Civic Education and the Mobilization of Political Participation in Developing Democracies
Steven E. Finkel
Department of Government and Foreign Affairs
University of Virginia
[email protected]
Paper prepared for the conference, “Political Participation: Building a Research Agenda,” Princeton
University, October 12-14, 2000. I thank Gwendolyn Bevis, Harry Blair, Larry Cooley, Paul Freedman,
Gary Hansen, James Gibson, Amanda Gouws, Robert Mattes, Chris Sabatini, Lynn Sanders, and Sheryl
Stumbras for advice, comments, criticisms, and assistance with this project.
Abstract
This paper examines the effect of adult civic education programs on political participation in two
developing democracies, the Dominican Republic and South Africa, on the core democratic value of
political tolerance. I first develop hypotheses about the effects of civic education on participation from
theories of political culture and recent work on recruitment and group mobilization. Using survey data
collected on participants in numerous civic education programs as well as control groups in both
countries, I then show that civic education has significant and substantively meaningful effects on locallevel political participation in four of the seven programs studied in South Africa and the Dominican
Republic, and that the results hold after controlling for potential biases related to the individual’s selfselection into the programs. Moreover, the effects of civic education on participation are largely
conditional in nature, dependent on variables such as the frequency and nature of the civic education
“treatment,” and the individual’s store of prior political and participatory resources. The results suggest
that civic education and other group mobilization processes are highly complementary in both countries;
civic education training stimulates individual political behavior in much the same way as does
participation in other kinds of secondary group activities. I discuss the implications of these findings for
the development of democratic political culture and for more practical considerations in the
implementation of civic education programs in democratizing contexts in the future.
2
Introduction
Can individuals in emerging democracies learn democratic values, skills, and participatory
orientations through civic education? Presumably the United States and many West European countries
believe the answer to be yes, since they have devoted considerable resources over the past several decades
to civic education as part of their larger efforts to provide democracy assistance and strengthen civil
society in emerging democracies around the world (Carothers 1996; 1999; Diamond 1995; Quigley
1997). As Quigley (1997, 564) notes, there are now a “plethora of public and private international actors
involved in these efforts .... [including]...most multilateral organizations, regional development banks,
major bilateral assistance programs, as well as literally thousands of non-governmental organizations.”
Some of these activities center around the training of lawyers, journalists and other social elites in the rule
of law, in assisting constitutional reform, and in strengthening democratic political parties and other
elements of a country’s newly emerging civil society.
Some activities, however, are aimed more directly at the mass level. Indeed, USAID’s own strategic
framework states explicitly that it seeks to “strengthen democratic political culture,” to promote
“acceptance by both citizens and political elites of a shared system of democratic norms and values,” and
to encourage citizens “to obtain knowledge about their system of government and act upon their values by
participating in the political and policy process” (USAID Democracy Strategic Framework 1998). To this
end, USAID and other donors have sponsored numerous programs directed explicitly at promoting
support for democratic norms, values, and behaviors among ordinary citizens. These efforts constitute
“civic education programs,” and range from the adoption of new curricula in primary and secondary
schools to teach young people about democracy, to programs that provide instruction about the social and
political rights of women, to voter education programs, to neighborhood problem-solving programs that
bring individuals in contact with local authorities for purposes of promoting collective action to benefit
local communities.
Surprisingly, until recently little effort has been made to assess the impact of civic education
programs on their target populations. There is an extensive literature on the effectiveness of school-based
3
civics education among children and young adults (e.g. Ichilov 1990; Jennings and Niemi 1974;
Morduchowicz, Catterberg, Niemi and Bell 1996; Niemi and Junn 1998; Slomcynski and Shabad 1999;
Torney-Purta, Schwille, and Amadeo 1999; Torney, Oppenheim and Farmer 1975). Among adult
programs, however, nearly all evaluations of civic education have looked exclusively at implementation
and management issues such as the numbers of people trained, the quality of teachers and trainers, the
quality of materials and whether the programs achieved their stated organizational goals (e.g. USAID
1992, 1989). Only in the last several years have efforts begun to evaluate the impact of civic education
on the democratic values, attitudes and activities of ordinary citizens who take part in these programs
(Bratton et al. 1999; Finkel et al., 1998, 2000; Finkel 2000).
In this paper I assess the effects of adult civic education programs on political participation. Recent
research documents the extremely low levels of social and political participation in many developing
democracies (e.g. Booth and Richard 1998; Bratton 1999), so such an investigation has the important
practical goal of illuminating whether civic education is a promising means for stimulating greater citizen
engagement in the political process. Moreover, the results can uncover the conditions under which civic
education is most (and least) effective, thereby providing critical information to policy-makers and donors
regarding the kinds of programs and methods that appear to be most capable of influencing ordinary
individuals to take part in politics.
Aside from these obvious practical concerns, however, examining the effect of civic education on
political participation has several more general theoretical aims. First, the results can shed light on the
extent to which democratic values and behaviors can be affected at all by short-term stimuli, as opposed
to changing more slowly due to long-term economic modernization, generational change, the activities of
political parties and governmental actors, and the gradual diffusion of democratic norms and ideals
through the international mass media (Almond and Verba 1963; Eckstein 1988; Gibson et al 1992;Weil
1989). A growing body of literature suggests that democratic political orientations are more susceptible
to influence from short-term economic, political, and experiential factors than traditional models of
socialization and democratic change allowed (Dalton 1994; Evans and Whitefield 1995; Reisinger et al
4
1994; Rose, Mishler, and Haerpfer 1998; Weil 1993). A finding that civic education can directly
influence political participation would lend significant support to the view that democratic political
culture in general is susceptible to short-term influence. And such changes may then at least partially be
within the reach of active efforts at manipulation by governments, political elites, or international actors.
Second, the examination of the civic education-participation relationship can provide an important
extension of recent theories that stress the role of group memberships, mobilization and recruitment in
determining mass political action in developing democracies (McDonough et al. 1998; Booth and Richard
1998; Bratton 1999). As opposed to the formal, classroom-based civics training that students receive in
school systems throughout advanced as well as developing democracies, adult civic education in most
democratizing contexts is conducted almost exclusively through secondary group associations. Hence the
investigation of civic education’s effect on political participation provides a means of assessing the ways
that the participatory appeals contained in democratic civics curricula are reinforced and amplified by
other group-related mobilization processes.
In this paper, I present the findings from two studies designed to evaluate civic education
programs in the Dominican Republic and South Africa.1 The project was conducted over the last three
years under the auspices of the United States Agency for International Development (USAID) and
Management Systems International, a Washington-based consulting firm. Using survey data collected on
participants in numerous civic education programs as well as control groups in both countries, I show that
civic education has significant and substantively meaningful effects on local-level political participation
in four of the seven programs studied in South Africa and the Dominican Republic, and that the results
hold after controlling for potential biases related to the individual’s self-selection into the programs.
Moreover, the effects of civic education on participation are largely conditional in nature, dependent on
theoretically relevant variables such as the frequency and nature of the civic education “treatment,” and
the individual’s store of prior political and participatory resources. The results suggest that civic
1
A third country, Poland, was also included in the first study. Basic findings from the Polish analysis can be found
in Finkel et al. (1998) and Sabatini et al. (1998).
5
education and other group mobilization processes are highly complementary in both countries; civic
education training stimulates individual political behavior in much the same way as does participation in
other kinds of secondary group activities. I discuss the implications of these findings for theories of the
development of democratic political culture, and for more practical considerations in the implementation
of civic education programs in democratizing contexts in the future.
Civic Education and the Mobilization of Political Participation: Theoretical Perspectives
How much impact on a political participation are civic education programs likely to have?
According to traditional views that root participation in the individual’s social-structural location, or in
basic attitudinal orientations acquired from a country’s political culture, the answer would appear to be
very little. On these views, change in democratic political culture should occur very slowly, primarily in
response to structural factors such as economic modernization (e.g. Lipset 1959), generational
replacement (Dalton 1994; Inglehart 1990; Jennings, van Deth et al. 1990), or the long-term experience of
citizens with rotations of power and a responsible opposition structure among the country’s political
parties (Weil 1989; 1993). The view of slow change in cultural values and participatory orientations was
echoed in much of the early literature in political socialization, which argued that orientations learned
early in life “structured” later adult attitudes and limited the extent to which basic values and preferences
would change in response to short-term stimuli (Hess and Torney 1967; Sears 1975).
Recent developments in the literature, however, suggest that aspects of democratic culture may be
more malleable than previously thought. A steady stream of findings over the past several decades has
shown that more immediate variables such as the individual’s perceptions of current economic conditions,
assessments of governmental competence, and experiences with governmental authority can affect
orientations such as the individual’s “normative commitment to democracy,” and the internalization of
democratic values, social and institutional trust, and political efficacy (e.g. Mishler and Rose 1997; Rose
and Mishler 1994; Dalton 1994; Brehm and Rahn 1997; Evans and Whitefield 1995; Mattes and Theil
1998). Similarly, Gibson (1998), Sniderman et al. (1996) and others have shown that individuals’
judgments about extending procedural democratic liberties towards one’s opponents can change
6
substantially in response to shifting political arguments, rhetoric, and attempts at political persuasion.
The widespread demonstration of such effects has led many to conclude that, although early socialization
and social-structural factors play a role in determining democratic orientations, these factors must be
augmented by variables related to adult political experiences. Mishler and Rose (1997, 434), for example,
explicitly posit what they call a “lifetime learning model,” where attitudes learned early in life are
continuously updated as these “early attitudes and beliefs are reinforced or challenged by subsequent
experiences.” Clearly, such a view allows a greater potential influence of civic education as another
short-term experiential effect on the individual’s overall orientation to democratic politics.
More recent political participation research provides even more optimism that civic education may
have a significant impact on the individual’s level of engagement with the political system. Following
Rosenstone and Hansen (1993) and Verba, Scholzman and Brady (1995), recent scholarship in both
advanced and developing democracies has emphasized clearly the role of active mobilization efforts in
stimulating individual political behavior. That is, individuals participate not only because they possess
the requisite socio-economic resources or favorable attitudinal dispositions, but also because they are
asked to participate, either by political parties, groups with which they are affiliated, family, friends, or
others in their social networks (e.g., Brady et al. 1999; Bratton 1999; Booth and Richard 1998; Huckfeldt
and Sprague (1992); Leighley 1996; Knoke 1990).2 Such an emphasis is echoed in the literature on social
movement participation, as a critical predictor in virtually all empirical analyses of participation in
movements concerning civil rights, nuclear energy, and the like is whether individuals are recruited or are
asked to participate or encouraged by others in the movement network (Eckberg 1988; Snow et al. 1986;
McAdam 1982, 1993; Walsh and Warland 1983).
