Do Service Provision NGOs Perform Civil Society Functions? Evidence of NGOs’ Relationship With Democratic Participation

Abstract

Scholars have long been interested in the relationship between civil society and democracy. Today, international donors promote civil society in developing countries, and they explicitly define civil society to include nongovernmental organizations (NGOs). It remains an open question, however, whether service provision NGOs in developing countries fulfill this civil society function. Some prominent scholarship argues that only NGOs that explicitly pursue advocacy perform civil society roles; service provision NGOs do not. I address this question, testing the relationship between individuals’ experience with, and perceptions of, NGOs and their political participation in voting, protests, and raising issues with public officials. Using data from Kenya, I find that individuals who report contact with, or positive views of, NGOs are more likely to engage in certain political activities, providing some evidence for the inclusion of NGOs in the concept of civil society. I theorize mechanisms by which service provision NGOs may affect various forms of participation.

Keywords

NGOs civil society service provision democratic participation protest Kenya

Introduction

Since Tocqueville (1835) described the role that voluntary citizen groups played in shaping American democratic society, scholars have been interested in the relationship between civil society and democracy. Studies from advanced industrialized countries show that individual participation in membership organizations creates social capital needed to monitor government performance, and that higher density of nonprofits at the community level is associated with a more vibrant democracy (Almond & Verba, 1963; Putnam, 1993). Drawing considerably on this research in developed countries, international donors now emphasize the need for civil society—often operationalized to mean nongovernmental organizations (NGOs)—as a crucial factor for the social, economic, and political development in the Global South. Representative of this stance, U.S. Ambassador Mark Palmer boldly asserted that “NGOs have assisted a massive expansion in freedom” (Senate Foreign Relations Committee, 2006) while former World Bank President Robert Zoellick (2011) declared that “A robust civil society can check on budgets, seek and publish information, challenge stifling bureaucracies, protect private property, and monitor service delivery.” Supporting this rhetoric, the World Bank deliberately increased the percentage of programs involving civil society organizations (CSOs), including NGOs, from 20% to 88% between 1990 and 2015 (World Bank, 2018).

Yet many prominent scholars remain skeptical of including NGOs, especially service-providing ones, in an understanding of civil society. Some have argued for decades that NGOs are unlikely to affect political change in developing countries because they “have failed to develop effective strategies to promote democratization,” especially individual-level citizenship practices (Banks et al., 2015; Edwards & Hulme, 1996, p. 965), or because their dependence on donors has resulted in misaligned priorities (Hearn, 2007) that stifle their advocacy efforts by shifting focus to service provision (Rahman, 2006) as they become more professionalized and “modern” (Chahim & Prakash, 2014).

Although these skeptical views exist, very few quantitative studies have actually tested whether political transformative effects are associated with NGOs in developing countries (exceptions include Booth & Richard, 1998; Boulding & Gibson, 2009; Brown et al., 2007). Here, I add to this limited scholarship, using original survey data from Kenya, to analyze the relationship between individual-level interaction with or views of service-providing NGOs and political behavior, including voting in national elections, engaging a local politician or administrator in a civic discussion, and protesting. Examining service-provision NGOs is especially important because it is often argued that these NGOs underperform their civil society function, even explicitly divorcing themselves from it (Banks et al., 2015). This makes them a “hard case”—one for which results are least expected.

I find suggestive evidence of a relationship between respondents’ reports of direct contact with NGOs and their political behavior. Individuals who report having had contact with NGOs—whether of their own volition or not—are more likely to protest and to raise issues with government officials locally, controlling for a number of factors. There is no consistently significant relationship between NGOs and voting behavior, which aligns with findings from the Americas (Boulding, 2010). My findings add validity to the promotion of NGOs as civil society actors.

Beyond testing this relationship, I contribute to theory-building in this article by proposing mechanisms by which service provision NGOs affect political participation. I not only elaborate why NGOs may affect participation generally, but also specifically examine why effects of NGOs may vary by type of participation.

Theorizing Civil Society, NGOs, and Democracy

For nearly 200 years, civil society—the sphere of organized activity between the family and the state—has been identified as a central feature of effective democratic governance (Almond & Verba, 1963; Putnam, 1993; and Tocqueville, 1835 are classic citations, but see also Paxton, 2002, pp. 257–260, and Skocpol and Fiorina, 1999 for excellent reviews). Putnam (1993), for example, demonstrates that areas of Italy with higher density of community organizations—soccer clubs, choral societies, women’s groups, and bowling leagues, among others—have more effective governments. He describes a link between a local “communitarian” position favoring collaboration to solve collective action problems, and “republican” positions that support democratic forms of government (Putnam, 1993, p. 87). Other scholars note that trust is central to the relationship between civil society and policy activism (Inglehart, 1997; Tavits, 2006). As such, decreases in social capital undermine the civic engagement necessary for strong democracy (Putnam, 2000; Skocpol, 2003).

Scholars have tested the “civil society hypothesis” in countries around the world and found that the theory “travels” (Baggetta, 2009a; Booth & Richard, 1998; Klesner, 2007; Lee, 2010; LeRoux & Krawczyk, 2014; Marsh, 2002; Menon & Daftary, 2011; Paxton, 2002; Sanyal, 2009; Seligson, 1999).¹ In Kenya, civil society—churches and religious groups, professional organizations and advocacy groups—has been credited with the arrival of multiparty elections in the 1990s, the opposition coalescence before the 2002 elections (Kanyinga, 2004), and the relatively quick resolution to the 2008 postelection crisis (Kanyinga, 2011).

Such ideas led scholars and practitioners to look to civil society as a locus of hope for deepening democracy in sub-Saharan Africa (Diamond, 1997; Gyimah-Boadi, 1996; Harbeson, 1994).² Donor agencies like the World Bank, United Nations Development Program (UNDP) and U.S. Agency for International Development (USAID) joined in, including NGOs in their notion of civil society. The World Bank was particularly influenced by Putnam’s (1993) ideas, which he was invited repeatedly to present to groups of senior staff (Bebbington et al., 2004). Donor funding to strengthen civil society has been considerable: In 1 year alone, the U.S. State Department and USAID spent approximately US$647 million on civil society programming (Tarnoff & Lawson, 2011), while The World Bank spent US$2.1 billion on community-driven development (World Bank, 2009).

