RESEARCH NOTE
Exploring the Political Diversity of Social Workers
Mitchell Rosenwald
barely audible conversation exists within
social work on the diversity of political ideology among social workers. Although diversity is a rich area of study in social work, a comprehensive exploration of social workers' political
ideologies remains largely absent (Rosenwald, 2004).
The assumption that social workers subscribe to
liberal economic, social, and moral values prevails,
as evidenced in NASW policy statements (National
Association of Social Workers [NASW], 2003), even
though this assumption has rarely been explicitly
and fully examined.This assumption occurs despite
the NASW Code ofEthics's, inclusion of respecting
fellow social workers' diversity of political belief
(NASW). As a result, ideologies of social workers
that differ from liberal political ideology may not
be represented.
A
The literature on political ideology provides some
insight into social \vorkers' political ideologies.
Political ideology, reflected in Democratic Party
membership, suggests social workers are predominantly liberal (Abbott, 1988, 1999; Epstein, 1969;
Reeser & Epstein, 1990). However, when ideology
was examined as political philosophy,findingswere
mixed; most social workers were liberal in Abbott's
(1988,1999) research but were fairly evenly liberal
and moderate (Hodge, 2003) or more moderate
(Varley, 1968) in other studies. Hodge also found
MSWs tended to be sUghdy more liberal than BSWs.
Social workers tended to be more liberal on general political ideology and specific policy positions
than people in other professions (Abbott, 1988;
Hendershot & Grimm, 1974;Jensen & Bergin, 1988;
Rubinstein, 1994) and the general public (Koeske
& Crouse, 1981; Hodge). Scant attention has been
given to studies involving social workers who idenPOLITICAL IDEOLOGIES
Political ideology refers to individuals' support of tify as radical left (but see Fisher,Weedman, Alex, &
Stout, 2001;Wagner, 1990) or radical right.
policy positions that reflect their attitudes on
society's relationship with technology, power disAlthough the literature provides a good introtribution, dependency, and nationalism (Gamson,
duction, the full range of social workers' political
1992).This support is commonly detailed among a
ideology was rarely examined as the central focus
multitiered ideological continuum (Brint, 1994;
of any study. No study examined multiple meaKnight, 1999; Lowi & Ginsberg, 1994; McKenna,
sures of political ideology as dependent variables.
1998). At one end of the continuum is a "radical
Therefore, the purpose of this exploratory study
left" ideology that focuses on major systemic changes
was to identify both the range and correlates of
to address oppression (Wagner, 1990)."Liberal"idesocial workers' political ideologies.
ology emphasizes government protection of individual rights (Lowi & Ginsberg) and the separation
METHOD
of church and state (Brint; McKenna). "Moderate"
The dependent variable of political ideology was
political ideology combines conservative and libprincipally measured by the 40-item Professional
eral views and favors rational, incremental change
Opinion Scale (POS) (Abbott, 1988) .The POS was
(McKenna). "Conservative" ideology emphasizes
selected as the main measure because it appeared to
the for-profit and voluntary sectors' abilities to adbe the most comprehensive and reliable scale that
dress social problems, socially traditional values, and
gauged political ideology by examining policy statesuspicion of government control (O'Connors &
ments linked to the social work profession.The POS
Sabato, 2000). Finally, at the opposite end of the
was based on NASW policy statements from 1985.
spectrum is the "radical right" political ideology,
Since that time, the positions reflected in these policy
which promotes policy relating to biblical literalstatements have not substantially changed as comism, the patriarchal family, and fiscal conservatism
pared with current NASW policy positions (per(Hyde, 1991).
sonal communication with A.Abbott, professor of
CCC Code: 1070-5309/06 $3.00 C2006 National Association of Sociai Workers
social work,West Chester University,West Chester,
PA, March 14, 2003), although clearly not all current policy positions from SocialWork Speaks (2003)
are reflected in the POS. The POS is divided into
four value dimension subscales: Respect for Basic
Rights (BRSS), Commitment to Individual Freedom (IFSS), Sense of Social Responsibility (SRSS),
and Support of Self-Determination (SDSS) (Abbott,
1988).Based on a five-point Likert scale (1 = conservative and 5 = liberal [accounting for reverse
scoring]), higher scores correspond with greater
liberalness (Abbott, 1988). Three new questions
linked to NASW policy statements ("Faith-Based
Initiatives," 2002; NASW, 2003) were added to the
POS to compensate for a few contemporary issues
not addressed and to compose a "POS+3" scale
(seeTable 1). Political ideology was measured using
a seven-point scale ranging from 1 = radical left to
7 = radical right on the self-ranked political ideology (SRPI) item (Knight, 1999). In sum, the study
comprises seven dependent variables: the POS, each
of the four POS's subscales, the POS+3, and the
SRPI item.