The recent emphasis on recruitment and mobilization in the participation literature has clear
implications for the analysis of the impact of civic education. Of course, to the extent that classroom2
The importance of such mobilization processes in stimulating participation in developing democracies is perhaps
even greater than in the U.S. and other advanced democracies. Bratton (1999) shows, for example, that
memberships in parties and voluntary associations are far stronger predictors of participation in Zambia than socio-
7
based or more formal forms of civic instruction contain appeals to take part in politics, we may conceive
of any type of civic education that contains appeals to participate as one characterized by “political
mobilization.” But there is an even deeper relationship between adult civic education and political
mobilization in developing democracies. Civic education in these contexts is conducted overwhelmingly
through secondary groups and associations, sometimes by labor, church, or trade associations, but more
frequently by what Carothers (1999) refers to as “advocacy NGOs.” These groups, with names such as
Grupo Accion por la Democracia in the Dominican Republic, Lawyers for Human Rights in South Africa,
and Constitutional and Reform Education Consortium in Kenya, are public interest or reformist groups
that are funded by the U.S. and European donors in the hopes that they can become part of a “diverse,
active, and independent civil society that articulates the interests of citizens and holds government
accountable…” (Carothers 1999, 87). Funding civic education in these settings is thus part of a general
strategy of strengthening democracy by strengthening the associations that mediate between citizens and
the state.
This suggests that civic education training in newly-democratizing contexts has at least two different
goals. One is the “public interest” goal, similar to that in classroom-based or formal civic education
programs, of providing information about democratic institutions, values, and norms to individuals in
order for them to better understand their rights and obligations in a democratic system. The other is for
advocacy NGOs to encourage individuals to participate in group activities, thereby strengthening their
membership base and increasing their influence as mediating interest groups within the political process.
Thus civic education in developing democracies is intimately bound up with processes of group political
mobilization. As such, the effects of advocacy-based civic education on political participation may be
quite substantial, as the “normal” appeals of civics training to participate in politics are augmented and
amplified by the dynamics of mobilization efforts of the groups themselves.
economic and attitudinal measures, as the upheavals present in a transitional society with a weakly established
democratic culture necessitate more formal institutions to link individuals to the political process.
8
Hypotheses
The linkage of civic education with group mobilization processes also provides insight into the
precise mechanisms through which civics training may lead to increased political participation. We may
classify these mechanisms as the direct, indirect, and conditional effects of civic education on
participation, with each figuring prominently in the group mobilization literature. First, civic education
should exert a direct effect on behavior, as participants will be exposed to both participatory appeals
contained in the civic education curriculum as well to potentially powerful behavioral cues emanating
from group leaders and other group members. To the extent that individuals are motivated to adhere to
the behavioral expectations of important others in their social networks, participation in group-based civic
education training may stimulate subsequent political participation as a response to these social and
behavioral cues (McAdam 1993; Rosenstone and Hansen 1993; Finkel and Opp 1991). Further, the
advocacy NGOs themselves may differ in the extent to which they directly encourage political
participation among their members, as some groups place greater emphasis on issues such as economic
development, labor or women’s and family rights (Carothers 1999). We may therefore expect that the
stance of the group regarding the desirability of participation should be an important determinant of the
subsequent political behavior of the individuals they train through civic education (Verba et al. 1995;
Pollack 1982; Leighley 1996).
Second, we may expect indirect effects of civic education on political participation, as democracy
training may influence a series of orientations, including political efficacy, civic skills, trust, and
tolerance, that feed into subsequent political participation. According to this process, civic education
produces behavioral change, i.e. increased participation, by changing the values, attitudes, knowledge and
self-concepts that predispose individuals to take part in politics. These indirect effects are likely to be
especially powerful in group-based civic education, as a wealth of previous research has found that
involvement in voluntary group associations affects precisely the same orientations that are expected to
influence subsequent participation. Perhaps the most celebrated recent example is the effect of secondary
group involvement on political and social trust, aspects of “social capital” that serve to orient the
9
individual to engage in collective political actions (Putnam 1993; Bratton 1999; Booth and Richard 1998).
Group associations have also been found to result in increase individuals’ adherence to democratic norms
such as tolerance and civic duty, as well as to indicators of “civic competence” such as political efficacy
and civic skills (Verba and Nie 1972; Verba et al. 1995; Booth and Richard 1998; Pollack 1982). Thus
the curricular goals of civic education join the group-based nature of the programs to produce the
expectation of indirect effects of civics training on democratic political participation.
Third, we expect that civic education will also have conditional effects on participation, as the
impact of group-based programs may differ depending on variables related to the individual’s specific
civic education experience, and on variables related to the individual’s demographic and political
characteristics. As scholars since Verba and Nie (1972) have found that active group involvement has
greater mobilization effects than passive group membership (Leighley 1996; Pollack 1982; McAdam
1993), we may hypothesize first that individuals who are more actively involved in civic education efforts
within the group will be more likely to be influenced by both the curricular aspects of civics training and
the other group processes that may stimulate participation. We should therefore observe, for example,
greater effects on participation among individuals who attend more frequent civic education sessions
within the group.
But “active involvement” in civics education efforts may go beyond simply attending more frequent
training sessions. The type of instruction the individual receives in civic education is likely to matter as
well. Much research in social psychology suggests that a significant source of attitudinal and behavioral
change is role-playing behavior within groups, as individuals come to adopt attitudes and cognitions that
are consistent with the behaviors that they are acting out (Zimbardo and Leippe 1991, 102-108; Fishbein
and Ajzen 1975, chapter 10). From the group mobilization perspective, this process may be likened to the
development of participatory skills through group involvement, as individuals “practice” participation
through group involvement and learn how to transfer these skills outside the group setting (Leighley
1996; Verba et al. 1995; Pollack 1982). It may be expected, therefore, that civic education programs that
make use of more active methodologies to instruct participants --- role playing, dramatizations, group
10
decision-making, and the like --- will have a greater effect on eventual individual participation than
lecture-based instruction.
Finally, the effects of group-based civic education on participation may be greater for certain
kinds of individuals than others. Though the goal of many advocacy NGOs conducting civic education is
to mobilize dispossessed, marginal, and previously inactive constituencies, it may nevertheless be the case
that individuals require sufficient prior resources in order to translate the appeals made through civic
education training into actual behavior. Verba et al. (1995, 16), for example, claim that “requests for
activity…seem to act as a catalyst for participation among those with the wherewithal and desire to
become active.” Similarly, Rosenstone and Hansen (1993, 173-174) show that for participation in acts
such as campaign volunteering or party contributions, where the absolute level of activity is relatively
low, mobilization efforts are most successful among those who are already most likely to participate.
And Brady et al (1999) show clearly that more politically engaged individuals, i.e., those with higher
levels of interest, efficacy, and information, are more likely than less-engaged prospects to “assent” to
requests by groups to become politically active.
Thus we may hypothesize that, among the individuals trained and targeted for mobilization by civic
education groups, individuals with higher levels of prior political resources will be more likely to ---- that
is, more able and willing to --- respond to the participatory cues that emanate from both the civic
education curriculum and the groups that conduct the training. We may expect, therefore, that the effects
of civic education will vary directly with the individual’s level of education and political interest, and will
be greater in many developing democracies among men than women as well. We also expect that more
socially-connected individuals, that is, those who are more active in other secondary groups and
associations, will have greater opportunities to reinforce the messages received in the advocacy group’s
civic education efforts than more socially-isolated individuals. Hence the individual prior level of “social
capital” may be an important facilitating factor in translating civic education messages into eventual
political behavior.
11
These effects, moreover, should exist over and above the potential effects of self-selection into
civic education programs. That is, it may be the case that individuals who possess higher levels of
resources and “participatory predispositions” will be those who are more likely to be targeted for civic
education training to begin with, or more likely to volunteer to attend civic education sessions regardless
of whether or not they were specifically asked to do so. Controlling for the possible confounding effects
of self-selection biases will occupy a good portion of the empirical analyses below. The conditional
effects hypothesis claims, however, that even after such selection effects are taken into account, the
effects of civic education training should be greater for individuals with greater amounts of political
resources and social connectedness.
Research Design and Measurement
The study examines these processes by comparing the levels of participation observed among
individuals trained in four U.S. Agency for International Development (USAID) civic education programs
conducted in the Dominican Republic in the mid to late 1990s, and three programs conducted in South
Africa between 1998 and 1999, with appropriate control groups in both countries. USAID’s Center for
Democracy and Governance initiated the evaluation of civic education efforts in the Dominican Republic
and Poland in 1996-97, with South Africa added in mid-1998. The Dominican Republic was selected for
several reasons, notably the scope of its civic education efforts since the early 1990s and the relative ease
of data collection due to the small size of the country. Equally important, the country was coming out of
period of semi-authoritarianism and afforded an excellent opportunity to assess the effects of civic
education during a particular kind of political transition (see Hartlyn 1998; Espinal 1996).
South Africa was included in the study in 1998 because USAID had long targeted the country for
democracy assistance, and there were a significant number of ongoing civic education programs over the
past year in preparation for the June 1999 elections. In addition, the USAID mission in Pretoria took
particular interest in the project in order to obtain information about how better to implement civic
education and other programs in the future.
Dominican Republic
12
The first of the programs studied was conducted by a national elections oriented nongovernmental organization, Participación Ciudadana (PC). For the 1996 presidential elections, PC
created another group, called La Red de los Observadores Electorales, to organize and train youth and
adults to serve as election observers in 1996 and to conduct a quick count of the vote. The program ran
from 1995 to mid-1996, although PC activities continued into 1997, still focused on elections. Of those
in the sample 14% of the respondents were exposed to PC and Red training sessions but did not
eventually work as election observers.
The second program was conducted by a newly formed non-governmental organization, Grupo
Accion por la Democracia (GAD). The program was conducted in two phases, with the first phase
dedicated to a general educational program concerning basic political rights and obligations in a
democracy, primarily through a lecture format. The second phase brought these people together to hold a
series of national and local issues forums to discuss problems and solutions in specific policy areas, such
as justice, health, and education. Local government authorities attended these forums as well. The two
phases were intended to create a national NGO with a network of local branches outside of Santo
Domingo and to mobilize citizens to participate in these new structures. The program ran from
November 1995 to October 1996.
The third program was part of a larger community finance and small business development program
for women conducted through women’s small business NGO, Asociación Dominicana para el Desarrollo
de la Mujer (ADOPEM). The program trained women community leaders in women's rights, democratic
values, democracy in the family, and self-esteem, using a classroom/workshop format, and ran from
January 1996 to January 1997. The fourth program studied was conducted by a local NGO affiliated with
a local radio station in La Vega, Radio Santa María (RSM). The project trained intermediaries (typically
leaders of rural towns) who then conducted civic education in their local communities. The subject
matter focused on civic knowledge and values, such as rights and duties in a democracy, the importance
13
of participation, and democracy in the family. RSM ran two consecutive projects, from 1994 to 1995 and
from 1995 to December 1996.
In all of the programs in the Dominican Republic except Radio Santa Maria, treatment samples
were drawn from lists of participants provided by the implementing organizations. For the Radio Santa
Maria program, only lists of the “leaders” or first-stage participants were maintained, and we obtained
names of ordinary participants through “snowball” sampling methods from interviewers with the firststage participants. The number of individuals interviewed from the four programs totaled 1018.