Early scholarship focused on NGOs took it as a given that NGOs served a civil society role (Bebbington, 1997; Clark, 1995; Heinrich, 2001). They witnessed Polish NGOs play a significant role in the country’s democratic transition by protecting human rights, changing public attitudes, and promoting the rule of law (Gotchev, 1998). Likewise, NGO civic education and paralegal programs in Kenya had a measurable effect on citizen understandings of politics (Orvis, 2003). In the past decade, scholars have examined the relationship between specific NGO programs and democracy and governance outcomes (Moehler, 2010), finding increases in collective action capacity (Fearon et al., 2015). Brass (2012) argues that interactions with, and learning from, NGO practices have increased democratic norms within the civil service in Kenya. In Latin America, Boulding similarly shows that NGOs can play a powerful role in stimulating protesting and activism toward electoral challenges (Boulding, 2010, 2014; Boulding & Gibson, 2009), and Nelson-Nuñez (2018) finds that greater levels of NGOs can facilitate citizen engagement with government. Murdie and Bhasin (2011) likewise find that some NGO activities are associated with increased protesting.

Nevertheless, there remains vocal skepticism about whether NGOs should be included in the category of CSOs (Hearn, 2000; Mercer, 2002)—especially service-provision NGOs. Banks et al. (2015, p. 709) argue that while NGOs excel in service delivery, they do so “at the expense of their civil society function,” the exercise of which potentially threatens relationships with their donors and host governments (cf., Chahim & Prakash, 2014; Hearn, 2007; Holmen & Jirstrom, 2009). Supporting this notion, NGOs in South Asia have deliberately behaved apolitically, with scholars asserting that this avoidance erodes democratic values, especially among the poor (Rahman, 2006; Ulvila & Hossain 2002). Ulvila and Hossain (2002) go further, asserting that NGOs are elite tools used to maintain power. Furthermore, Lewis (2004) argues that patronage relationships align nominally “civic” organizations with the state and state interests, undermining their democratizing potential. In Jordan, moreover, even when NGOs attempt to engage in civic activities, they are stymied by deliberate state interventions to limit their effects (Wiktorowicz, 2002). In addition, sometimes, participation reduction is unintentional: Nelson-Nuñez (2018) finds that when NGO provision results in sufficient service coverage, citizens petition the government less.

Despite the skeptics, I hypothesize that there exists a positive association between service provision NGO activity and democratic behavior, mirroring the link between civic groups and democracy. Arguably, NGOs positively affect political participation not only through activities like direct civic education and rule of law promotion, but also through several mechanisms related to service provision. I review these mechanisms at length in the “Discussion” section of this article but only summarize them here as data limitations prevent testing them in this article. First, for decades, NGOs have emphasized active “participatory development,” meaning community engagement in decision-making in their programs (Chambers, 1983; Korten, 1980; Uphoff et al., 1998). Inclusion in collective decision-making is meant to empower the individuals participating (Carr, 2011; Hershey, 2019), who can use the skills they practice with the NGO as they participate in the broader public realm (Dodge & Ospina, 2016). Working together is also meant to build social capital, which facilitates individuals to overcome collective action problems around some forms of participation, like voting (Dupuy & Prakash, 2020). Second, service provision NGOs bring a range of information, finance, education, and other resources that can foster individual and collective political participation by making them easier to achieve (Boulding, 2010; Murdie & Bhasin, 2011; Nelson-Nuñez, 2018). Third, NGO staff often work collaboratively with government staff and invite government officials to their informal meetings and formal events (Abers & Keck, 2009; Brautigam & Segarra, 2007; Nurul Alam, 2011). In doing so, they can make it easier for citizens, who may also be at these events, to engage politicians and civil servants. Finally, the act of service provision by nonstate actors may spur protest or contacting, in that it may provide new hope to people that change is possible (Brass, 2016). Or a partial provision of services may lead non-recipients to demand complete services from government (Hern, 2017).

Data From Kenya

Data for this article come from an original survey of about 500 individuals at the household level in three areas of Kenya. The survey measured respondents’ subjective assessment of their recent experiences with NGOs; their perceptions of NGOs, community-based organizations (CBOs), and the government; a number of political behaviors; and a variety of demographic and socioeconomic characteristics. The sample includes responses from middle-class areas of two cities, the capital, Nairobi, and Machakos Town, and from rural areas of the district of Machakos and Mbeere. In Nairobi, respondents were selected from long-standing communities comprising, largely, educated and formal-sector employed—but not elite—Kenyans. Another 100 respondents lived in Machakos Town (population 150,000) and represent urban viewpoints outside of the capital. Representative samples of the full Machakos and Mbeere district were also completed. Because some survey respondents were below voting age at the time of the election discussed, which occurred nearly a year before the survey, only respondents 19 years or older are included in the analyses of voting patterns.

Kenya is a prime case study for this sort of research as the country falls near the middle of less-developed countries internationally: Kenya ranks 147th out of 189 countries in the Human Development Index (UNDP, 2019), putting it near the bottom of the middle human development category. Likewise, Freedom House considers Kenya “partly free” in its annual Freedom in the World index, placing it squarely in the middle of the range of countries (Freedom House, 2018). Unlike some extremely poor or weak African countries, moreover, Kenya has developed manufacturing, technology, and export sectors, making it comparable with many of the countries of Latin America and Asia.

Kenya is also well-suited due to its lively and long-standing CSOs, including NGOs.³ Community-based CSOs have a long history in Kenya. The country’s motto, Harambee, means “let’s pull together” and is a reference to ubiquitous local self-help organizing that has occurred since precolonial times. After independence, harambee groups collaborated with government offices to contribute to service provision, especially in schools and clinics. A number of more formal, usually larger, and often international NGOs grew rapidly in the 1990s and 2000s, largely in response to economic hardship and state downsizing during neoliberal austerity measures. Although staffed almost entirely by Kenyans, the overwhelming majority of their funding comes from international sources (Brass, 2016). These organizations primarily provide services related to poverty reduction in Kenya (see Figure 1)—only 4% of organizations at the time of research were “democracy and governance” organizations (Brass, 2016; NGOs Co-ordination Board, 2009). These NGOs are well-regarded by Kenyans. When surveyed about whether various organizations have the interests of the people in mind, more Kenyans (70%) responded positively about NGOs than about civil servants (53%) or politicians (34%; Brass, 2016). In the same survey, 82% of respondents who expressed an opinion agreed with the statement, “NGOs in general share my values and do the right thing.”

Figure 1.

Distribution of NGO activities by primary sector of engagement: Author coding.