The study contained 15 independent variables
drawn fi-om the literature (Abbott, 1988,1999; Brint,
1994; Fisher et al., 2001;Jensen & Bergin, 1988;
Kornblum, 1997).They are personal characteristics
from analysis because they had more than 20%
missing data.Therefore, 294 surveys were analyzed,
which corresponds with a fair response rate of 52.6%
(Rubin & Babbie, 2001). Sufficient power was
achieved based on the sample size. Based on an a
posteriori analysis, using N = 138 as the lowest
number of participants in the regression models,
power of 0.80 (.7986) was achieved.
RESULTS
The sample was predominantly fetnale (85.6%),
white (80.1%), 45 years of age on average, Protestant (36.1%), and fairly strongly religious or spiritual {M = 1.96 with 1 = very strong and 4 = not
strong at all). In addition, most participants were
Democrat (78.1%) and heterosexual (93.7%) and
varied on income, with 25.5% earning between
$40,000 and $49,999 and 25.2% earning more than
$60,000. The majority of participants worked full
time (72.9%), held MSWs (83.6%), and were licensed at the LCSW-C level (59.8%). Participants
tended to work in public settings (36.6%) and in
nonprofit settings (35.5%). Finally,participants had
an average of approximately 13 years of licensed
experience and tended to work in clinical and direct social work practice (52.6%).
(gender, age, race, religion/spirituality, religiosity, sexual Range of Political Ideology
orientation, socioeconomic status), professional charac- The POS (M= 158.38, SD = 13.52) points correteristics {degree achieved as one proxy for education,
sponded with a fairly liberal political ideology; with
employment status, type of work setting, years of work
the addition of three items for the POS+3, the mean
experience, type of social work function, licensure level increased almost 10 points. Regarding the four
[created by author]), and two significant interac-
subscales, the SDSS had the highest mean (44.41,
tion effects {degree by years of [licensed] experience [r = SD = 4.00), followed by the BRSS (M = 44.07, SD
.25, p < .01] and gender by highest social work degree = 3.08), the SRSS (M= 37.38, SD = 5.18), and the
[X'(3, N = 285) = 8.69, p = .03]).
After a pilot test {N = 10) identified no major
problems, the study's sample was drawn from the
2003 membership list of the state of Maryland's
social work licensing board. Proportional random
sampling was conducted to ensure licensed social
workers from all four licensure levels (LSWA, LGSW,
LCSW, and LCSW-C) were represented. Participants received the survey, along with a cover letter,
self-addressed stamped envelope, and $1 as a token
of appreciation. A follow-up reminder postcard was
sent to all participants a week later (Dillman,2000).
Of the approximately 11,000 licensed social workers in Maryland, a sample of 558 participants was
obtained. Three hundred questionnaires were received for the data analysis, but six were removed
122
IFSS (M = 32.69, SD = 6.02) (Table 1).
The single item on SRPI (M= 3.39, SD = 0.92)
corresponds with a political ideology that is between liberal and moderate. No one political ideology was held by the majority of participants.The
largest self-ranking category was "liberal" (40.6%),
followed by "moderate" (34.4%). Slightly more than
half of participants (55.2%) ranked their political
ideology from liberal to radical left, and 10.4% indicated they were right of center (from conservative to very conservative). No one reported her or
his ideology as "radical right."
Correlates of Political Ideology
The significant correlates for each of the hierarchal regression models appear in Table 2. The best
Social Work Research VOLUME 30, NUMBER 2 JUNE 2006
Table 1: Means and Standard Deviations of Professional
Opinion Scale + rhree Items for Social Workers
1. All direct-income benefits to welfare recipients should be in the form of cash.
2. When they are old enough, children should have the right to choose their religion, including the option
to choose none.
3. The employed should have more government assistance than the unemployed.
4. Sterilization is an acceptable method of reducing the welfare load.
5. Counseling should be available to women who ask for abortions.
6. There should be a guaranteed minimum income for everyone.
7. Couples should decide for themselves whether they want to become parents.
8. The federal government has invested too much money in the poor.
9. The government should not redistribute the wealth.
2.19
.90
4.19
3.63
4.20
.98
.91
1.05
.80
4.49
3.59
4.62
.58
4.18
.88
3.58
1.08
10. Retirement at age 65 should be mandatory.
11. Women should have the right to use abortion services.
4.37
4.34
12.
13.
14.
15.
The dying have a right to be informed of their prognoses.
The FBI (government) should keep files on individuals wirh minority political affiliation.