The strategy for obtaining appropriate control samples was to select non-participants at random in
each of the regions where the programs were conducted. The sampling began with a national stratified
random sample of 50 municipalities, as the PC program operated nation-wide, and GAD operated in all
areas except for Santo Domingo, the country’s capital. Individuals were selected for inclusion in the
sample in proportion to the population of the selected municipality. This control sample was then
supplemented with an oversample of individuals in La Vega, where the Radio Santa Maria program
operated, and an oversample of women in the four areas where ADOPEM conducted its training. The
number of individuals interviewed for the control groups was 1017. Appendix A-1 summarizes the
participant and control samples for each of the four Dominican programs.
The in-country survey was conducted by the Instituto de Estudios de Población y Desarrollo
(IEPD), the statistical office affiliated with PROFAMILIA. Data were collected from February to April
1998. The response rate for the survey was an excellent 90.5%, with 98% response for the participant
sample and 83.7% response for the control group. Due to the lack of appropriate Census-type data, it is
impossible to assess definitively the representativeness of the sample, but the age, educational level, and
marital status of our control sample closely resemble the levels seen in the 1993 DEMOS survey
conducted on behalf of USAID, which at the time represented the last official survey of political values of
the Dominican Republic population before the current study.
14
South Africa
The South African study included three programs that conducted civic education among black
and coloured adults. The first was run by the National Institute for Public Interest Law and Research
(NIPILAR). NIPILAR is “the lead organization of a Consortium operating at the national level in the
field of public interest law, rights education with emphasis on women and children’s rights, as well as the
Constitution and Bill of Rights education. The activities aim to promote the 1) respect, practice and
fulfillment of human, legal, and civil rights; 2) respect for the rights of women and children and 3) a
widespread awareness of human rights and democracy” (USAID/Pretoria Activity Summary 1998). One
of the main civic education programs conducted by NIPILAR over the past several years was its
Women’s Rights program, designed to promote awareness of the United Nations Women and Children’s
Rights Convention.
The second program was operated through the Community Law Centre-Durban (CLC). CLC is
part of the Consortium described above, and thus has many of the same goals and activities as NIPILAR.
CLC, however, operates almost exclusively within the province of KwaZulu Natal, where NIPILAR does
not operate. Its primary activities are to coordinate approximately 30 rural legal advice offices in the
province. The advice offices provide assistance to community members on legal and human rights issues.
Democracy and civic education workshops are also conducted through the advice centers.
The third South Africa program was conducted by Lawyers for Human Rights (LHR). LHR is a
national organization aiming to increase the awareness of human and democratic rights in South Africa.
The organization holds an extensive series of workshops yearly on democracy and human rights issues,
with different aspects of democracy receiving particular emphasis in different years. Workshops in the
last two years have emphasized the Constitution and the Bill of Rights, and participation in politics,
respectively.
The first two groups were selected primarily because of the support that USAID-South Africa has
provided for their civic education efforts. LHR was included in the study in order to examine a nonUSAID-funded group, and because it is a well-known NGO promoting democracy and human rights in
15
South Africa. LHR also conducted civic education in eight of the nine provinces of South Africa;
NIPILAR was also more or less national in focus, while CLC-Durban’s area of operation was mainly
within one province, KwaZulu Natal.
The three NGOS operate in generally similar fashion with regard to their civic education activities.
Representatives from the central offices train a core group of individuals, called “paralegals,” in
democracy and human rights instruction. These activities, generally known as Training of Trainers
(ToT), consume a considerable amount of the group’s time and resources. The paralegals then go on to
operate offices in villages and towns across country, from which they provide a number of services for
individual residents. Some of these services have nothing to do with civic education, for example
providing advice on economic development or labor law. However, the paralegals are also expected to
conduct community workshops on different aspects of democratic governance and human rights, and
these activities are the focus of our study. According to interviews with the groups’ staff conducted by
members of our research team, the number of workshops throughout the country are claimed to be in the
hundreds yearly by LHR and NIPILAR, reaching many thousands of ordinary citizens.3
As in the Dominican Republic, the treatment group interviews were obtained through sampling
lists of civic education participants provided by the three NGOs (either national representatives or the
facilitators or paralegals who ran the civic education programs in each of the areas). Participants were
selected systematically from the lists whenever addresses and contact information was provided. In
regions where no lists of names and addresses existed, the facilitators or paralegals themselves located the
requisite number of participants and provided contact information to the South African survey
organization, Markinor, which collected the data. The sample of participants is shown in Appendix A-2.
The control group of non-participants in South Africa was designed slightly differently than in the
Dominican Republic. Instead of aiming to produce a random sample of the South African black or
coloured population, we attempted to introduce more rigorous experimental control at the outset by
16
“mirroring” the participant sample on a number of important demographic dimensions. Interviewers were
instructed to conduct an interview with a civic education participant selected according to the procedures
just described, and then to conduct an identical interview in the same area with a person who had not
participated in civic education. The control group respondent was to be the same race, gender, and age
group as the participant. Interviewers were instructed to make a systematic selection of houses, beginning
with the third house from the civic education participant who had been interviewed, in order to find an
appropriate non-participant for inclusion.
These sampling procedures produced a total of 1550 interviews for the study, with the final data
collection conducted between 10 May and 1 June 1999. The questionnaires were translated into Zulu,
Xhosa, Tswana, Northern Sotho, Southern Sotho, and Afrikaans so that respondents could be interviewed
in the language with which they felt most comfortable. The final sample consisted of: 475 adult
participants in civic education, and 475 adult non-participants who were matched on race, gender, and
age. 10 individuals from the treatment group were eliminated from the analysis because it became
unclear in the course of the interview how many workshops they had attended, or whether they had been
exposed to civic education “treatment” at all.
Measurement of Political Participation
The dependent variable in the analysis in both countries consists of four behaviors that commonly
take place at the local or community level: taking part in organized community problem-solving activity;
attending a local government meeting; working in an election campaign; and contacting a local elected
official. I focus on local-level participation because of the emphasis placed in civic education training on
relating abstract concepts about democracy and citizen participation to local-level political issues and
institutions. The scale used in the analysis thus comprises five points, running from zero (no behaviors)
to four (all behaviors). The reliability of the scale is .64 in the Dominican Republic, and .77 in South
Africa.
3
The research team included myself and Sheryl Stumbras of Management Systems International, Robert Mattes, a
public opinion and democracy specialist at IDASA (Institute for Democracy in South Africa), and Dumisani Hlophe,
17
Independent Variables
I include a number of attitudinal and demographic control variables in the analysis. A series of
democratic orientations were included as potentially intervening variables in the “indirect effects” model
described above: political knowledge, civic skills, political efficacy, political tolerance and political trust.
Details on these questions and the resultant scales can be found in the Appendix.
I also included a series of demographic and political control variables, including educational
attainment, age, income, gender, race, religiosity (church attendance), interest, media exposure, previous
voting behavior, and the number of voluntary organizations to which the individual belongs. Details on
these items are also found in the Appendix.
Statistical Methods
After presenting the simple bivariate relationships, I estimate several models that attempt to
isolate the effect of civic education on political participation, controlling for other known determinants of
participation, as well as controlling for the selection biases that are inherent in the civic education
programs examined in the two countries. The fundamental problem in assessing the effect of civic
education on participation (or other democratic orientations) is that, in the absence of a pre-test and/or
randomized assignment of individuals to civic education “treatments,” we cannot rule out with absolute
certainty the possibility that individuals who were trained in civic education workshops already possessed
those attributes that correlate with participation, or that pre-disposed them to increased participation
tolerance in the absence of any “treatment” whatsoever. Indeed those attributes, such as education, group
memberships, political interest, and the like, are exactly the factors that may lead individuals to attend
civic education workshops in the first place. Thus any observed difference between civic education
participants and the control group on participation many be due to the pre-existing differences on these
other variables.4
a PhD student specializing in democratization and civil society at the University of Durban-Westville.
4
These problems are endemic in the evaluation of policy programs in general. For overviews of these issues and
some of the available solutions, see Achen (1986), Moffitt (1991), and Stromsdorfer and Farkas (1980).
18
The most basic approach for dealing with these selection biases is to include all other variables
that are known to be related to both civic education exposure and participation into the statistical model.
These variables are entered along with the treatment variables in an OLS multiple regression analysis (or
equivalently, through analysis of covariance methods), which estimates the effect of the treatment on
participation after taking into account the differences between the treatment and the control groups on the
other variables included in the model. Thus the treatment effect in these models represents the estimate of
how much civic education affects participation, over and above the effects of all the observed control
variables. I present these estimates below under the label “Regression Model --- OLS”.
As many scholars have shown, however, the OLS approach fails to account for an additional, and
potentially important, source of bias in the estimation of treatment effects. As Achen (1986), Barnow et
al (1980), Heckman and Robb (1985), and others suggest, there may be differences between the treatment
and control group on relevant unmeasured variables that influence both the decision to attend a treatment
program as well as the program’s desired outcome. For example, individuals who decide to attend civic
education workshops may differ from other individuals not only in terms of observed characteristics such
as group memberships, educational attainment, and political interest, but also in terms of unobserved
variables such as their intrinsic predisposition toward democracy, their motivation to succeed in a
democratic society, their need for sociability, and the like. If these factors related to self-selection are
also positively (or negatively) related to participation, then estimates of the treatment effect of civic
education will be biased, as the estimated regression coefficient for attending civic education would
include some of the effect of these unmeasured variables as well.
More technically, the problem exists because of the potential for a correlation between the error
terms in the selection equation (i.e. the decision to attend a civic education workshop) and the outcome
equation (the prediction of political tolerance), due to unmeasured factors or to random perturbations that
influence both the decision to participate and the outcome in question (Berk and Ray 1982; Breen, 1996,
35). To correct this problem, which biases estimates of coefficients in the outcome equation, Heckman
19
(e.g. 1992) has proposed the following two-step procedure, also discussed at length in Achen (1986),
Green (1993, 713-174), Vella (1997, 135-138), and Winship and Morgan (1999, 669-687).
In the first step, the decision to participate in the treatment program is modeled via probit
analysis, with individuals who were exposed to the treatment T coded as “1” and the control group coded
as “0.” Thus
(1) Ti = γ k w i + υ i
where the w represent all independent variables that predict whether individuals attend civic education
workshops or not, the γ k are their respective regression coefficients, and the υ i are assumed to be
normally distributed. The “generalized probit residuals” (Vella 1997, 136) from this equation are then
calculated as
(2) λ
i
= φ (γ k w i ) / Φ (γ k w i )
for the civic education participants (T = 1) and
(3) λ
i
= − φ ( γ k w i ) /( 1 − Φ ( γ k w i ))
for the control group (T = 0) , where φ ( γ k w i ) represents the height of the normal distribution (the
probability density) at the point γ k w i , and Φ ( γ k w i ) represents the cumulative probability at the same
point.5
The second step in the process is to use those residualized estimates as an additional independent
variable in the outcome equation, as in:
(4) y i = β
k
x i + βt T + βλ λ i + ε i
where the x represent all independent variables that affect the outcome in question, T represents the
treatment, and the β are respective regression coefficients. It can then be shown that the regression
coefficient for the generalized residual term, β λ , is an estimate of Rho ( ρ ), the correlation between the
20
errors in the selection and outcome equations, multiplied by the outcome equation’s standard error of
estimate ( ρ σ ε ) (Greene 1993).