NGO and Participation Measurement

Three measures of democratic participation, the dependent variable, are used in the analyses. Respondents were told that these “activities represent actions people sometimes take as citizens” before being asked about whether they had done them, making the link to democratic activity explicit for the respondents. The first measure examines whether respondents voted in the national election that occurred less than a year prior to the survey implementation. Of the 476 survey respondents, aged 19 years or older, 87% had voted. The second examines how often the respondent has joined with others to raise an issue with a local politician or administrator in the past 1 year as raising an issue with a government official is a common civic-minded act. Approximately, a quarter of the respondents reported that they had done so. The third examines respondents’ experience with demonstrations or protests, an even less-common civic activity, within the past year. Fifty-nine respondents, or 12%, reported participation in a protest or demonstration. (Two people, representing less than one half of 1% of respondents, did not respond.) Unlike some settings, protests in sub-Saharan Africa tend to be “valence protests,” meaning that they are aimed at communicating preferences on key issues to government in situations where voting is insufficient—the goal is not regime or government change (Harris & Hern, 2019). This type of protest can be compatible with democratic expression where there exists a minimum level of institutional support for democracy (Boulding, 2014). In Kenya, specifically, protests are not necessarily high risk as there are fairly strong guarantees of freedom of expression and assembly, and Kenya holds regular elections. Table 1 presents descriptive statistics for all variables.

Table 1.

Descriptive Statistics for Variables Used in Regressions.

Variable	Obs.	M	SD	Minimum	Maximum
Dependent variables
Has protested	499	0.12	0.32	0	1
Has raised issues	499	0.24	0.43	0	1
Voted in prior year election	476	0.87	0.33	0	1
Independent variables
NGOs in area	439	0.39	0.49	0	1
Gone to NGO once	498	0.15	0.36	0	1
Gone to NGO 2+ times	498	0.09	0.29	0	1
NGO has come once	469	0.15	0.36	0	1
NGO has come 2+ times	469	0.12	0.33	0	1
Controls
Education	498	4.39	1.68	1	8
Gender (male = 0, female = 1)	501	0.55	0.5	0	1
Age	501	34	13	18	99
District	501	N/A	N/A	N/A	N/A

Independent variables come from the same survey and include measures of respondents’ subjective assessment of their contact with and perceptions of NGOs. Two variables measure whether the respondent recalled, in the past year, either deliberately seeking out an NGO, or having had an NGO approach their community to provide goods or services—and if so, the number of times. Specifically, the questions were as follows: “In the past year, how many times have you gone to an NGO seeking training, information, a service or for a physical good?” and “In the past year, how many times do you know of any NGO coming to this community to offer training, information, a service or for a physical good?” A considerable number of people had these experiences: 24% of respondents reported having sought out an NGO, and 27% reported having had their community approached by one. Only three respondents did not know the answer to the question about going to an NGO. Thirty-two respondents did not know the answer to the question about whether NGOs had approached their community.

Should NGOs play a civil society function, there should be a stronger relationship among individuals who report contact with NGOs multiple times than those who report experiencing NGOs only once. Thus, analyses in this article distinguish between a respondent recalling having gone to or having been approached only once (“Gone to NGO once” and “NGO has come once”) or 2 or more times (“Gone to NGO 2+ times” and “NGO has come 2+ times”). A third measure of NGOs assesses whether the respondent perceived there to be NGOs working in the place where they lived or not. For this question, initial responses included “very many,” “some,” “very few,” and “none,” which were dichotomized to a value of 1 if the respondent perceived there to be any NGOs in the area.

Because scholars have noted that individuals who proactively seek experiences in general tend to also seek political experiences specifically, the latter two measures are important checks on the measure about seeking out NGOs because they do not face the issue of “joiners” (Almond & Verba, 1963; Verba et al., 1995). That said, individuals who responded positively to having sought out NGOs were only statistically different in minor ways than those who had never gone to an NGO. On average, NGO seekers were slightly wealthier (in the second-to lowest instead of the lowest of six wealth categories) and slightly more educated (having finished more secondary school than their counterparts)—with the note that these two variables are fairly highly correlated—and were more likely to be male. But they were no different on age, urban/rural location, ethnicity, views about the current state of the economy, or views on CBOs. Likewise, NGO seeking respondents were no more likely to be registered voters, to correctly name their member of parliament, to trust the government of Kenya generally or the president specifically, or to have used force or violence for a political cause.

Analyses control for respondents’ education level, location, gender, and age. Analyses also included measure of housing type and radio use. They are not reported because neither covariate had a nonzero coefficient at even 50% significance level. This yields five covariates: education level (which proxies for social class due to multicollinearity), gender, age, and two dummy variables indicating the respondent’s district (Machakos or Mbeere, or if each dummy is zero, Nairobi). None of the variables of interest are highly correlated with any of the control variables.

Analysis

I use logistic regression to test the association between individuals’ reported experiences of NGOs and their likelihood of political participation through voting, through community action, and through protest. If NGOs providing services represent civil society actors, we expect to find a positive correlation between contact with NGOs and each of these three behaviors. All regressions use the following model, with standard errors clustered by the sampling unit, as the survey employed clustered sampling, with no more than 10 individuals selected for any cluster:

\begin{array}{l} P r (d e m o c r a t i c a c t i v i t y_{i} = 1) \\ = f [\begin{array}{l} β_{0} + β_{1} (N G O a c t i v i t y_{i}) + β_{2} (e d u c a t i o n_{i}) \\ + β_{2} (g e n d e r_{i}) + β_{2} (a g e_{i}) + β_{2} (d i s t r i c t_{i}) + u_{i} \end{array}] \end{array}

Independent variables of interest are included in the models one at a time because they are fairly highly correlated and because they all measure the underlying question regarding NGOs as civil society actors. The strategy of employing them one at a time in the model thus affords us a robustness check for our results.

Findings reported in Tables 2 to 4 suggest that NGO activity is associated with democratic participation in the ways expected by civil society theory, and that even service provision NGOs can fulfill civil society roles. In all models, statistically significant NGO measures are positively associated with democratic participation measures. Summarizing, the results reveal the following: There are significant, positive correlations between an individual reporting having gone to an NGO, or having an NGO visit, and the likelihood of that individual protesting. Likewise, there are significant, positive correlations between having gone to an NGO, or perceiving there to be NGOs in the area, and raising issues with government officers, all else held constant. There are no significant correlations, however, between individuals’ interactions with or perceptions of NGOs and voting.⁴

Table 2.

Effects on Likelihood of Participation in Demonstrations or Protests.

