacy subscales each
consist of 15 items. The Importance of Advocacy and Actual Advocacy scales each
consist of 5 items. The Level of Advocacy Scale consists of three items to measure
advocacy at the individual, school/community, and national levels. Response alterna-
tives for each item range from 1 (Strongly Disagree) to 5 (Strongly Agree).

In to reduce the amount of time required for respondent participation, the
Attributes Scale was excluded from use in this study, decreasing the number of items
on the Advocacy Characteristic Scales for the present research from 88 to 73.
Cronbach’s alphas reported by Paylo for the Attitudes, Behaviors, Skills,
Knowledge, Importance of Advocacy, and Actual Advocacy scales were .83, .91,
.88, .91, .88, and .85 respectively. Reliability estimates for the same scales for the
present study were .77, .91, .87, .89, .87, and .84 respectively.

Political characteristics

Political characteristics measured for the purpose of this study included political party
affiliation, political ideology, and political involvement. For political party affiliation,
participants chose from Independent, Democrat, Republican, and Other. Political
ideology and political involvement were measured using items taken from the
American National Election Studies Time Series Study (ANES n.d.). Since 1948,
ANES has conducted nationwide surveys to measure various political characteristics
of the U.S. electorate during presidential election years. The political ideology mea-
sure asked: When it comes to politics do you usually think of yourself as (a)
Extremely Liberal, (b) Liberal, (c) Slightly Liberal, (d) Moderate, (e) Slightly
Conservative, (f) Conservative, (g) Extremely Conservative, or (h) Haven’t thought
much about this?

To measure political involvement, participants answered Yes or No concerning their
participation in seven types of political activity, such as attending political rallies or
giving money to an individual candidate running for office during the last presidential
campaign. Additionally, participants were also asked to indicate whether they voted in
the last presidential election, as voting is traditionally considered a key indicator of
political involvement.

454 Int J Adv Counselling (2014) 36:450–467

Results

Perceptions of Social Justice Advocacy and Agreement With ACA Using its Resources
for Social Activism

As shown in Table 1, descriptive statistics indicate that participants in this study generally had
positive perceptions of social justice advocacy according to scores obtained on the Advocacy
Characteristics Scales (item means, standard deviations, medians, and modes are presented in
the Appendix). Differences between counselors and counselor educators and scores on the
Advocacy Characteristics Scales were tested using a 2 x 6 multivariate analysis of variance
(MANOVA). The main effect for the interaction between status as a Counselor or Counselor
Educator and scores on the Advocacy Characteristic Scales, Wilks’ Λ=.96, F(6, 206)=1.43,
p=.20, partial η2=.04, was not statistically significant. Follow-up univariate ANOVAs were,
therefore, not interpreted.

Regarding social activism, descriptive statistics showed that the majority of participants
either strongly agreed or agreed with ACA using its resources to advocate for social issues. In
response to the question, To what extent do you agree with ACA using its resources to advocate
for social issues?, 33.64 % (n=72) of participants indicated that they Strongly Agree;
43.93 % (n=94) said that they Agree; 12.62 % (n=27) were Neutral; 3.74 % Disagreed
(n=8); and 6.07 % Strongly Disagreed (n=13). Results from an independent samples
t-test indicated that there was not a statistically significant difference between Counselors
(M=3.97, SD=1.04) and Counselor Educators (M=3.94, SD=1.13) in terms of their level of
agreement with ACA using its resources to advocate for social issues (t(211)=−.22, p=.83,
d=.03).