According to this model, the difference between individuals in the civic education group and the
control group on tolerance will be equal to:
(5) β t + ρ σ
ε
[φ
i
/(Φ i (1 − Φ i )
]
or the “true” effect of the treatment, β t , plus some bias term that is a function of Rho, the standard error
of estimate σ ε , and the generalized probit residuals from the selection equation. If Rho is positive, this
means that, other things being equal, the bias term will be larger, and the estimated “true” effect of civic
education on participation will be smaller than in an OLS formulation.6 If Rho is negative, bias term will
be smaller, and the estimate of the “true” effect of civic education on participation will be
correspondingly larger than the estimated obtained in OLS. If Rho is statistically indistinguishable from
zero, then the unmeasured factors that lead individuals to participate in civic education programs, over
and above the variables that are included in the participation equation, are irrelevant for the prediction of
participation. In that case the results from what I will refer to as the “Self-Selection” or “Heckman
Model” and the OLS regression model will be substantively equivalent. The models were estimated
using LIMDEP 7.0.
Results
Bivariate Findings
Table 1 displays the simple percentage differences in self-reported participation between
individuals who were exposed to civic education in the four Dominican programs and three South African
programs, along with the percentages for individuals who received no civic education (the control group).
5
This expression is often referred to as the Inverse Mills Ratio (IMR). The IMR decreases monotonically for the
treatment group as the probability that T = 1 increases, and decreases monotonically for the control group as the
probability that T = 0 increases.
6
Intuitively, a positive Rho means that individuals who were treated have larger error terms in the outcome equation
than individuals in the control group. OLS ignores this fact and produces an exaggerated impact of the treatment on
21
(Table 1 goes here)
As can be seen in both countries, there are substantial differences between civic education participants
and the control group on all of the political participation items. Individuals who were exposed to civic
education in the Dominican Republic, for example, were roughly twice as participatory as individuals in
the control group, with the largest effects being seen for community problem-solving participation and
attendance at local municipal meetings. The differences vary substantially between programs as well, as
it appears that GAD participants in particular are extremely active, with ADOPEM participants only
somewhat more participatory than the control group, and the two other programs falling in between. On
the overall participation scale, GAD participants average a very high 2.31 behaviors (out of 4), over three
times the size of the control group’s average of .73, while the other program means are roughly 1.5 to 2
times the mean of control group. Clearly civic education participants in the Dominican Republic are
much more active in a variety of political behaviors than the average Dominican individual, though of
course we cannot yet claim that such differences can be attributed to the civic education experience.
Differences of slightly smaller magnitude can be found between civic education participants and the
control group in South Africa.7 In the bottom half of Table 1, it can be seen that participation in two
programs – Lawyers for Human Rights (LHR) and the NIPILAR women’s rights program – is associated
with significantly higher levels of participation on each of the four political participation items than the
control group. Participants in the CLC program, however, which operated exclusively in the politicallytroubled KwaZulu Natal region, showed no differences in participation over the control group. The
differences for LHR and NIPILAR over the control group are of similar magnitude for each participation
item, with both LHR and NIPILAR respondents showing participation rates at roughly 1.5 to 2.9 times
the control group average. On the overall scale, the control group mean is just under 1 behavior (out of 4
the outcome. A negative Rho indicates the reverse; smaller error terms in the outcome equation for treated
individuals, and consequently an underestimation by OLS of the “true” causal impact of the treatment.
7
Recall, however, that the control group in South Africa consisted of individuals who were closely matched to the
CE sample on age, race, gender and place of residence; to the extent that these factors are relevant to political
participation, smaller bivariate differences would be expected to emerge before controlling for these variables in
multivariate analyses.
22
total), with both LHR and NIPILAR respondents reporting between 1.6 and 1.8 behaviors on average.
Clearly, participants in two of the three South African programs are also more politically active at the
local level than non-participants.
Multivariate Analysis and Controls for Selection Effects
Tables 2A and 2B present the results of multivariate analysis of the effects of civic education on
the overall political participation scale in the Dominican Republic and South Africa, respectively. In
model (1) in both tables, the effects of each of the civic education programs are shown, controlling for a
series of demographic and political variables that may relate to both political participation and to the
likelihood of exposure to civic education. Model (2) displays the effects of the same independent
variables in the context of the self-selection model, which introduces an added control for unobserved
factors that may influence both the probability of exposure to civic education and participation. This
control variable is denoted as “Inverse Mills Ratio (Lambda)” in Model 2 in both countries.
(Tables 2A and 2B go here)
The results of these analyses indicate that a significant portion of the bivariate differences between
Civic education participants and the control group on political participation can be attributed to the
selection process; that is, CE participants in general have many of the demographic and political
characteristics that predispose them to be more active politically than control group individuals. In Model
1 in the Dominican Republic, for example, it can be seen that controlling for these factors --- in particular
the individual’s level of involvement in secondary associations, political interest, urban residence and
gender --- completely eliminates the observed differences from Table 1 in participation rates for PC and
ADOPEM individuals compared to the control group. Nevertheless, significant differences in overall
political participation rates remain for GAD respondents and individuals trained in the Radio Santa Maria
program, with the GAD effect of .73 being about one-third of its bivariate value, and the RSM effect of
.31 being about one-half its bivariate size. The effects for both programs in the Dominican Republic are
substantively meaningful as well, with CE participation being among the strongest predictors of political
behaviors among all variables in the table.
23
The significant effect of the GAD and RSM civic education programs also persists after controlling
for potential biases in the context of the Heckman self-selection in Model (2).8 As discussed above, the
first stage in the self-selection process is to model explicitly the decision to attend civic education
workshops, and to produce an estimated residual from this equation for both civic education participants
and the control group which is introduced into the model predicting political participation. The results of
the probit model predicting civic education participation indicate that a series of demographic (age,
education, gender, rural residence) and political factors (interest, group memberships, prior voting
behavior) are associated with exposure to the civic education “treatment.” And as can be seen in Model
(2) in Table 2A, the coefficient estimate for the Inverse Mills Ratio is .09, which also translates into a .09
estimate for Rho, the correlation between the error terms in the civic education treatment and participation
equations (as Rho equals the coefficient estimate multiplied by 1.03, the standard error of estimate for the
participation equation). This estimate is statistically significant, indicating that there is a small positive
residual correlation between civic education treatment and political participation, once the observed
variables are taken into account.
Substantively, however, the residual error correlation between the treatment and outcome equations
has little relatively impact on the model’s overall results. The effect for GAD is nearly identical to its
OLS estimate and the effect for the RSM treatment falls by about 16% to .26. There are concomitant
increases in the estimates of some of the control variables, as the effect of group memberships, political
interest, urban residence, and prior voting behavior each register increases over their OLS values.9 But
8
I report the program effects only for GAD and RSM in the Heckman model in Table 2. The estimated effects for
PC and ADOPEM, already insignificant in the OLS formulation in Model 1, were also insignificant in Heckman
models that were estimated separately.
9
This occurs, as Achen (1986) explains, because these variables all have significant positive effects in the treatment
equation. Individuals in the treatment group, for example, are disproportionately active in secondary groups, so
individuals in the treatment group who are inactive in secondary groups must have had large positive error terms in
the treatment equation. By virtue of the positive correlated error in the treatment and outcome equations, they are
also somewhat more likely to participate politically, over and above their objective score on the group memberships
variable. So the treatment group is composed of individuals who are highly active in secondary groups and atypical
individuals who are inactive in secondary groups yet otherwise predisposed to participate politically. Similarly, the
control group is composed of individuals who are less active in secondary groups and atypical individuals who are
more active in secondary groups yet otherwise predisposed not to participate politically. Hence the estimated effect
24
the overall conclusion from the two models is clear: civic education treatment in two of the four
Dominican programs has substantial effects on the individual’s propensity to participate in politics at the
local level, over and above the fact that treated individuals differ from the control group on a host of
observed (and unobserved) factors that predict political participation as well.
Table 2B shows the similar analyses conducted for the South African sample. Model (1) shows that,
controlling for a series of demographic and political factors, individuals who were treated in the LHR and
NIPILAR programs were significantly more participatory than individuals in the control group. These
differences ---- about .4 on the overall participation scale --- are approximately 50% the size of the initial
observed differences seen in Table 1. As in the Dominican Republic, the civic education effects are
among the strongest in the Table, with differences between the two treatment groups and the control
group being larger than, for example, the relatively sizeable differences in participation between men and
women and between black and coloured respondents.
The self-selection model (2) in the table shows a nearly identical set of results, primarily because the
estimated correlation between the error terms in the selection and outcome equations is statistically
insignificant. There is, in fact, a slight negative estimate for the effect of the Inverse Mills Ratio
(Lambda), which translates into an estimated error term correlation of -.03. This pattern, though
statistically insignificant, is congruent with the findings from a previous paper on political tolerance
(Finkel 2000), where it was found that individuals in the civic education treatment groups were also
predisposed towards lower levels of tolerance than the control group. In the selection equation for South
Africa, the findings indicate that key variables such as education and income are in fact negatively related
to CE treatment, in contrast to the pattern seen in the Dominican Republic. This means that civic
education is attracting individuals who possess some characteristics that are negatively related to
participation in the South African context, and the negative value for Rho in the Heckman model suggests
that unmeasured factors in the selection process to a small degree reinforce this tendency. The effect is
of group membership is dampened in the OLS model, as OLS does not take into account the negative relationship
between group memberships and the error term of the treatment equation.
25
not large enough to produce differences between the OLS and self-selection models; nevertheless it shows
that predispositions to participation are not necessarily higher among the types of individuals who are
participants in democracy training sessions, and this makes successful civic education more difficult to
achieve in the South African context.
How can the variations in program effects within and across countries be explained? It was
hypothesized above that the degree to which individuals who were trained in particular groups became
active politically would vary directly with the importance placed on participation by the group itself and
the resultant behavioral cues imparted by the group leadership and other group members. This cannot be
proven definitely with the data at hand, but such an interpretation is almost entirely consistent with the
results seen in Tables 2A and 2B. The participants from Participacion Ciudinada (PC), for example,
were given civics training specifically to prepare them for roles as observers in the 1996 Presidential
elections, not to mobilize local-level participation more generally. Similarly, ADOPEM’s primary
emphasis was in the area of women’s development in the economic sector; its civic education functions
were of secondary importance. By contrast, the raison d’etre of the GAD program was to promote local
level problem-solving and community action, while Radio Santa Maria in the Dominican Republic, and
LHR and NIPILAR in South Africa were all broad-based training programs emphasizing knowledge,
skills, values and participation through community workshops conducted by the advocacy groups’
“paralegals”. The only finding that does not conform to this hypothesis is the low level of mobilization
seen among CLC participants in South Africa. This program’s emphasis and general structure was
similar to RSM, LHR and NIPILAR, yet the results were much less satisfactory in terms of stimulating
post-workshop political participation.
The Indirect Effects of Civic Education
The results thus far suggest that at least some civic education programs can have a substantial
stimulating effect on local-level political participation in both the Dominican Republic and South Africa.