Variable	Model 1	Model 2	Model 3
Gone to NGO once	−0.0104 (–0.02)
Gone to NGO 2+ times	1.005** (2.01)
NGO has come once		1.135** (2.29)
NGO has come 2+ times		0.998* (1.74)
NGOs in area			0.681(1.60)
Education	−0.0468(–0.33)	−0.101(–0.63)	−0.0700(–0.46)
Gender	−1.261*** (–4.01)	−1.299*** (–3.61)	−1.119** (–2.88)
Age	0.0148(1.44)	0.0177(1.48)	0.0186** (2.05)
Machakos	−1.213** (–2.47)	−1.357** (–2.33)	−1.325** (–2.47)
Mbeere	0.759(1.13)	0.691(0.97)	0.493(0.71)
Constant	−1.758** (–2.03)	−1.753* (–1.89)	−1.892** (–2.16)
N	495	466	437
Pseudo R²	.144	.169	.132

Note. Coefficients are log-odds from logistic regression analysis. T-statistics are in parentheses.

p < .1. **p < .05. ***p < .001.

Table 3.

Effects on Likelihood of Raising an Issue With a Government Official.

Variable	Model 1	Model 2	Model 3
Gone to NGO once	0.256(0.89)
Gone to NGO 2+ times	0.733** (2.04)
NGO has come once		0.491(1.42)
NGO has come 2+ times		0.610(1.45)
NGOs in area			0.561* (1.95)
Education	−0.148(–1.61)	−0.150(–1.55)	−0.135(–1.31)
Gender	−0.772** (–3.18)	−0.832** (–3.20)	−0.653** (–2.52)
Age	0.0111(1.52)	0.0142* (1.79)	0.0135(1.56)
Machakos	−0.575(–1.19)	−0.609(–1.19)	−0.523(–1.02)
Mbeere	0.593(1.05)	0.556(0.94)	0.564(0.97)
Constant	−0.536(–0.68)	−0.622(–0.77)	−0.839(–0.94)
N	495	466	437
Pseudo R²	.084	.087	.072

Note. Coefficients are log-odds from logistic regression analysis. T-statistics are in parentheses.

p < .1. **p < .05. ***p < .001.

Table 4.

Effect on Likelihood of Having Voted in the Recent Presidential Election.

Variable	Model 1	Model 2	Model 3
Gone to NGO once	−0.0564(–0.14)
Gone to NGO 2+ times	0.522(0.90)
NGO has come once		0.657(1.18)
NGO has come 2+ times		−0.472(–1.08)
NGOs in area			−0.304(–0.99)
Education	0.0444(0.45)	−0.00944(–0.10)	0.0404(0.43)
Gender	−0.824** (–2.43)	−0.822** (–2.34)	−1.003** (–2.93)
Machakos	0.182(0.52)	0.332(0.96)	0.545(1.37)
Mbeere	0.222(0.57)	0.327(0.82)	0.543(1.16)
Age	0.0394** (2.05)	0.0350* (1.73)	0.0363* (1.76)
Constant	0.806(0.91)	1.136(1.24)	0.907(0.93)
N	470	442	413
Pseudo R²	.054	.058	.062

Note. Coefficients are log-odds from logistic regression analysis. T-statistics are in parentheses.

p < .1. **p < .05. ***p < .001.

Looking with more detail into these findings, Model 1 across the tables includes respondents’ recollection that they went to an NGO seeking training, services, or for a physical good. This model shows a significant correlation between going to an NGO more than once and civic participation in both protest and in raising issues with government officials, whereas going to an NGO only once does not show any significant effect on any of our dependent variables. There is no evidence for a link between going to an NGO and voting, whether the respondents went to an NGO once or more than once.

Although this model provides some evidence to support the notion that NGOs act as civil society agents, I acknowledge that the design of this variable may raise questions. Specifically, studies have shown than people who proactively seek social activities also tend to participate in political activities (Almond & Verba, 1963; Verba et al., 1995). It is plausible that this correlation holds true in the Kenyan context—those who report to have sought NGOs for training, services, or for a physical good may simply be seekers or “joiners” (Baggetta, 2009b). If this were the case, this unobserved factor could explain the correlation.

Although there is not a variable in this survey that allows us to control for whether or not an individual is a “joiner,” Models 2 and 3 across Tables 2 to 4 attempt to correct for this issue. Here, the variables measure whether (and, if so, how many times) a respondent’s community has been approached by an NGO or whether the respondent perceives there to be NGOs in the area. Because respondents were quite unlikely to have had influence over whether or not NGOs came to their area, these variables do not face the “joiners” issue. In addition, we have less reason to question the direction of causality with the variable measuring NGOs coming to a community: There is no theoretical reason to believe that politically active individuals are more likely to be approached by NGOs. Applying this model shows a significant correlation between a respondent reporting having been approached by an NGO and the likelihood of protest, whether the NGO approached the respondent once or more than once. There is no statistically significant relationship between this variable and raising an issue or voting although the positive coefficient results may be substantively meaningful.⁵ There is a significant correlation between perceptions of NGOs in the area and raising an issue with government officials.

Interpretation of Results

As log-odds are nonintuitive for most readers, I used Clarify (Tomz et al., 2003) to generate predicted probabilities that facilitate interpretation. Tables 5 and 6 show the predicted probability of a respondent having protested, or raised an issue with the government, as associated with the various measures of NGOs. The tables report nonsignificant results because the direction of the result is generally consistent across all models and power is limited by the relatively small sample size. Results reflect the mean values for all other variables in the model, meaning they represent an average-aged (34-year-old) male resident of average education level in each of the three districts. I report males for these tables as they are significantly more likely to engage in both of these activities than are females.

Table 5.

Predicted Probabilities of Having Protested.

Variable	Range of change	District	Predicted probability of protesting when X = 0	Predicted probability of protesting when X = 1	Change in the predicted probability of protesting
Gone to NGOs	From never to 2 or more times	Machakos	0.07	0.17	0.10**
		Mbeere	0.34	0.56	0.22**
		Nairobi	0.2	0.4	0.20**
NGOs came to area	From never to 2 or more times	Machakos	0.05	0.33	0.28**
		Mbeere	0.31	0.75	0.44**
		Nairobi	0.17	0.61	0.44**
NGOs in area	From unaware to aware of NGOs	Machakos	0.06	0.1	0.04
		Mbeere	0.27	0.41	0.14
		Nairobi	0.18	0.3	0.12

p < .1. **p < .05. ***p < .001.