Political Ideology and Perceptions of Social Justice Advocacy

Table 2 presents a summary of the political characteristics that were measured during this study
compared with national percentages where available. A 6 × 7 MANOVAwas conducted to test
differences in perceptions of social justice advocacy as measured by the Advocacy
Characteristic Scales among the various political ideologies. Results of the MANOVA showed
a statistically significant main effect for the interaction between scores obtained on the
Advocacy Characteristic Scales and political ideology, Wilks’ Λ=.63, F(42, 932.15)=2.27,
p<.01, partial η2=.07. Follow-up univariate ANOVAs revealed statistically significant differ- ences among participants from the various categories of self-identified political ideology for Attitudes, F(7, 155.73)=5.39, p<.01, partial η2=.16; Behaviors, F(7, 478.29)=6.67, p<.01, partial η2=.19; Skills, F(7, 277.71)=5.97, p<.01, partial η2=.17; Knowledge of Advocacy, F(7, 169.11)=3.94, p<.01, partial η2=.12; Importance of Advocacy, F(7, 81.14)=6.55, p<.01, partial η2=.18; and Actual Advocacy, F(7, 53.98)=3.67, p<.01, partial η2=.11. Post-hoc comparisons using Tukey Honestly Significant Difference tests (Pallant 2007) were conducted to determine the nature of the differences among the groups. Comparisons indicated that the Extremely Liberal group had significantly higher scores on Attitudes than Liberal, Slightly Liberal, Moderate, Slightly Conservative, Conservative, and Extremely Conservative groups. The Extremely Liberal group also had significantly higher scores when compared with the rest of the groups on Behaviors, Skills, Knowledge, Importance, and Actual Advocacy. Comparisons further indicated that the Extremely Conservative group had signif- icantly lower scores on the Importance scale when compared to Extremely Liberal, Liberal, Slightly Liberal, and Moderate groups. See Table 1 for means and standard deviations by groups. Int J Adv Counselling (2014) 36:450–467 455 Table 1 Means and Standard Deviations for the Advocacy Characteristic Scales Ratings Summed Scores M SD M SD Attitudesa Total 4.27 .39 64.12 5.78 Extremely Liberal 4.63 .22 69.50 3.35 Liberal 4.32 .33 64.84 5.00 Slightly Liberal 4.28 .33 64.17 4.89 Moderate 4.28 .39 64.18 5.85 Slightly Conservative 4.14 .31 62.13 4.69 Conservative 4.07 .44 61.07 6.67 Extremely Conservative 3.97 .65 59.50 9.75 Behaviorsa Total 3.82 .62 57.27 9.30 Extremely Liberal 4.50 .51 67.50 7.62 Liberal 3.84 .52 57.59 7.75 Slightly Liberal 3.78 .55 56.65 8.31 Moderate 3.89 .56 58.35 8.39 Slightly Conservative 3.53 .76 52.96 11.44 Conservative 3.52 .58 52.80 8.67 Extremely Conservative 3.16 .91 47.38 13.61 Skillsa Total 4.09 .49 61.34 7.42 Extremely Liberal 4.62 .44 69.33 6.61 Liberal 4.08 .42 61.27 6.37 Slightly Liberal 3.96 .46 59.41 6.95 Moderate 4.17 .46 62.56 6.87 Slightly Conservative 3.83 .47 57.48 7.11 Conservative 4.01 .48 60.15 7.19 Extremely Conservative 3.58 .89 53.75 13.30 Knowledgea Total 4.29 .46 64.37 6.91 Extremely Liberal 4.67 .42 70.06 6.31 Liberal 4.30 .44 64.45 6.60 Slightly Liberal 4.23 .38 63.41 5.71 Moderate 4.33 .46 65.00 6.89 Slightly Conservative 3.98 .51 59.74 7.72 Conservative 4.25 .40 63.76 6.01 Extremely Conservative 3.83 .70 57.50 10.54 Importanceb Total 3.98 .77 19.89 3.85 Extremely Liberal 4.61 .57 23.06 2.84 Liberal 4.05 .67 20.25 3.37 Slightly Liberal 3.95 .60 19.77 2.99 Moderate 4.12 .73 20.53 3.67 Slightly Conservative 3.61 .79 18.07 3.97 Conservative 3.61 .76 18.03 3.79 Extremely Conservative 2.55 1.48 12.75 7.41 Actual Advocacyb Total 3.43 .81 17.16 4.04 456 Int J Adv Counselling (2014) 36:450–467 Social and Political Characteristics and Perceptions of Social Justice Advocacy Six separate stepwise multiple regressions were conducted to describe the relationship between the dependent variables (i.e., each of the Advocacy Characteristics Scales), and the study’s independent variables (i.e., highest degree obtained, gender, age, race, sexual orientation, income, party affiliation, and political involvement). There is little empirical research or theory concerning the relationship among social justice advocacy in counseling and the independent variables tested in this study to guide their ing into a regression equation. For this reason, stepwise regression was utilized following recommendations by made Garson (2010), who suggested that stepwise regression is preferable in exploratory phases of research. Table 3 shows the unstandardized regression coefficient (B), standard error for B, standardized regression coefficient (β), partial correlation, R2, adjusted R2, and significance value for each regression analysis conducted on the Attitudes, Behaviors, Skills, Knowledge, Importance, and Actual Advocacy scales. Attitudes Results from the first stepwise multiple regression showed that political involvement (F(1, 189)=5.80, p=.02, R2=.03, adjusted R2=.03) was significantly related to scores on the Attitudes scale, explaining 3 % of the total variance. Higher political involve- ment resulted in higher Attitudes scores. The remaining variables did not enter into the equation. Behaviors Stepwise multiple regression analysis showed that both political involvement and African