Yet it remains unclear whether the effects are the result of the behavioral cues and direct mobilization
appeals made by the groups conducting the civic education training, or whether civic education and other
26
group-related processes influence participation indirectly, through their influence on other democratic
skills, values, or participatory orientations. Model (3) in Tables 2A and 2B show the results related to this
hypothesis in the Dominican Republic and South Africa, respectively. In each country, the same
Heckman self-selection model from (2) was estimated again, after including five important democratic
orientations: political knowledge, civic skills, efficacy, tolerance, and political trust.
The results show weak support for the “indirect effects” model. In Table 2A, it can be seen that only
two of the intervening variables, political knowledge and civic skills, are themselves significant
determinants of local level participation, while efficacy, tolerance and trust are irrelevant, once other
variables are taken into account. And controlling for all of these factors in Model (3) decreases the effect
of the GAD program on participation by only about 10%, while the effect of the RSM program increases
somewhat over the “reduced form” estimates from Model (2).10 This indicates that the overwhelming
portion of the initial civic education influence on participation cannot be explained though indirect
processes; by far the more substantial effects appear to be the result of direct political mobilization from
participation in the group-based civic education programs.
Table 2B shows a very similar set of results from South Africa. Among the potentially intervening
variables, knowledge and efficacy have direct effects on participation, while skills, tolerance and trust are
irrelevant. Controlling for all these factors in turn reduces the impact of the two significant civic
education programs by only 13% in the case of NIPILAR, and 25% in the case of LHR. As in the
Dominican Republic, then, there is some indirect effect of civic education on participation through
knowledge and efficacy, but most of the direct effects persist after taking these variables into account.
Previous analyses of these programs have found significant positive effects of civic education on
political tolerance in both countries (Finkel 2000) , and significant effects (negative in the Dominican
Republic, positive in South Africa) on trust in social and political institutions (Finkel et al 2000; Finkel
and Stumbras 2000). But neither of these factors appears to be directly related to political participation.
10
The RSM coefficient increases because of an unexpectedly negative effect of civic education training on skills
among participation in that program.
27
Thus civic education affects a range of important democratic political orientations, but the bulk of its
effect on participation does not operate through attitudinal change or changes in civic competence,
knowledge, or skills. The evidence thus far suggests that civic education training --- at least training that
is conducted within 4 of the 7 programs examined in the overall study --- represent direct mobilizing
instruments for local level participation in both the Dominican Republic and South Africa.
“Conditional Mobilization”: The Role of Civic Education Frequency and Pedagogical Methods
According to the “conditional mobilization” model discussed above, the effects of civic education on
participation should depend on factors related to the nature of the individual’s civic education experience,
and factors related to the individual’s previous store of political resources. That is, we expect to find
greater effects when individuals receive more frequent exposure to the mobilizing messages of civic
education, when those messages are taught through more intensive, involving participatory
methodologies, and when the individual has sufficient political resources to act on the messages received
through civic education training. In this section I test the conditional mobilization effects of factors
related to the civic education training per se; in the following section I analyze the role of the individual’s
prior political resources in facilitating the acceptance of mobilization appeals.
To measure the frequency of exposure to civic education, respondents were asked how often they
had attended sessions sponsored by the particular program in which they were engaged. Response
categories ranged from once, twice, three times, to four or more exposures. To measure the extent of
exposure to participatory teaching methods, respondents were asked which of the following methods were
used in the programs they attended (breaking into small groups, staging plays or dramatizations, playing
games, problem-solving, and simulations or role-playing). In the Dominican Republic respondents were
asked further whether they had much opportunity to express their own views in the program, and in South
Africa whether workshops had included “staging mock trials of legal proceedings,” or “staging mock
elections or other kinds of political activities.” The scales thus ran from 0 to 6 in the Dominican Republic,
and 0 to 7 in South Africa. Both variables were then entered into the Heckman self-selection model in
place of the simple dummy variable for program participation seen in Tables 2A and 2B. The analyses
28
were limited to those programs where significant effects were found in the earlier tables. Thus PC and
ADOPEM respondents were eliminated from the Dominican Republic analysis, and CLC respondents
were eliminated from the analysis for South Africa.
The results strongly support the conditional mobilization hypotheses in both country contexts. In the
Dominican Republic, the results indicate that each training session the individual attended was associated
with an increase in local-level participation of .09 (s.e. of .04, p<.05), while each participatory method
used in CE instruction was associated with an increase in local-level participation of .20 (s.e. of .04,
p<.01). The corresponding results in South Africa were .14 for CE frequency (s.e. of .05, p<.01), and .22
for participatory teaching methodologies (s.e. of .06, p<.01). Thus individuals who were trained more
frequently, with more participatory instructional methods, showed greater increases in local-level
participation over the control group than civic education participants who were trained less frequently,
with more traditional classroom or lecture-based teaching methodologies.
(Table 3 goes here)
I summarize these effects in Table 3 by showing the predicted increment in local participation over
the control group for individuals whose civic education experience was either more or less frequent, and
accompanied by greater or fewer participatory teaching methodologies. As can be seen, in the Dominican
Republic (Table 3A), individuals who were trained once or twice with fewer than 4 participatory methods
registered, on average, a .44 increase in participation over the control group. This figure increases to .62
for individuals who were trained more often (between 3 and 4 times), but with few participatory
methodologies. Larger increases still are seen for individuals whose training encompassed more active
methodologies, with an increment of over 1 behavior on the 4 point participation scale being registered
for individuals who were training often and with many active methods. The effects are somewhat more
dramatic for the South African sample: individuals who are trained infrequently and with more
traditional methods register .65 increments on the participation scale over the control group; this figure
more than doubles to 1.37 behaviors for individuals who were trained both frequently with many active
29
methods.11 Thus, the effectiveness of mobilization appeals in civic education depend directly on the
nature of the individual’s experience with the training itself: the more exposure to civic education, the
more individuals come into contact with mobilizing appeals by the group; and the more participatory
methods that are used in instruction, the more individuals develop “practice” in political participation that
subsequently influences later activities outside of the civic education context.
“Conditional Mobilization”: The Role of Political Resources
The second set of hypotheses in the “conditional mobilization” model concern the effects of civic
education on individuals who possess different amounts of participatory resources such as education,
political interest, and memberships in voluntary associations. As suggested by recent mobilization
literature (Rosenstone and Hansen 1993; Brady et al. 1999), the effects of active recruitment efforts by
parties, groups, and other mobilization agents should depend on the individual’s prior dispositions to
participate. If individuals do not possess a sufficient amount of information, skills, or social attachments,
they may be unable to act on the information imparted in civic education training, and hence unable to
translate mobilization appeals into actual political participation. This suggests an interaction effect
between exposure to civic education and various indicators of prior participatory resources and skills.
To test these hypotheses, I created simple interaction terms between the civic education treatment
and a set of demographic and political factors that correspond to participatory resources in both countries:
education, group memberships, political interest, gender (male) and, in South Africa, race (black). In
South Africa especially, there was very high multicollinearity between the various civic education and
resource products. I therefore created dichotomous indicators of group memberships (zero or 1 group
11
Interpretation of larger effects in Table 3 for South Africa should be tempered by the fact, compared to the
Dominican Republic, a smaller proportion of the South African civic education sample was trained frequently (33%
versus 43%), and a smaller proportion was trained with more active teaching methodologies (48% versus 51%). The
table thus shows the potential effect of civic education in both contexts under certain conditions; the favorable
conditions were met more frequently among programs in the Dominican Republic. The distinction between actual
and potential civic education effects will be discussed further in the conclusion.
30
versus 2 or more) and political interest (divided at the 50th percentile).12 The civic education treatment
variables, resources, and the interaction terms were then entered into the Heckman self-selection model of
Table 2. I show the effects of civic education treatment, prior resources, and the interaction terms in
Table 4; the effects of the other demographic and political control variables are omitted to simply the
presentation of results.
(Table 4 goes here)
Table 4 indicates that the individual’s prior political resources condition the effects of CE in both
countries in important --- though not fully consistent ---ways. In the Dominican Republic, all four
interaction terms attain statistical significance, despite a reasonable amount of collinearity between the
civic education and resources product terms. The model R-squared of .37, moreover, represents a
substantial improvement over the .34 value from Table 2. Substantively, the results indicate that CE has
larger effects on local participation among men than women, and stronger effects among politically
interested and socially attached individuals as well. For example, the marginal effect of GAD civic
education training for women is .17, while the effect is over three times that amount (.53) for Dominican
men. The social and structural barriers to mobilizing women to participate in the Dominican context are
thus reinforced by civic education training., despite the fact that women in the GAD treatment group
participate at rates similar to men in the control group (.21). Thus civic education serves to mobilize
those with fewer social resources (women) to some degree, but greater effects are seen for those whose
resources facilitate the translation of mobilization messages into actual behavior. Similar interaction
effects are seen in the Dominican Republic for political interest and group memberships: more highly
interested, socially attached individuals are substantially more influenced by civic education mobilization
appeals training than less interested, socially isolated trainees.
12
This strategy has the drawback of preventing the testing of curvilinear relationships between civic education and
prior resources (such that treatment at very high levels of resources may be less effective than treatment at moderate
levels). I tested trichotomous versions of education, interest, and group membership variables in the subsequent
analysis and found no evidence of improvement in the model fit.
31
The same general pattern of interaction effects is seen in South Africa, though the collinearity
between the various product terms renders the conclusions somewhat more tentative. It can be seen that
significant interactions exist between civic education and political interest, and between civic education
and the individual’s other group memberships. The interaction effects for education, gender and race are
irrelevant. This indicates that highly interested individuals who are trained in civic education sessions in
South Africa are substantially more likely to participate in subsequent political activities than less
interested trainees, and individuals who belong to at least two secondary associations participate more
frequently as a result of civic education training as well. In all of these cases, mobilization efforts within
the context of civic education are more successful when individuals have higher levels of prior resources,
presumably reflecting greater ability and willingness to translate the messages of civic education training
into concrete action.
There is one exception, however, to this general pattern, and that is the negative interaction in the
Dominican Republic between civic education exposure and education. This indicates that, controlling for
all other variables, individuals at higher education levels are less influenced by civic education appeals
than individuals at lower levels of education. The effect, however, should be understood within the
context of the overall effect of education on political participation in the Dominican Republic. It can be
seen from Table 2A that education has a slight negative impact on local level participation in general, a
pattern that conforms with Bratton’s (1999) results in Zambia as well. Thus it is not always the case that
educational attainment is positively associated with political participation in developing democracies.
What does appear to be the case, however, is that civic education reinforces the resource-disparities that
exist for political participation otherwise. That is, variables such as political interest, gender (male) and
group memberships are positively related overall to political participation in the Dominican Republic, and
the interaction of civic education and these resources on participation is therefore positive. Education is
negatively related to participation, and the negative interaction of civic education with education
reinforces this effect as well. This pattern of a reinforcement effect of civic education on resource-based
32
disparities in participation, again, is fully consistent with a view of civics training as deeply embedded in
more traditional group mobilization dynamics of developing democracies.