Table 6.

Predicted Probabilities of Having Raised an Issue With a Government Official.

Variable	Range of change	District	Predicted probability of protesting when X = 0	Predicted probability of protesting when X = 1	Change in the predicted probability of protesting
Gone to NGOs	From never to 2 or more times	Machakos	0.21	0.41	0.20**
		Mbeere	0.45	0.67	0.22**
		Nairobi	0.32	0.54	0.22**
NGOs came to area	From never to 2 or more times	Machakos	0.2	0.43	0.23
		Mbeere	0.45	0.69	0.24
		Nairobi	0.32	0.57	0.25
NGOs in area	From unaware to aware of NGOs	Machakos	0.19	0.28	0.09*
		Mbeere	0.4	0.54	0.14*
		Nairobi	0.29	0.41	0.12*

p < .1. **p < .05. ***p < .001.

Table 5 reveals predicted probabilities of having protested in the recent past. Looking at the level of change in the predicted probability of having protested (rightmost column), the largest increase is associated with NGOs having come to the respondent’s community to provide goods or services. Thus, for example, an average man in Machakos district who reports that he has not had service provision NGOs come to his community has only a 5% chance of having protested, whereas one who has had NGOs work in his community two or more times has a one in three chance of having protested, an increase of more than 25%. In Nairobi and Mbeere districts, the effect is even greater—In Nairobi, having had NGOs visit the community two or more times to provide goods or services results in an increase from a 17% probability to more than 60% probability and, in Mbeere, from about 30% to 75%. Table 6 shows that the predicted change in an individual having raised an issue with a government official is also large and positive when a respondent reports having had NGOs come to their community 2 or more times, but it is not statistically significant and should therefore be interpreted cautiously.

Results are similar and consistently statistically significant, but are smaller, for the variable assessing whether respondents had sought out NGOs for goods or services in the previous year. Across all districts, an average respondent had an approximately 10% to 20% higher likelihood of having protested or raised an issue with government officials if they had gone to an NGO for services or goods two or more times in the prior year than individuals who had never gone to an NGO.

Although perception of NGOs’ presence and their moral legitimacy are not statistically associated with protesting, they are positively and statistically significantly associated with increased probability of raising an issue with a government official, as seen in Table 6. Respondents who report that they strongly agree that NGOs are morally legitimate organizations have, on average, an approximately 90% predicted probability of raising an issue with a government official. Those who strongly disagree with the idea of NGOs as morally legitimate have a 26%, 57%, or 38% predicted probability of raising an issue in Machakos, Mbeere, and Nairobi, respectively. Although not shown, those who strongly believe NGOs to be legitimate actors have an average predicted probability of voting of 98%, versus approximately 92% for those who strongly believe NGOs to be illegitimate.

Discussion and Conclusion

The findings of this study reveal an apparent relationship between individuals’ reports of NGOs and their likelihood of some types of democratic participation in the Kenyan capital and two rural districts. Contact with NGOs is linked to an increased likelihood of an individual participating politically by as much as 44%. This article thus adds evidence supporting the idea that even service provision NGOs serve a civil society function. It also further confirms that the theory of “civil society,” a term popularized by a French social theorist examining the United States in the 1800s, “travels.”

The relationship holds true for less-common, higher cost activities like protesting and raising an issue with a local government official, but not for the relatively easier act of voting. This finding accords with work examining civil society membership organizations in the United States as well as with work on NGOs in Latin America (Baggetta, 2009b; Boulding, 2010; Crenson & Ginsberg, 2004). Although scholars have often found that voting is affected differently from other forms of participation (Brass et al., 2019; Dionne et al., 2014; Hern, 2017; Mattes & Mughogho, 2009; Rosenstone & Hansen, 1993), it is worth exploring possible reasons for this variation. The simplest explanation for this Kenyan data is that nearly 90% of respondents in the sample who were eligible to vote did so, regardless of their contact with, or perceptions of NGOs, leaving little room for expansion.

But this may not be the full story. Scholarship has also shown that voting can be seen as less effective than other forms of participation, particularly in situations with weak democratic institutions (Boulding, 2014; Harris & Hern, 2019). Across Africa, voting is often not seen as an effective tool to voice individual or collective concerns (Dionne et al., 2014; Wantchekon, 2003), because voting is heavily orchestrated by local leaders, brokers, and politicians, and because political parties tend to lack issue-based platforms (Bratton et al., 2004). In Kenya, specifically, national elections are largely seen as ethnic censuses, were for decades largely noncompetitive in a single-party system, and, in recent years, have been marred by significant and repeated irregularities and violence (Kanyinga, 2011).

In such settings, raising an issue directly to a government official or protesting can be seen by citizens as a more effective form of democratic expression than voting (Boulding, 2014). Given Kenya’s electoral violence, they are possibly safer than voting as well. Where electoral institutions are insufficient, moreover, people tend to see protests not as calls for regime change, but as a way for citizens to voice their demands on particular issues, especially service provision issues (Harris & Hern, 2019). As such, they are often not even especially contentious, instead often focusing on an issue a government is aware it is underperforming on (Harris & Hern, 2019). Thus, it is possible that the broader institutional context in which NGOs act is more important than anything NGOs are doing (Boulding, 2014).

At the same time, it is worth systematically theorizing factors about NGO contact that could make individuals more likely to participate politically as this is an avenue ripe for future research. Several possibilities exist. And all of these causal paths plausibly explain why the frequency of contact with NGOs appears to be important: One experience with an NGO may be insufficient for producing political participation outcomes; repeated interaction may be needed.

First, exposure to NGO programming—even service provision NGOs—may facilitate increased civic participation due to NGOs’ general emphasis on participatory development approaches. Most NGOs explicitly strive to use participatory approaches in their work: Scholars have been writing about “participatory rural appraisals,” “reversals of learning,” “assisted self-reliance,” moving up the “ladder of participation,” and other methods for decades (Chambers, 1983; Korten, 1980; Uphoff et al., 1998). These strategies aim to empower individuals and engender collective action in communities to identify their own service provision needs related to education, health, sanitation, agriculture, or other service sector, and to design their own solutions, facilitated by the NGOs. NGOs using these participatory approaches often work by creating or drawing on existing community groups, thereby creating or reinforcing spaces for people to interact and associate with one another and build social capital, following classing notions of civil society (Boulding, 2010). But they also allow individuals to practice citizenship skills and gain cognitive orientations that make their participation more likely (Dodge & Ospina, 2016).