Abduction by parents who do not have custody should be viewed as a family, not a legal, matter.
The government should not subsidize family-planning programs.
4.80
4.22
16.
17.
18.
19.
The mandatory retirement age protects society from the incompetency of the elderly.
Welfere mothers should be discouraged from having more children.
Family planning should be available to all adolescents.
Capital punishment should not be abolished.
20. The government should provide a comprehensive system of insurance protecting against loss of income
because of disability.
1.27
4.34
4.14
4.44
.75
1.05
.50
.99
.82
2.63
.89
.(>!
1.07
4.23
3.09
1.34
.94
4.07
4.11
.80
21. Mandatory retirement based on age should be eliminated.
22. The death penalty is an important means for discouraging criminal activity.
3.68
1.20
23. Local governments should be monitored on the enforcement of civil rights statutes.
24. The aged require only minimum mental health services.
4.04
.78
4.49
.67
25. Welfare workers should keep files on those clients suspected of fraud.
2.23
4.44
.89
.70
.(>(,
.60
26. Only medical personnel should be involved in life and death treatment decisions.
27. Pregnant adolescents should be excluded from school.
28.
29.
30.
31.
32.
Students should be denied government funds if they participate in protest demonstrations.
Juveniles do not need to be provided with legal counsel in juvenile courts.
Corporal punishment is an important means of discipline for aggressive, acting-out adolescents.
Unemployment benefits should be extended, especially in areas hit by economic disaster.
It would be better to give welfare recipients vouchers or goods rather than cash.
4.51
4.57
4.60
.84
4.31
4.10
.64
.92
.77
2.56
1.04
3.17
34. The government should have primary responsibility for helping the community accept a returning offender. 2.81
35. Efforts should be made to increase voting among minorities.
4.20
36. "No-knock" entry, which allows the police entrance without a search warrant, encourages police to
violate the rights of individuals.
3.75
37. Family planning services should be available to individuals regardless of income.
4.60
38. Older persons should be sustained to the extent possible in their own environments.
4.43
1.10
.92
33. The gap between poverty and affluence should be reduced through measures directed at redistribution
of income.
.79
1.00
.58
.82
39. The child in adoption proceedings should be the primary client.
4.34
40. A family may be defined as two or more individuals who consider themselves a family and who assume
protective, caring obligations to one another.
4.19
.95
41. Faith-based delivery of social services is an effective method of helping people in need.
42. Special laws for the protection of lesbians' and gay men's equal rights are not necessary.
43. Social services should be provided to illegal immigrants.
2.47
3.87
3.28
1.02
RoSENWALD / Exploring the Political Diversity of Social Workers
.83
1.05
1.18
123
Table 2: Significant Variables in Multiple Regression Models
Correlated with Social Workers' Political Ideology
Variable
Professional Opinion Scale (A^= 139)
Age
Independent*
Professional Opinion Scale +3 (A'^= 138)
Age
Independent
Individual Freedom Subscale (N= 161)
.416
11.884
.366
2.816
.006
.249
2.047
.043
2.809
2.308
.006
.445
.353
14.363
.273
-2.965
.208
-.185
.394
-2.036
.044
3.348
.001
5.453
6.816
.371
.295
3.003
2.952
.003
.004
Race
Age
2.731
.160
.213
.378
2.321
3.152
.022
.002
Democrat
4.639
5.293
.384
3.134
.002
.281
2.806
.006
-.710
-1.183
-.845
-.350
-.238
-3.306
-.550
-5.639
-3.120
.001
.000
Race
Age
Democrat
Independent
.023
Social Responsibility Subscale (A^ = 163)
Independent
Self-Ranked Political Ideology {N= 180)
Other religious or spiritual affiliation''
Democrat
Independent
Work status
-.247
-.154
-2.035
.002
.044
Note: All missing variable cases were excluded listwise.
'Democrat and Independent are attributes of the political affiliation variable.
"Other religious/spiritual affiliation is an attribute of the religious or spiritual affiliation variable.
regression model was for SRPI, which explained
37.6% of the variance in political ideology and had
four significant correlates. Two of the correlates
related to political party, where participants who
affiliated with the Democratic (B = -1.183, ( =
-5.639, p < .001) and Independent political parties (B = -.845, t = -3.120,p = .002) scored approximately one point lower on the SRPI scale
than those afFiHating with the Republican Party.