Conclusion
This study of the impact of adult civic education on political participation in two developing
democracies, the Dominican Republic and South Africa, showed that democracy training has significant
effects on local-level participation in four of the seven programs examined in the two countries. The
effects remained significant in multivariate models, and in models that attempted to control for the selfselection processes that are inherent in the implementation of civic education programs. The effects were
only modestly explained through the impact of civic education training on other democratic orientations
such as efficacy, knowledge, tolerance, and trust; almost all of the direct effect of civic education
persisted after these potentially intervening variables were taken into account. Finally, the effects of civic
education on participation in both countries varied considerably, depending on the frequency of the
individual’s exposure to civic education training, the extent to which the training was conducted with
active, participatory teaching methodologies, and the level of the individual’s prior participatory
resources. In short, civic education can mobilize individuals in developing democracies to participate in
politics, but not unqualifiedly so. The success of civic education efforts depends directly on the same
factors that mitigate or enhance successful group mobilization in general.
The findings have important theoretical implications for our understanding of the development of
democratic culture, along with several practical implications for the implementation of civic education
programs in emerging democracies. First, they lend additional credence to the growing claim that
democratic political culture can change significantly in response to short-term stimuli. As suggested by
the “lifetime learning model” (Mishler and Rose 1997), democratic orientations and behaviors may be
altered under the right conditions. The findings thus provide an interesting twist to Dalton’s assertion in
the East German context that “democratic norms are not learned through formal education and
indoctrination but through experience with the democratic process” (1994, 490). The analysis here
suggests that successful civic education in developing democracies may be viewed profitably as a
33
combination of “formal indoctrination” and direct political experience; that is civic education exposes
individuals to both curricular instruction and group-related mobilization appeals, and this combination of
influences appears to be highly capable of effecting substantial short-term change in the individual’s
propensity to engage the political system. Thus experience is critical to the development of democratic
attitudes and behavior, but this does not mean that it is necessarily outside the reach of the advocacy
groups that conduct training in democracy.
The results found here, in combination with those reported in previous work (Finkel 2000; Finkel et
al., 2000; Finkel and Stumbras 2000; Sabatini et al, 1998), lead to two additional conclusions. First, it
appears that the effects of civic education on local-level participation are larger in magnitude than those
seen for almost all other democratic orientations, including tolerance, trust, knowledge, efficacy, and
support for the rule of law. Further, at least in the short-term, these orientations themselves do not
necessarily appear to be powerful predictors of individual participation (see also Bratton 1999). This
suggests that changes in political participation may be easier to effect than changes in more general
democratic orientations; thus it may be the case that one promising route for the development of
democratic culture may be for groups to emphasize political mobilization in the hopes that individual
activity will stimulate the development of more durable democratic attitudes in the future (but see Bratton
et al. 1998). Such a process, in fact, may well describe the ways that democratic attitudes, values and
norms can develop in transitional systems with weakly formed democratic political cultures.
Second, the results suggest that the effectiveness of adult civic education in developing democracies
depends crucially on the advocacy NGOs and other groups that conduct the civic education training. The
strategy of funding such groups as a means of strengthening civil society and the democratization process
has been severely criticized, as it is often claimed that advocacy groups represent only a narrow part of
many countries’ emerging civil society, that they are “top-down” or “elite” associations that can only be
sustained through external funding, and that they are sometimes prone to corruption and mismanagement
(e.g. Carothers 1999; Ottaway and Chung 1999; Lasota 1999). The present study cannot speak to all of
these issues, yet the results here suggest that such groups may be highly effective mobilizing agents for
34
citizen participation precisely because they are often directly focused on that task. By contrast, the
behavioral cues for political participation that emanate from many other civil society groups are likely to
be much more muted. Moreover, the fact that advocacy NGOs draw many of the participants for civic
education training from existing civil society associations, and the fact that civic education appears to
have greater effects among individuals who are already members of other secondary groups, suggests that
advocacy groups are able to use existing civil society groups effectively to further their own aims. To this
extent the strategy of funding explicitly political organizations to achieve explicitly political goals seems
to make a good deal of sense.
At the same time, the results point to several limitations of civic education as a means of developing
democratic political culture. The results demonstrate that when individuals are trained frequently and are
trained with active, participatory methodologies, changes in participation can be of substantial magnitude.
But in most cases only a small portion of individuals who receive civic education instruction are exposed
to these beneficial pedagogical conditions. For example, only one-third of all civic education recipients
in South Africa attended three or more workshops, and less than half were trained with a large number of
active participatory teaching methodologies. The more intensive programs in the Dominican Republic
were more clearly designed to deliver frequent exposure, yet even in that context just over 40% of civic
education recipients were trained three times or more times. The results thus illuminate the difference
between the potential effect of civic education on political participation and the actual effects witnessed
in these democratizing contexts. Given the severe barriers to the implementation of civic education,
ranging from financial constraints, logistical difficulties in reaching potential respondents, and political
turmoil in certain areas, this limitation raises serious issues for policy makers regarding the feasibility of
utilizing civic education as part of a democratization program in many instances (Carothers 1999).
Finally, the interaction effects seen between civic education and the individual’s prior political
resources also serve as a reminder that civic education cannot by itself overcome the inequitable
distribution of politically-relevant resources that characterize many democratic settings. As the results
from South Africa make clear, it is not necessarily the case that civic education participants are drawn
35
from the elite and participatory strata in a given country. Advocacy groups often are able to bring
marginalized and previously inactive citizens into the civic education activities of the group. But, as is
the case in advanced democracies (Brady et al., 1999), the result of mobilization efforts by civic
education groups appears to reinforce many of the resource-based disparities in political participation in
developing contexts. Individuals require sufficient political resources and sufficient opportunities to
translate the mobilization appeals contained in civic education into concrete political action. Thus, civic
education can and does affect the political participation of resource-poor individuals, but the greater
effects seen among the resource-rich tends to exacerbate the existing “stratification of participation” in
developing democracies.
36
References
Almond, G. and Verba, S. 1963. The Civic Culture: Political Attitudes in Five Nations. Princeton
University Press, Princeton.
Barnow, Burt S., Glen G. Cain, and Arthur S. Goldberger. 1980. Issues in the Analysis of Selectivity
Bias. Pp. 43-59 in Stromsdorfer and Farkas (eds.), Evaluation Studies Review Annual, Volume 5.
Beverly Hills: Sage Publications.
Berk, Richard A., and Subhash C. Ray. 1982. Selection Biases in Sociological Data. Social Science
Research 11: 352-398.
Booth, John, and Patricia Richard. 1998. Civil Society, Social Capital and Democratization in Central
America. Journal of Politics 60: 780-801.
Brady, Henry E., Kay Lehman Schlozman and Sidney Verba. 1999. Prospecting for Participants:
Rational Expectations and the Recruitment of Political Activists. American Political Science
Review. 93:153-169.
Bratton, Michael. 1999. Political Participation in a New Democracy: Institutional Consideration from
Zambia. Comparative Political Studies 32:549-588.
Bratton, Michael and Philip Alderfer, with Georgia Bowser and Joseph Temba. 1999. The Effects of
Civic Education on Political Culture: Evidence from Zambia. World Development 27: 807-824.
Breen, Richard. 1996. Regression Models: Censored, Sample-Selected, or Truncated Data. Thousand
Oaks, CA.: Sage Publications.
Brehm, John, and Wendy Rahn. 1997. Individual-Level Evidence for the Causes and Consequences of
Social Capital. American Journal of Political Science 41: 999-1023.
Carothers, Thomas. 1999. Aiding Democracy Abroad: The Learning Curve. Washington, D.C.:
Carnegie Endowment for International Peace.
Carothers, Thomas. 1996. Assessing Democracy Assistance: The Case of Romania. Washington, D.C.:
Carnegie Endowment for International Peace.
Dalton, Russell. 1994. Communists and Democrats: Democratic Attitudes in the Two Germanies. British
Journal of Political Science 24: 469-493.
Diamond, Larry. 1995. Promoting Democracy in the 1990s: Actors and Instruments, Issues and
Imperatives. New York: Carnegie Corporation of New York.
Eckberg, Douglas Lee. 1988. The Physicians' Anti-Abortion Campaign and the Social Bases Of Moral
Reform Participation. Social Forces 67:378-398.
Eckstein, H. 1988. A Culturalist Theory of Political Change. American Political Science Review 82: 789804.
37
Espinal, R. 1996. The Dominican Republic: An Ambiguous Democracy. In J. Dominguez and A.
Lowenthal, eds., Constructing Democratic Governance: Mexico, Central America and the
Caribbean in the 1990s. Johns Hopkins University Press, Baltimore, MD.
Evans, G. and Whitefield, S. 1995. The Politics and Economics of Democratic Commitment: Support
for Democracy in Transitional Societies. British Journal of Political Science 25: 485-514.
Finkel, Steven E. 2000. Can Tolerance be Taught: Adult Civic Education and the Development of
Democratic Values." Paper prepared for delivery at the 2000 Annual Meeting of the American
Political Science Association, Washington, D. C., August 31-September 3, 2000.
http://pro.harvard.edu/abstracts/011/011031FinkelStev.htm (until August 2001).
Finkel, Steven E. and Karl-Dieter Opp. 1991. Party Identification and Participation in Collective
Political Action. Journal of Politics 53:339-371.
Finkel, Steven E., Christopher A. Sabatini, and Gwendolyn G. Bevis. 2000. Civic Education, Civil
Society, and Political Mistrust in a Developing Democracy: The Case of the Dominican Republic.
World Development, November, forthcoming.
Finkel, Steven E. and Sheryl Stumbras. 2000. Civic Education in South Africa: The Impact of Adult and
School Programs on Democratic Attitudes and Participation. Report Prepared for U.S.A.I.D.
Contract No. AEP-I-00-96- 90012-00, Task Order No. 10, Washington, D.C.: Management
Systems International, February 8, 2000.
Finkel, Steven E., Christopher A. Sabatini, and Gwendolyn G. Bevis. 1998. Civic Education and the
Development of Democratic Political Culture: Evidence from the Dominican Republic and Poland.
Paper presented at the Annual Meetings of the Midwest Political Science Association, Chicago,
Illinois.
Fishbein, Martin and Icek Ajzen. 1975. Belief, Attitude, Intention and Behavior: An Introduction to
Theory and Research. Reading, Massachusetts: Addison-Wesley Publishing Company.
Gibson, James L. 1998. A Sober Second Thought: An Experiment in Persuading Russians to Tolerate.
American Journal of Political Science 42: 819-850.
Gibson, James L., Raymond Duch, and Kent Tedin. 1992. Democratic Values and the Transformation of
the Soviet Union. Journal of Politics 54: 329-371.
Greene, William. 1993. Econometric Analysis. New York: Prentice Hall.
Hartlyn, J. 1998. The Struggle for Democratic Politics in the Dominican Republic. The University of
North Carolina Press, Chapel Hill, NC.
Heckman, James J. 1992. Selection Bias and Self-Selection. Pp. 201-224 in Eatwell, John, Murrary
Milgate, and Peter Newman (eds.), The New Palgrave Econometrics. London: MacMillan.
Heckman, James J. and Richard Robb. 1985. Alternative Methods for Evaluating the Impact of
Interventions. Pp. 156-245 in Heckman, James J., and Burton Singer (eds), Longitudinal Analysis of
Labor Market Data. Cambridge: Cambridge University Press.
38
Hess, R. D., and Judith V. Torney. 1967. The Development of Political Attitudes in Childhood. Chicago:
Aldine.