The increased likelihood of participation in protests and demonstrations may be an indirect effect of NGOs’ general participatory strategy, with its empowerment rhetoric and focus on collective organizing. The large increase on the likelihood of having raised an issue with a politician or government administrator following NGO contact also makes sense in this light and in the Kenyan context. The Kenyan government has, at most points in its history, frowned on overtly political NGO activities but not on service provision ones. This is reflected numerically: Only 4% to 10% of NGOs registered in Kenya are explicitly “governance” organizations focused on such things as human rights, democracy, or anti-corruption, whereas at least 90% provide services (Brass, 2016, NGOs Co-ordination Board, 2009, p. 34).

That said, it has been well documented that NGOs’ success rates at achieving “truly participatory development” or “transformative participation” are lower than their rhetoric suggests (Carr, 2011; Hershey, 2019; Mercer, 2003; White, 1996). Thus, it may not be that all NGOs or NGO programs are linked to greater democratic participation, but only those that succeed in their participatory mobilization. (Or it could be that spillover from participatory development practices is not a key causal path.)

Instead, a second mechanism by which NGOs may lead to increased democratic participation is through the flow of resources (Boulding, 2010; Murdie & Bhasin, 2011; Nelson-Nuñez, 2018). Service provision NGOs bring a range of financial, educational, informational, and other resources that can facilitate individual and collective political participation. For example, an NGO might bring resources to a community to facilitate business development (through business-related trainings, donations of supplies, or microfinance). If successful, these NGO resources can facilitate more efficient businesses and concomitantly more free time that individuals can use to organize collectively or engage the government. Likewise, an NGO might provide adult or youth education, and the skills learned could make it easier to contact a government official or organize people for protest. The same effects could stem from donated technology or provision of internet access by NGOs. Or an NGO might bring information to a community about a new government program that could cause residents to seek out government officials to obtain the services. Such resources are less likely to affect voting rates, especially as election day is a public holiday in Kenya and voting is done in person. International NGOs may have this effect more than local ones as they generally are better resourced.

Third, NGOs may facilitate increased political participation by directly linking citizens to government officials, which would make it easier and more likely that individuals would contact them. Over the past decades, NGOs have increasingly implemented their work in collaboration with government ministries (Abers & Keck, 2009; Brautigam & Segarra, 2007; Nurul Alam, 2011). As a result, NGO workers have close connections with their civil servant counterparts to whom they can refer citizens or groups. NGO and government personnel, moreover, frequently travel together to the communities in which they collaboratively work—often in NGO vehicles or using petrol that the government office lacks (Brass, 2016)—making it rather easy for citizens to raise an issue with a government official. NGOs workers endorsing individual or group claims can lend legitimacy to their requests as well (Nelson-Nuñez, 2018).

Finally, the very act of NGOs providing services may spur protest or contacting action. On one hand, it is possible that the mere arrival of NGOs providing new services in an area brings hope to the people that change is possible (Brass, 2012). On the contrary, NGO services are likely insufficient to meet demand, but may facilitate an increased demand for services from the government under these conditions (Nelson-Nuñez, 2018). The partial cover of services itself may motivate citizens to demand more through protest or contact (Hern, 2017).

Additional research is needed to understand which of these causal mechanisms explain the association between NGO provision of services and some forms of increased democratic participation. In the meantime, the findings of this article have practical import for Kenyan and international policy makers, donors, and NGOs. This is particularly true in the light of repeated electoral irregularities in Kenya—from mass postelection violence in 2008 to annulled presidential elections in 2017. Both may be related to the Kenyan government under Uhuru Kenyatta signaling a growing resistance to NGOs, particularly human rights and democratization NGOs—a trend that has become common around the world, as governments place restrictions on NGO activities (DeMattee, 2019; Dupuy & Prakash, 2020; Dupuy et al., 2016). The government of Uhuru Kenyatta, among other things, blamed NGOs for his being charged by the International Criminal Court in association with the 2008 postelection violence, refused to implement the NGO-lauded Public Benefits Organizations Act of 2013 despite orders from the Kenyan High Court in 2017, and instead periodically proposed very restrictive legislation on NGOs. In a country long admired as a bastion of peace and a Western ally in the volatile Horn of Africa region, these threats to democratic consolidation and civil society activities have been regarded as unacceptable by many Kenyans. It is possible that Kenyatta understands that even service provision NGOs increase some forms democratic participation among citizens, as has been shown in this article. Going forward, where NGO numbers continue to grow—even service provision NGOs—results from this article suggest we might expect to see a more vibrant democracy.

Footnotes

Author’s Note

This article would not have been possible without the Kenyans who were willing to participate in the study. I am grateful for research assistance from Mark O’Dell, as well as productive feedback from participants at the Stanford Center on Philanthropy and Civil Society (PACS) Junior Scholar Forum, the Midwest Group in African Political Economy (MGAPE), the African Studies Association Annual Meeting, and the American Political Science Association Annual Meeting. All remaining errors are my own.

Declaration of Conflicting Interests

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

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author received financial support from the National Science Foundation, the University of California, Berkeley, and Indiana University, Bloomington.

ORCID iD

Jennifer N. Brass

Notes

Author Biography

Jennifer N. Brass studies service provision and governance in developing countries. She authored Allies or Adversaries? NGOs and the State in Africa (Cambridge University Press), as well as a range of articles and chapters. She has conducted field research in Senegal, Djibouti, Kenya, and Uganda. Her doctorate is in political science.

References

Abers

R. N.

Keck

M. E.

(2009). Mobilizing the state: The erratic partner in Brazil’s participatory water policy. Politics & Society, 37(2), 289–314.

Almond

Verba

(1963). The civic culture. Princeton University Press.

Baggetta

(2009a). Civic opportunities in associations: Interpersonal interaction, governance experience and institutional relationships. Social Forces, 88(1), 175–199.

Baggetta

(2009b, August). Making citizens: How associations stimulate individual civic engagement through the development of civic mindfulness [Conference session]. American Sociological Association Annual Meeting, San Francisco, CA, United States.

Banks

Hulme

Edwards

(2015). NGOs, states, and donors revisited: Still too close for comfort? World Development, 66, 707–718.

Bebbington

(1997). New states, new NGOs? Crises and transitions among rural development NGOs in the Andean region. World Development, 25(11), 1755–1765.