American race had a significant influence on Behaviors scores. Political involvement was entered in step 1, explaining 6 % of the total variance of the Behaviors scores (F(1, 189)= 13.68, p<.01, R2=.07, adjusted R2=.06). African American race was entered into the regres- sion equation with political involvement at step 2, explaining 10 % of the total variance (F(2, 188)=11.39, p<.01, R2=.11, adjusted R2=.10). As shown, both political involvement and African American race had a positive effect on scores on the Behaviors scale, indicating that Table 1 (continued) Ratings Summed Scores M SD M SD Extremely Liberal 4.11 .84 20.56 4.19 Liberal 3.34 .80 16.71 4.01 Slightly Liberal 3.42 .67 17.11 3.37 Moderate 3.61 .76 18.03 3.81 Slightly Conservative 3.19 .62 15.97 3.12 Conservative 3.11 .76 15.53 3.83 Extremely Conservative 2.95 1.10 14.75 5.50 Note. Response choices range from 1 Strongly Disagree to 5 Strongly Agree a Highest rating=5. Total possible summed score=75 b Highest rating=5. Total possible summed score=25 Int J Adv Counselling (2014) 36:450–467 457 participants with higher political involvement and those who identified as African American were more likely to agree with items on the Behaviors scale. Skills Stepwise multiple regression showed that political involvement (F(1, 189)=5.80, p<.01) had a significant influence on Skills scores. As in the Attitudes and Behaviors models, political involvement had a small influence on Skills scores. R2 was .05, and adjusted R2 was .04, indicating that political involvement accounted for 4 % of the variance in this model. Participants with higher political involvement scores had higher Skills scores. Table 2 Frequencies and percentages of participant political characteristics compared to national percentages during 2004 election Counselor Educator Counselor National a Variable Category f % f % % Political Party Independent 21 20.79 27 24.11 10.00 Democrat 54 53.47 56 50.00 49.00 Republican 18 17.82 17 15.18 41.00 Other 7 6.93 9 8.04 — Missing 1 .99 3 2.68 — Political Ideology Extremely liberal 11 10.89 7 6.25 2.00 Liberal 29 28.71 38 33.93 9.00 Slightly liberal 14 13.86 15 13.39 12.00 Moderate 16 15.84 24 21.43 26.00 Slightly conservative 9 8.91 6 5.36 13.00 Conservative 15 14.85 15 13.39 16.00 Extremely conservative 2 1.98 2 1.79 3.00 Haven’t thought much about this 3 2.97 4 3.57 20.00 Missing 2 1.98 1 .89 — Political Involvement (Yes) Talk to any people and try to show them why they should vote for or against one of the parties or candidates? 63 62.38 66 58.93 48.00 Go to any political meetings or rallies? 40 39.60 36 32.14 7.00 Do any work for one of the parties or candidates? 20 19.80 16 14.29 3.00 Wear a campaign button, put a campaign sticker on your car, or place a sign in your window or in front of your house? 42 41.58 39 34.82 21.00 Give money to an INDIVIDUAL CANDIDATE running for public office? 44 43.56 43 38.39 — Give money to a POLITICAL PARTY during this election year? 34 33.66 23 20.54 — Give any money to ANY OTHER GROUP that supported or opposed candidates? 16 15.84 23 20.54 — Vote? 96 95.05 110 98.21 77. 00 Note. a ANES n.d. Data missing from empty cells not reported by ANES 458 Int J Adv Counselling (2014) 36:450–467 Knowledge Results from the stepwise analysis on the Knowledge scores showed that political involvement and highest degree at the master’s level had a significant influence. Political involvement was entered in step 1, explaining 5 % of the total variance of the Knowledge scores (F(1, 189)=10.14, p<.01, R2=.05, adjusted R2=.05). Master’s degree was entered into the regression equation at Table 3 Summary of Stepwise Regression Analyses for Variables Influencing Advocacy Characteristics Scales Scores Step and Variable B SE B β Partial r R2 adj R2 Attitudes Step 1 0.03 0.03 Political involvement 0.48 0.20 0.17* 0.17 Behavior Step 1 0.07 0.06 Political involvement 1.16 0.31 0.26** 0.26 Step 2 0.11 0.10 Political involvement 1.26 0.31 0.28** 0.29 African American race 6.13 2.10 0.20** 0.21 Skills Step 1 0.05 0.04 Political involvement 0.80 0.25 0.21** 0.21 Knowledge Step 1 0.05 0.05 Political involvement 0.75 0.23 0.23** 0.23 Step 2 0.08 0.07 Political involvement 0.65 0.23 0.20** 0.20 Master’s degree −2.41 0.98 −0.18* −0.18 Importance Step 1 0.06 0.06 Political involvement 0.46 0.13 0.25** 0.25 Step 2 0.10 0.09 Political involvement 0.56 0.13 0.30** 0.29 Age −0.06 0.03 −0.18* −0.18 Step 3 0.13 0.11 Political involvement 0.56 0.13 0.31** 0.31 Age −0.49 0.03 −0.14* −0.14 African American race 2.26 0.88 0.18** 0.18 Step 4 0.15 0.13 Political involvement 0.58 0.13 0.31** 0.31 Age −0.49 0.03 −0.14* −0.14 African American race 2.15 0.87 0.17* 0.17 Specialist degree −3.43 1.46 −0.16* −0.17 Actual Advocacy Step 1 0.03 0.02 Political involvement 0.33 0.14 0.17** 0.17 Step 2 0.06 0.05 Political involvement 0.37 0.14 0.19** 0.19 African American race 2.36 0.94 0.18** 0.18 *p<.05, **p<.01 Int J Adv Counselling (2014) 36:450–467 459 step 2, increasing the total variance explained to 7 % (F(2, 188)=8.25, p<.01, R2=.08, adjusted R2=.07). As shown, while political involvement was associated with higher scores on the Knowledge scale, highest degree at the master’s level …