Two other correlates in the SRPI model emerged
with significant findings: (1) an "other" religious
or spiritual affiliation {B = -.710, t = -3.306, p =
.001), where participants scored almost one point
lower than participants who were Protestant, and
(2) work status (B = -.350, t = -2.035, j? = .044),
with participants who worked part time in social
work scoring almost half a point higher on the
SRPI scale than those who worked full time. Finally, race and age emerged as significant correlates in the other regression equations, with participants who were white and older tending to be
more liberal than those who were not white and
124
younger. The BRSS and SDSS were not significant in their final models.
DISCUSSION
As Dinerman (2003) observed, the profession has
"grown sloppy in assuming that the prevailing beliefs of our environment are, indeed, held by all" (p.
251).To this end, it was important to examine the
political diversity of social workers, which uncovered a range of their political ideologies. Most striking from the data is that social workers' political
ideology is not a liberal monolith.
With respect to self-ranked political ideology, a
slim majority reported they were liberal or very
liberal (approximately 53% total), which counters
the stereotype of a profession dominated by liberal
political ideology. Indeed, almost as many participants reported they were moderate to very conservative in ideological thought. Having more than a
third of the sample identifying as moderate suggests that the liberal versus conservative dialogue is
too simple and perhaps based on stereotypes.Those
Social Work Research VOLUME 30, NUMBER 2 JUNE 2006
with a radical left perspective were a small percentage ofthe sample. The absence ofthe radical right
suggests that either there are no social workers who
consider themselves as radical right, at least in this
sample, or the term may be unpopular for self-ranking and used more as a label by others.
The means ofthe POS-related items suggest an
overall liberal tendency among members, although
the BRSS and SDSS subscales had higher liberal
scores than the IFSS and SRSS subscales. Upon
further review, the findings confirm Brint's (1994)
conclusion that social workers tend to be more liberal on social issues than economic issues, as the top
six liberal items (means between 4.5 and 5) related
to social welfare and four ofthe six most moderate
to conservative items related to economic welfare
(all means under 3) (Table 1). Detailing economic
welfare, for example, social workers tended to be
more liberal when welfare was needed to help with
an unexpected crisis (that is, disaster, disability) and
more moderate to conservative when advocating
for clients on welfare to have fewer children and to
record those who "commit fraud." Finally,fivecorrelates from the regression models (political party
affiliation, age, race, religious or spiritual afEliation,
and work status) support their importance in the
literature (for example, Abbott, 1988, 1999) and
suggest variables to examine in future studies.
This study's limitations require a cautious interpretation of the findings. Conceptually, using the
POS (developed to measure social workers' values)
may have limited validity when applied to measuring social workers' political ideology, because some
POS items might have more significance to political ideology than others. More conceptual development and psychometric testing of a social work
political ideology scale completely based on social
work policy statements (NASW, 2003) would be
usefiil, in addition to reconciling Abbott's (1988)
POS subscales with Brint's (1994) typology of political ideology.Although the POS and POS+3 had
good rehability (a = 0.85 and a = 0.86, respectively), the four subscales had fair reliabilities ranging from 0.65 to 0.78. The study's response rate,
approximately 53%, is a fair response rate for a mailed
questionnaire (Dillman, 2000; Rubin & Babbie,
2001), yet it does not reflect almost half of the
sample; participants who responded may be different from those who did not respond. More attention to increasing the response rate would be helpful (Dillman). Finally, because the sampling frame
ROSENWALD / Exploring the Political Divenity ofSocial Workers
was licensed social workers in Maryland, the results
of the study cannot be generalized to nonlicensed
social workers in Maryland or to any other social
workers beyond that state. National studies would
be useful.
This exploratory study's central focus on political ideology raises the volume of the barely audible conversation of political ideology in social
work. Considering its range and correlates showcases the richness of this diversity variable.The intent of this study is to help establish political diversity as a legitimate diversity variable worthy of
serious study and to add to the literature on social
work and diversity. M'l'iH
REFERENCES
Abbott.A. A. (1988). Professional chokes: Values at work.
Silver Spring, MD: National Association ofSocial
Workers.
Abbott, A. A. (1999). Measuring social work values: A
cross-cultural challenge for global practice.
International Social Work, 42, 455-470.
Brint, S. (1994). In an age of experts:The changing role of
professionals in politics and public life. Princeton, NJ:
Princeton University Press.
Dillman, D. A. (2000). Mail and Internet surveys.Ttte tailored
design method (2nd ed.). New York: John Wiley &
Sons.
Dinerman, M. (2003). Fundamentahsm and social work.
Affilia, J 5,249-253.
Epstein, I, (1969). Professionatization and social work
activism. Unpublished doctoral dissertation,
Columbia tjniversity, New York.
Faith-based initiatives discussed. (2002, November).
NASW News, pp. 1,10.
Fisher, R.,Weedman, A.,Alex, G., & Stout, K. D. (2001).