Bebbington

Guggenheim

Olson

Woolcock

(2004). Exploring social capital debates at the world bank. The Journal of Development Studies, 40(5), 33–64. https://doi.org/10.1080/0022038042000218134

Booth

J. A.

Richard

P. B.

(1998). Civil society, political capital, and democratization in Central America. Journal of Politics, 60(3), 780–800.

Boulding

C. E.

(2010). NGOs and political participation in weak democracies: Subnational evidence on protest and voter turnout from Bolivia. Journal of Politics, 72(2), 456–468.

10.

Boulding

C. E.

(2014). NGOs, political protest and civil society. Cambridge University Press.

11.

Boulding

C. E.

Gibson

C. C.

(2009). Supporters or challengers? The effects of nongovernmental organizations on local politics in Bolivia. Comparative Political Studies, 42, 479–500.

12.

Brass

J. N.

(2012). Blurring boundaries: The integration of NGOs into the governance of Kenya. Governance, 25(2), 209–235.

13.

Brass

J. N.

(2016). Allies or adversaries: NGOs and the state in Africa. Cambridge University Press.

14.

Brass

J. N.

Harris

MacLean

L. M.

(2019). Is there an anti-politics of electricity? Access to the grid and reduced political participation in Africa [Afrobarometer working paper 182]. https://www.afrobarometer.org/publications/wp182-there-anti-politics-electricity-access-grid-and-reduced-political-participation

15.

Bratton

(1994). Civil society and political transition in Africa. Institute for Development Resources Reports. https://africanphilanthropy.issuelab.org/resources/19673/19673.pdf

16.

Bratton

Mattes

Gyimah-Boadi

(2004). Public opinion, democracy, and market reform in Africa. Cambridge University Press.

17.

Brautigam

D. A.

Segarra

(2007). Difficult partnerships: The World Bank, States, and NGOs. Latin American Politics and Society, 49(4), 149–181.

18.

Brown

D. S.

Brown

J. C.

Desposato

S. W.

(2007). Promoting and preventing political change through internationally funded NGO activity. Latin American Research Review, 42(1), 126–138.

19.

Carr

E. C.

(2011). Delivering development: Globalization’s shoreline and the road to a sustainable future. Palgrave Macmillan.

20.

Chahim

Prakash

(2014). NGOization, foreign funding, and the Nicaraguan civil society. Voluntas, 25(2), 487–513.

21.

Chambers

(1983). Rural development: Putting the last first. Longman.

22.

Clark

(1995). The state, popular participation, and the voluntary sector. World Development, 23(4), 593–601.

23.

Crenson

M. A.

Ginsberg

(2004). Downsizing democracy: How America sidelined its citizens and privatized its public. Johns Hopkins University Press.

24.

DeMattee

A. J.

(2019). Covenants, constitutions, and distinct law types: Investigating governments’ restrictions on CSOs using an institutional approach. Voluntas, 30, 1229–1255.

25.

Diamond

. (1997). Introduction: In search of consolidation. In Diamond

Plattner

M. F.

Chu

Y.-H.

Tien

H.-M.

(Eds.), Consolidating the third wave democracies (pp. xiii–xlvii). Johns Hopkins Press.

26.

Dionne

K. Y.

Inman

K. L.

Montinola

G. R.

(2014). Another resource curse? The impact of remittances on political participation [Afrobarometer working paper 145]. https://afrobarometer.org/publications/wp145-another-resource-curse-impact-remittances-political-participation

27.

Dodge

Ospina

S. M.

(2016). Nonprofits as schools of democracy: Comparative case study of two environmental organizations. Nonprofit and Voluntary Sector Quarterly, 45(3), 478–499.

28.

Dupuy

Prakash

(2020). Why restrictive NGO foreign funding laws reduce voter turnout in Africa’s national elections. Nonprofit and Voluntary Sector Quarterly. Advance online publication. https://doi.org/10.1177/0899764019897848

29.

Dupuy

Ron

Prakash

(2016). Hands off my regime! Governments’ restrictions on foreign aid to non-governmental organizations in poor and middle-income countries. World Development, 84(Suppl. C), 299–311.

30.

Edwards

Hulme

(1996). Too close for comfort? The impact of official aid on nongovernmental organizations. World Development, 6, 961–973.

31.

Fearon

J. D.

Humphreys

Weinstein

J. M.

(2015). How does development assistance affect collective action capacity? Results from a field experiment in post-conflict Liberia. American Political Science Review, 109(3), 450–469.

32.

Foley

M. W.

Edwards

(1996). The paradox of civil society. Journal of Democracy, 7(3), 38–52.

33.

Freedom House. (2018). Freedom in the world 2017, Kenya. https://freedomhouse.org/country/kenya/freedom-world/2017

34.

Gotchev

(1998). NGOs and promotion of democracy and civil society in east-Central Europe [NATO Research Fellowships Programme]. North Atlantic Treaty Organization.

35.

Gugerty

M. K.

Kremer

(2008). Outside funding and the dynamics of participation in community associations. American Journal of Political Science, 52(3), 585–602.

36.

Gyimah-Boadi

(1996). Civil society in Africa. Journal of Democracy, 7(2), 118–132.

37.

Harbeson

J. W.

(1994). Civil society and political renaissance in Africa. In Harbeson

J. W.

Rothchild

Chazan

(Eds.), Civil society and the state in Africa (pp. 1–32). Lynne Reinner.

38.

Harris

A. S.

Hern

(2019). Taking to the streets: Protest as an expression of political preference in Africa. Comparative Political Studies, 52(8), 1169–1199.

39.

Hearn

(2000). Aiding democracy? Donors and civil society in South Africa. Third World Quarterly, 21(5), 815–830.

40.

Hearn

(2007). African NGOs: The new compradors? Development and Change, 38(6), 1095–1110.

41.

Heinrich

V. F.

(2001). The role of NGOs in strengthening the foundations of South African democracy. Voluntas, 12(1), 1–15.

42.

Hern

(2017). In the gap the state left: Policy feedback, collective behavior, and political participation in Zambia. Studies in Comparative International Development, 52(4), 510–531.

43.

Hershey

(2019). Whose agency: The politics and practice of Kenya’s HIV-prevention NGOs. University of Wisconsin Press.

44.

Holmen

Jirstrom

(2009). Look who’s talking! Second thoughts about NGOs as representing civil society. Journal of Asian & African Studies, 44(4), 429–448.

45.

Howard

M. M.

(2003). The weakness of civil society in post-communist Europe. Cambridge University Press.

46.

Inglehart

(1997). Modernization and postmodernization: Cultural, economic, and political change in 43 societies. Princeton University Press.