Huckfeldt, Robert and John Sprague. 1992. Political Parties and Electoral Mobilization: Political
Structure, Social Structure, and the Party Canvass. American Political Science Review 86:70-86.
Ichilov, Orit (editor). 1990. Political Socialization, Citizenship Education, and Democracy. New York:
Teachers College Press.
Inglehart, Ronald. 1990. Culture Shift. Princeton: Princeton University Press.
Jennings, M. Kent., and Richard Niemi. 1974. The Political Character of Adolescence. Princeton:
Princeton University Press.
Jennings, M. Kent, Jan Van Deth et al. (editors). 1990. Continuities in Political Action. New York: de
Gruyter.
Judd, Charles and David Kenny. 1981. Estimating the Effects of Social Interventions. New York:
Cambridge University Press.
Knoke, David. 1990. Networks of Political Action: Toward Theory Construction. Social Forces 68:
1041-1063.
Lasota, Irena. 1999. Sometimes Less is More (Response to Ottaway and Chung). Journal of Democracy
10: 125-128.
Leighley, Jan. 1996. Group Membership and the Mobilization of Political Participation. Journal of
Politics 58:447-463.
Lipset, Seymour M. 1959. Some Social Requisites for Democracy. American Political Science Review
53: 69-105
Mattes, Robert, and Hermann Thiel. 1998. Consolidation and Public Opinion in South Africa. Journal
of Democracy 9 (1): 95-110.
McAdam, Doug. 1982. Political Process and the Development of Black Insurgency, 1930-1970.
Chicago: University of Chicago Press.
McAdam, Doug and Ronnelle Paulsen. 1993. Specifying the Relationship between Social Ties and
Activism. American Journal of Sociology 99:640-667.
McDonough, Peter, Doh C. Shin and Jose Alvaro Moises. 1988. Democratization and Participation:
Comparing Spain, Brazil, and Korea. Journal of Politics 60:919-921.
Mishler, William, and Richard Rose. 1997. Trust, Distrust, and Skepticism: Popular Evaluations of Civil
and Political Institutions in Post-Communist Societies. Journal of Politics 59: 418-451.
Moffit, Robert. 1991. Program Evaluation with Nonexperimental Data. Evaluation Review 15: 291314.
39
Morduchowicz, Roxana, Edgardo Catterberg, Richard G. Niemi, and Frank Bell. 1996. Teaching
Political Information and Democratic Values in a New Democracy: An Argentine Experiment.
Comparative Politics 28: 465-476.
Niemi, R.G., and Junn, J. 1998. Civic Education: What Makes Students Learn. New Haven: Yale
University Press.
Ottaway, Marina and Theresa Chung. 1999. Toward a New Paradigm. Journal of Democracy 10:99113.
Petty, R. E., and J.T. Cacioppo. 1984. Source Factors and the Elaboration Likelihood Model of
Persuasion. Advances in Consumer Research 11: 668-672.
Pollack, Philip H. III. 1982. Organizations as Agents of Mobilization: How Does Group Activity Affect
Political Participation? American Journal of Political Science 26:485-503.
Putnam, Robert. 1993. Making Democracy Work. Princeton: Princeton University Press.
Quigley, Kevin F.F. 1997. Political scientists and assisting democracy: too tenuous links. PS: Political
Science & Politics 30: 564-568.
Reisinger, William., Arthur Miller, Vicki Hesli, and Kristen Maher. 1994. Political Values in Russia,
Ukraine and Lithuania: Sources and Implications for Democracy. British Journal of Political
Science, 24: 183-223.
Rose, Richard, William Mishler and Christian Haerpfer. 1998. Democracy and Its Alternatives:
Understanding Post-Communist Societies. Baltimore: The Johns Hopkins University Press.
Rose, Richard, and William Mishler. 1994. Mass Reaction to Regime Change in Eastern Europe:
Polarization or Leaders and Laggards? British Journal of Political Science 24: 159-82.
Rosenstone, Steven J. and John Mark Hansen. 1993. Mobilization, Participation and Democracy in
America. New York: MacMillan Publishing Company.
Sabatini, Christopher, Gwen Bevis and Steven Finkel. 1998. The Impact of Civic Education Programs
on Political Participation and Democratic Attitudes. Report Prepared for U.S.A.I.D. Contract No.
AEP-5468-I-00-6012-00, Delivery Order No. 5, Washington, D.C.: Management Systems
International, February 2, 1998.
Sears, David. 1975. Political Socialization. In Fred Greenstein and Nelson Polsby, eds. Micropolitical
Theory, volume 2 of The Handbook of Political Science. Reading, MA.: Addison-Wesley,
93154.
Slomczynski, K.M., and G. Shabad, (1998) “Can Support for Democracy and the Market Be Learned in
School? A Natural Experiment in Post-Communist Poland.” Political Psychology 19: 749-779.
Sniderman, Paul, J. F. Fletcher, P.E. Tetlock, and R. M. Boyd. 1996. The Clash of Rights. New Haven:
Yale University Press.
40
Snow, David A., E. Burke Rochford, Jr., Steven K. Worden, and Robert D. Benford. 1986. Frame
Alignment Processes, Micromobilization, and Movement Participation. American Sociological
Review 51:464-481.
Stromsdorfer, Ernst, and George Farkas. 1980. Introduction and Methodology. Pp. 13-41 in
Stromsdorfer and Farkas (eds.), Evaluation Studies Review Annual, Volume 5. Beverly Hills: Sage
Publications.
Torney, Judith, A.N. Oppenheim, and R.F. Farmer. 1975. Civic Education in Ten Countries. New York:
John Wiley.
Torney-Purta, Judith, John Schwille and JoAnn Amadeo. 1999. Civic Education Across Countries:
Twenty-Four National Case Studies from the IEA Civic Education Project. Amsterdam: IEA.
United States Agency for International Development (USAID). 1998. Democracy Strategic Framework,
Global Bureau for Democracy and Governance.
United States Agency for International Development (USAID). 1989. Final Evaluation of AID Grant to
Inter-American Institute of Human Rights. Lockett, Danuta; Kruvant, Charito, et al. USAID Bureau
for Latin American and the Carribbean, Office of Democratic Initiatives.
United States Agency for International Development (USAID). 1992. Evaluation of Voter Education
Programs in Chile, Inter-American Institute of Human Rights--Civitas: Final Report. Checchi and
Company Consulting. USAID Bureau for Latin American and the Carribbean, Office of Democratic
Initiatives.
Vella, Francis. 1998. Estimating Models with Sample Selection Bias: A Survey. The Journal of Human
Resources 33: 127-169.
Verba, Sidney and Norman Nie. 1972. Participation in America. New York: Harper and Row.
Verba, Sidney, Kay Lehman Schlozman and Henry Brady. 1995. Voice and Equality: Civic Voluntarism
in American Politics. Cambridge: Harvard University Press.
Walsh, Edward J. and Rex H. Warland. 1983. Social Movement Involvement in the Wake of a Nuclear
Accident: Activists and Free Riders in the Three Mile Island Area. American Sociological Review
48:764-781.
Weil, F. D. 1989. The Sources and Structure of Legitimation in Western Democracies: A Consolidated
Model Tested with Time-Series Data in Six Countries Since World War II. American Sociological
Review 54: 682-706.
Weil, F.D. 1993. The Development of Democratic Attitudes in Eastern and Western Germany in a
Comparative Perspective. Research on Democracy and Society 1: 195-225.
Winship, Christopher, and Stephen L. Morgan. 1999. The Estimation of Causal Effects from
Observational Data. Annual Review of Sociology 25: 659-706.
Zimbardo, Philip G. and Thomas Leippe. 1991. The Psychology of Attitude Change and Social
Influence. Philadelphia: Temple University Press.
41
Table 1
Participation by Civic Education Program
Dominican Republic and South Africa
DOMINICAN REPUBLIC
Number of Cases
Control
Group
1019
GAD
PC
ADOPEM
RSM
267
222
211
318
Organized Effort to Solve Community
Problem
33%
75%
64%
56%
62%
Attended Municipal Meeting
13%
53%
24%
19%
25%
Contacted Local Official
15%
54%
25%
16%
26%
Volunteered for Local Political Post (Town
Council, School Board, etc.)
12%
49%
19%
15%
19%
Overall Participation Scale (0-4)
ETA
SOUTH AFRICA
Number of Cases
.73
2.31
1.32
1.07
.41
Control
Group
475
LHR
CLC
219
99
NIPILAR
147
Organized Effort to Solve Community
Problem
38%
59%
40%
61%
Attended Local Council/Government Official
Meetings
30%
52%
21%
50%
Contacted Local Official
14%
29%
9%
34%
Worked for Party or Candidate in Election
Campaign
10%
27%
9%
29%
Overall Participation Scale (0-4)
ETA
.92
1.67
.30
Program Names, Dominican Republic:
GAD:
Grupo Accion por la Democracia
PC:
Participación Ciudadana
ADOPEM: Asociación Dominicana para el Desarrollo de la Mujer
RSM:
Radio Santa María
Program Names, South Africa
LHR:
Lawyers for Human Rights
CLC:
Community Law Centre – Durban
NIPILAR: National Institute for Public Interest Law and Research
42
.80
1.73
1.33
TABLE 2A
THE EFFECTS OF CIVIC EDUCATION ON POLITICAL PARTICIPATION,
DOMINICAN REPUBLIC
Model 3
Model 1
Model 2
Self-Selection
Self-Selection
With Intervening
OLS
Variables
B
s.e.
B
s.e.
B
s.e.
Civic Education Programs
GAD
.08
.08
.08
.73
.71
.64
RSM
.07
.08
.08
.31
.26
.31
PC
.03
.07
ADOPEM
.03
.08
Demographic Controls
Age
Age-Squared
Gender
Education
Income
Employed
Church
City
.01
-.01
.27
-.04
.03
.21
-.01
.10
.01
.01
.06
.02
.03
.06
.01
.05
.00
-.01
.27
-.06
-.01
.16
-.02
.15
.01
.01
.05
.02
.03
.06
.01
.05
.00
-.01
.19
-.09
-.02
.16
-.02
.14
.01
.01
.05
.02
.03
.06
.01
.05
Political Controls
Group Memberships
Political Interest
Media Use
Voted in 1996
2.17
.19
.24
.05
.19
.03
.03
.08
2.37
.22
.23
.23
.20
.03
.03
.08
2.26
.18
.18
.19
.19
.03
.03
.08
.12
.28
.05
.04
.01
.02
.06
.04
.03
.01
.07
-1.81
.04
.26
Democratic Orientations
Political Knowledge
Civic Skills
Political Efficacy
Political Tolerance
Trust in Institutions
Inverse Mills Ratio (Lambda)
Constant
-1.09
R-Squared
Number of Cases
.35
2037
.09
-1.48
.24
.35
2037
Coefficients in Bold: p<.05
Italicized Bold: p<.10
43
.04
.24
.37
2037
TABLE 2B
THE EFFECTS OF CIVIC EDUCATION ON POLITICAL PARTICIPATION,
SOUTH AFRICA
Model 3
Model 1
Model 2
Self-Selection
Self-Selection
With Intervening
OLS
Variables
B
s.e.