Graduate education for social change: A study of
political social workers. Jowrna/ of Community Practice,
9, 43-64.
Gamson,W.A. (1992). Talking politics. Cambridge, England:
Cambridge University Press.
Hendershot, G. E., & Grimm, J.W. (1974). Abortion
attitudes among nurses and social workers. American
fournal of Public Health, 64, 438-441.
Hodge, D. R. (2003).Value differences between social
workers and members ofthe working and middle
classes. Social Work, 48, 107-119.
Hyde, C. A. (1991). Did the New Right radicalize the women's
movement? A study of change in feminist social movement
organizations, 1977 to 1987. Unpublished doctoral
dissertation. University of Michigan, Ann Arbor.
Jensen,J. P., & Bergin,A. E. (1988). Mental health values
of professional therapists: A national interdisciplinary
survey. Professional Psychology: Research and Practice,
19, 290-297.
Knight, K. (1999). Liberalism and conservatism. In J. P.
Robinson, P R. Shaver, & L. S.Wrightsman (Eds.),
Measures of political attitudes (Vol. 2, pp. 59-158). San
Diego: Academic Press.
Koeske, G. F, & Crouse, M. A. (1981). Liberalismconservatism in samples of social work students and
professionals. Social Service Review, 55, 193—205.
Kornbluni,W. (1997). Sociology in a changing world (4th
ed.). Fort Worth,TX: Harcourt Brace College
Publishers.
125
Lowi,T.J., & Ginsberg, B. (1994). American government:
Freedom and power (3rd ed.). New York: W.W.
Norton.
McKenna, G. (1998). The drama of democracy: American
government and politics (3rd ed.). Boston: McGrawHill.
National Association of Social Workers. (2003). Social work
speaks: Policy statements of the National Association of
Social Workers, 2003-2006 (6th ed.). Washington,
DC: NASW Press.
O'Connors, K.,& Sabato, L.J. (IQOQ). American government: Continuity and change. New York: Longman.
Reeser, L. C , & Epstein, I. (1990). Professionalization and
activism in social work:The sixties, the eighties, and the
future. New York: Columbia University Press.
Rosenwald, M. (2004, March). Is there room for one more?
Exploring political ideologies of licensed social workers.
Paper presented at Council on Social Work
Education's Annual Conference, Anaheim, CA.
Rubin, A., & Babbie, E. (2001). Research methods for social
work (4th ed.).Belmont, CA:Wadsworth.
Rubinstein, G. (1994). Political attitudes and religiosity
levels of Israeli psychotherapy practitioners and
students. American Journal of Psychotherapy, 48, 441—
454.
Varley, B. K. (1968). Social work values: Changes in value
commitments of students from admission to MSW
graduation. JoMrna/ of Education for Social Work, 4, 6 7 85.
Wagner, D. (1990). The quest for a radical profession: Social
service careers and political ideology. Lanham, MD:
University Press of America.
Mitchell Rosenwald, PhD, is assistant professor of social
work, Binghamton University, P. O. Box 6000, Binghamton,
NY 13902; e-mail: mrosenwa@binghamton.edu.
Original manuscript received September 28, 2004
Final revision received September 29, 2005
Accepted October 28, 200S
Sqcial
Workers Do
2nd Edition I
Maigaret Gibelman I
What Social Workers
Do, 2nd Edition is a
panoramic view of the
social work profession
in action. Extensive
case studies and
vignettes highlight the
intersection between
practice functions,
practice settings, and
practice areas, connecting what appear to be
diverse specializations.
1
,
;
;
;
•
|
Written in a lucid and engaging style that is
refreshingly jargon-free, the book flows easily
from the definitions and context of the
profession to fields of practice, issues in macro
practice, and the future of the profession. It
synthesizes source materials, current research,
case studies, and insights from social work
experts in wide-ranging fields of practice,
including mental health, health care, children
and families, aging, and more.
As with the first edition. What Social Workers
Do is perhaps the most comprehensive resource
guide currently available on the social work
profession. It is a "must-have" addition to
college and high school libraries, BSW and
MSW classrooms, and the desks of social
workers at every career level. Plus, it is an
invaluable reference tool for agencies,
policymakers, legislators, and executives.
j
|
j
I ISBN: (W71O1-364-9. 2005. Item #3649. $49.99.
NASW PRESS
ORDER TODAY!
CALL: 1-800-227-3590
Refer to Code ASWD05
Visit our web site at: www.naswpress.org
to learn about other NASW Press publications.
#NASW
Ntuioaol Auocidiiod ol SocicI V
126
Social Work Research VOLUME 30, NUMBER 2 JUNE 2006