47.

Kanyinga

(2004). Civil society formations in Kenya: A growing role in development and democracy. In Okello

(Ed.), Civil society in the third republic (pp. 152–159). National Council of NGOs.

48.

Kanyinga

. (2011). Stopping a conflagration: The response of Kenyan civil society to the post-2007 election violence. Politikon, 38(1), 85–109.

49.

Klesner

J. L.

(2007). Social capital and political participation in Latin America. Latin American Research Review, 42(2), 1–32.

50.

Korten

D. C.

(1980). Community development and rural development: A learning process approach. Public Administration Review, 40(5), 480–511.

51.

Lee

A.-R.

(2010). The quality of social capital and political participation in South Korea. Journal of East Asian Studies, 10(3), 483–505.

52.

Lee

A.-R.

Glasure

Y. U.

(2007). Social capital and political participation in South Korea. Asian Affairs: An American Review, 34(2), 101–118.

53.

LeRoux

Krawczyk

(2014). Can nonprofit organizations increase voter turnout? Findings from an agency-based voter mobilization experiment. Nonprofit and Voluntary Sector Quarterly, 43(2), 272–292.

54.

LeVan

A. C.

(2011). Questioning Tocqueville in Africa: Continuity and change in civil society during Nigeria’s democratization. Democratization, 18(1), 135–159.

55.

Lewis

(2004). On the difficulty of studying “civil society”: Reflections on NGOs, state and democracy in Bangladesh. Contributions to Indian Sociology, 38(3), 299–322.

56.

Marsh

(2002). Social capital and grassroots democracy in Russia’s regions: Evidence from the 1999-2001 gubernatorial elections. Demokratizatsiya, 10(1), 19–36.

57.

Mattes

Mughogho

(2009). The limited impacts of formal education on democratic citizenship in Africa [Afrobarometer working paper 109]. https://afrobarometer.org/publications/wp109-limited-impacts-formal-education-democratic-citizenship-africa

58.

Menon

Daftary

(2011). The impact of associational membership on political engagement: A comparative investigation of Brazil and India. International Social Work, 54(1), 81–96.

59.

Mercer

(2002). NGOs, civil society and democratization: A critical review of the literature. Progress in Development Studies, 2(1), 5–22.

60.

Mercer

(2003). Performing partnership: Civil society and the illusions of good governance in Tanzania. Political Geography, 22, 741–763.

61.

Moehler

D. C.

(2010). Democracy, governance, and randomized development assistance. ANNALS of the American Academy of Political and Social Science, 628(1), 30–46.

62.

Murdie

Bhasin

(2011). Aiding and abetting: Human rights INGOs and domestic protest. Journal of Conflict Resolution, 55(2), 163–191.

63.

Nelson-Nuñez

(2018). Substitution or facilitation: Service-delivery NGOs and political engagement in the Peruvian Amazon. Comparative Political Studies, 52(3), 445–477.

64.

Newton

(1997). Social capital and democracy. American Behavioral Scientist, 40(5), 575–586.

65.

Nongovernmental Organizations Co-ordination Board. (2009). Report on the national validation survey of NGOs. Government of Kenya.

66.

Nurul Alam

S. M

. (2011). Health service delivery: The state of government-non-government relations in Bangladesh. Public Administration and Development, 31(4), 273–281.

67.

Orvis

(2003). Kenyan civil society: Bridging the urban-rural divide? Journal of Modern African Studies, 41(02), 247–268.

68.

Paxton

(2002). Social capital and democracy: An interdependent relationship. American Sociological Review, 67(2), 254–277.

69.

Putnam

(1993). Making democracy work: Civil traditions in modern Italy. Princeton University Press.

70.

Putnam

(2000). Bowling alone: The collapse and revival of American community. Simon & Schuster.

71.

Rahman

(2006). Development, democracy and the NGO sector. Journal of Developing Societies, 22(4), 451–473.

72.

Rosenstone

S. J.

Hansen

J. M.

(1993). Mobilization, participation, and democracy in America. Longman.

73.

Sanyal

(2009). From credit to collective action: The role of microfinance in promoting women’s social capital and normative influence. American Sociological Review, 74(4), 529–550.

74.

Seligson

A. L.

(1999). Civic association and democratic participation in central America. Comparative Political Studies, 32(3), 342–362.

75.

Senate Foreign Relations Committee. (2006). Promotion of democracy by nongovernmental organizations: An action agenda: Testimony by ambassador Mark Palmer. U.S. Senate.

76.

Skocpol

(2003). Diminished democracy: From membership to management in American civic life. University of Oklahoma Press.

77.

Skocpol

Fiorina

M. P.

(Eds.). (1999). Civic engagement in American democracy. Brookings Institution Press.

78.

Tarnoff

Lawson

M. L.

(2011). Foreign aid: An introduction to U.S. programs and policy. DIANE.

79.

Tavits

(2006). Making democracy work more? Exploring the linkage between social capital and government performance. Political Research Quarterly, 59(2), 211–225.

80.

Tocqueville

A. D.

(1835). Democracy in America (2000 ed.). HarperCollins.

81.

Tomz

Wittenberg

King

(2003). CLARIFY: Software for interpreting and presenting statistical results. https://gking.harvard.edu/clarify

82.

Ulvila

Hossain

(2002). Development NGOs and political participation of the poor in Bangladesh and Nepal. Voluntas, 13(2), 149–163.

83.

United Nations Development Programme. (2019). United Nations human development report. United Nations.

84.

Uphoff

Esman

M. J.

Krishna

(1998). Reasons for success: Learning from instructive experiences in rural development. Kumarian Press.

85.

Verba

Schlozman

K. L.

Brady

H. E.

(1995). Voice and equality: Civic voluntarism in American politics. Harvard University Press.

86.

Wantchekon

(2003). Clientelism and voting behavior: Evidence from a field experiment in Benin. World Politics, 55(3), 399–422.

87.

White

S. C.

(1996). Depoliticising development: The uses and abuses of participation. Development in Practice, 6(1), 6–15.

88.

Wiktorowicz

(2002). The political limits to nongovernmental organizations in Jordan. World Development, 30(1), 77–93.

89.

World Bank. (2009). World bank–civil society engagement: Review of fiscal years 2007 to 2009.

90.

World Bank. (2018). Civil society. http://www.worldbank.org/en/about/partners/civil-society

91.

Zoellick

(2011). The Middle East and North Africa: A new social contract for development speech given at the Peterson Institute for International Economics. World Bank.