B
s.e.
B
s.e.
Civic Education Programs
LHR
.09
.13
.13
.40
.44
.33
NIP
.11
.14
.14
.47
.52
.45
CLC
-.04
.12
Demographic Controls
Age
Age-Squared
Gender
Education
Income
Employed
Race
Church
City
.26
-.03
.32
.15
.05
.17
.29
-.03
.16
.16
.03
.08
.03
.03
.08
.13
.03
.09
.26
-.03
.32
.15
.05
.16
.29
-.03
.16
.16
.03
.08
.03
.03
.08
.12
.03
.09
.18
-.02
.24
.06
.05
.12
.23
-.03
.14
.16
.03
.08
.03
.03
.08
.12
.03
.08
Political Controls
Group Memberships
Political Interest
Media Use
Voted in 1994
Voted in 1995
1.24
.28
.15
.21
-.00
.16
.06
.06
.12
.09
1.23
.27
.15
.21
-.01
.16
.06
.06
.12
.09
1.05
.16
.10
.18
-.06
.16
.06
.06
.12
.09
.23
.06
.20
.01
.03
.04
.08
.05
.04
.02
-.01
-1.77
.07
.30
Democratic Orientations
Political Knowledge
Civic Skills
Political Efficacy
Political Tolerance
Trust in Institutions
Inverse Mills Ratio (Lambda)
Constant
-2.09
R-Squared
Number of Cases
.34
940
-.03
-2.10
.28
.34
940
Coefficients in Bold: p<.05
Italicized Bold: p<.10
44
.07
.28
.38
940
Table 3
Predicted Increment in Participation for Different Levels of Civic Education Frequency and
Participatory Training Methodologies
(A) Dominican Republic
Frequency of Civic Education Exposure
Percent of
CE Sample
1–2
3–4
.54
.72
49%
More than 4
.94
1.12
51%
Percent of CE Sample
57%
43%
Less than 4
Participatory
Methodologies
(B) South Africa
Frequency of Civic Education Exposure
1–2
3–4
.65
.93
53%
More than 4
1.09
1.37
48%
Percent of CE Sample
67%
33%
Less than 4
Percent of
CE Sample
Participatory
Methodologies
Note: Estimates based on regression models including frequency of exposure and number of reported
participatory methodologies along with demographic and political controls from Table 1, Models 1 and 2.
As discussed in text, Dominican Republic analysis limited to GAD and Radio Santa Maria programs and
control groups (N=1562), and South Africa analysis limited to Lawyers for Human Rights, NIPILAR and
control groups (N=843).
45
Table 4
Interaction Effects of Civic Education Exposure and the
Individual’s Participatory Resources
Dominican Republic
B
s.e.
B
South Africa
s.e.
.17
.24
.32
2.79
-.31
.21
.40
2.79
-.03
1.24**
.19**
.21**
.02
.30
.04
.07
.16*
1.02
.18
.36*
.08
1.03
.19
.10
.31
.17
Civic Education
GAD (DominicanRepublic)
/LHR (South Africa)
RSM (Dominican Republic)
/NIPILAR (South Africa)
Participatory Resources
Education
Group Memberships
Political Interest
Gender
Race
Civic Education *
Resources
CE * Education
CE * Group Memberships
CE * Political Interest
CE * Gender
CE*Race
R-Squared
Number of Cases
-.08*
1.72**
.16*
.36**
.04
.45
.07
.12
.37
1532
*
.02
.26*
.39**
-.02
-.10
.36
841
**
p<.05
p<.01
*
Note: Control variables in Table 1, model 2 included in the analysis but not shown.
46
.05
.15
.14
.16
.23
Appendix
Measurement of Control Variables and Sample Information
Democratic Orientations
Political Knowledge. In the Dominican Republic, I added the number of correct answers out of four
questions concerning knowledge of incumbents and the electoral process (“when are the next presidential
elections,” who is the mayor of this municipality”). In South Africa I added the number of correct
answers out of four questions concerning knowledge of the institutional structure of the government
(“how long is the President’s term in office,” “who has the power to decide whether a law is
constitutional”). The reliability of the scale was .66 in the Dominican Republic, .76 in South Africa.
Civic Skills. The questionnaire contains four questions asking the respondent to rate the extent to which
“in general, you communicate your ideas with others (cooperate with other people, speak in public, know
better whom to contact to get things done, direct a group) better, worse, or the same as most other people
you know?” Answers of “better than” were scored as two; an answer of “same as” as one; and “worse
than others” as zero. These scores were then added and divided by four to create a scale from 0 to 2.
The reliability of the scale was .75 in the Dominican Republic, .76 in South Africa.
Political Efficacy. Respondents in the Dominican Republic were asked three questions on efficacy, all
asking the respondent to agree or disagree on a four-point scale to a series of questions concerning their
views of their influence on the political system. The three questions were: (1) Sometimes politics and
government are so complicated that people like me can’t understand what is going on; (2) People like me
have no say in what the government does; and (3) There is no point in getting involved in politics because
I would have no influence anyway. In South Africa, the first two questions were supplemented with: (1) I
feel well prepared for participating in political life; and (2) If I wanted to discuss my political views, I
would know where and how to contact elected officials. The reliaability of the scale was .68 in the
Dominican Republic, .63 in South Africa.
Political Tolerance. Respondents were asked nine standard questions to test willingness to extend
freedoms of association, participation and speech to three unpopular groups in each country: atheists,
racists (“persons who believe blacks to be genetically inferior”) and sexists (“persons who believe women
to be genetically inferior”) in South Africa, atheists, militarists, and communists in the Dominican
Republic. For each group, it was asked whether such a person should be allowed to speak publicly in
your locality; should be allowed to vote, and should be allowed to organize peaceful demonstrations to
express his/her point of view. Answers are on a four-point agree/disagree scale. It was found in both the
South Africa survey that the questions regarding voting tolerance were distinct in respondents’ minds
from tolerance for speaking in public and organizing a peaceful demonstration. I created a tolerance
variable for the six non-voting questions (2 questions for the 3 unpopular groups) in South Africa by
averaging the six scores; in the Dominican Republic the scale contained all nine items. The reliaability of
the scale was .78 in the Dominican Republic, .87 in South Africa.
Trust in Institutions. Respondents were asked how much they trusted a series of thirteen political and
social institutions. I created a 0-6 sub-scale for Political Trust by focusing on the six institutions that
clearly relate to institutions of the South African government (President, provincial legislature, local
council, constitutional court, national government, African National Congress). In the Dominican
Republic, the scale consisted of a 0-7 count of the number of the following institutions in which the
individual reports Amuch trust:@ Judicial System, Police, Army, Parliament, Local Officials, Political
System, and the Business Community. The reliability of the scale was .85 in the Dominican Republic,
.83 in South Africa.
47
Demographic and other Control Variables
Education. 7 categories ranging from no education to profession training beyond a university degree.
Age. 5 categories, grouped from 18-24, 25-34,35-49,50-64, and 65 and older.
Income. 7 grouped categories in the Dominican Republic, 15 grouped categories in South Africa.
Church Attendance. Frequency of attending religious services on a 0-5 scale, 0 being “no religion” to “5”
for “every day”.
Dummy variables were created to signify: gender, race, urban residence, employment status
Political Interest. In the Dominican Republic, I averaged two questions measuring local and national
political interest on a four-point scale from “none” to “much” interest. In South Africa, a question on
interest in the “current election campaign” was added, and in that country the response categories ranged
from “very little” to “a great deal” of interest on a three-point scale.
Media Exposure. In the Dominican Republic, I averaged two questions on a four-point scale measuring
the respondent’s attentiveness to information about politics on “television and radio” and then in
“newspapers” (“never” to “every day or almost every day”). In South Africa the radio and television
questions were separated, resulting in a three-question scale for attentiveness, with each measured on the
same four-point scale as in the Dominican Republic.
Previous Voting Behavior. Respondents were asked if they voted in the previous Presidential election
(1996 in the Dominican Republic, 1994 in South Africa). South African respondents were also asked if
they voted in the 1995 local community elections.
Group Memberships. Respondents were asked if they were members of a series of voluntary associations
in each country. In the Dominican Republic, the groups were: parent’s association, peasant’s association,
women’s association, union, church group, neighborhood or community group, occupational or
professional association, cultural organization, sports or hobby club, or cooperative group. In South
Africa, the groups were: community or neighborhood group, church group, youth group, sports or hobby
club, union, women’s group, or cultural group. In each country the number of groups reported was
divided by the number of groups asked about; thus the variable measures the proportion of groups asked
about to which the respondent claims membership. The reliability of the scale is .57 in the Dominican
Republic, .70 in South Africa.
48
Table A-1
Characteristics of Civic Education and Control Group Samples,
Dominican Republic
Participación
Ciudadana
LOCATION
National
Sampling
method
N
La Vega
Sampling
method
random,
from lists
250
Civic Education Program
GAD
ADOPEM Radio Santa
Radio Santa
Maria
Maria (indirect)
(direct)13
random,
from lists14
247
N
San Pedro de
Macoris
Sampling
method
N
San Cristobal
Sampling
method
N
Herrera
Sampling
method
N
Sabana Perdida
Sampling
method
N
TOTALS
Control
Group
random
stratified
695
random,
from lists
random,
from lists
20115
152
snowball from
RSM-Direct
participants
153
189
random
50
random
50
random
50
random
250
247
201
152
153
13
50
2084
The Radio Santa Maria project was a training of trainers activity. The implementing organization only
maintained lists of the "direct" participants or trainers that it trained. After interviewing the "direct" participants, the
interviewers would ask for names of "indirect" participants they had trained who would then be interviewed. This
second set constitutes the "indirect" respondents.
14
The national sample for the GAD program excluded Santo Domingo.
15
The total N of 201 for the ADOPEM program was drawn from San Pedro de Macoris,
San Cristobal, Herrera, and Sabana Perdida as well as from La Vega.
49
Table A-2
Characteristics of Civic Education and Matched Control Group Samples, South Africa
PROVINCE
Civic Education Program
CLC-Durban NIPILAR
LHR
Total N
100
150
225
Control
Group
475
14
11
25
10
8
7
14
11
25
10
8
7
30
55
5
AREA
Eastern Cape
Mt. Fletcher
Qumbu
Free State
Bethlehem
Bloemfontein
Botshabelo
Phuthaditjaba/Witsieshoek
Gauteng
30
30
Bronkhorstspruit
Johannesburg
KwaMhlanga/Bronkhorstsp
ruit
Pretoria
Soweto
Vaal Area
West Rand (Johannesburg)
20
Mpumalanga
Northern Province
Cornfields
Durban
Eshowe
Muden
Pinetown
River View
Sankotshe
20
5
10
30
KwaZulu-Natal
25
5
5
10
25
25
13
25
12
25
25
100
Bosbokrand
Dennilton
Graskop
Leandra
Siyabuswa
25
10
15
15
30
10
10
Lebowakgomo
Nebo
Nebo/Sekhukhune
Pietersburg
Warmbaths
5
10
25
8
8
10
30
50
9
25
5
10
5
10
55
25
13
25
12
25
25
125
10
15
15
5
10
55
18
10
8
10
9
55