Sage Journals: Discover world-class research

Abstract

Across Africa, people have varied understandings of what “democracy” is, with implications for democratic support, political participation, and willingness to resist democratic backsliding. Drawing on insights from literature about varied understandings of democracy and partisan bias in perceiving democratic backsliding, I argue that partisanship influences the ways that citizens understand democracy—specifically, whether they understand it in procedural or instrumental terms--but that this relationship depends on the nature of political competition within a country. Using cross-sectional survey data from the Afrobarometer, alongside data from the World Values Survey over time in Ghana and Zimbabwe, this study demonstrates ruling party partisans are more likely to understand democracy in instrumental terms, while opposition party partisans understand it in more procedural terms. This partisan gap in understandings of democracy is more pronounced in countries with higher levels of clientelism and lower levels of multiparty competition.

Keywords

partisanship understandings of democracy Africa multipartyism clientelism

Introduction

Across Africa, majorities indicate that they prefer democracy to any other regime type.¹ Yet, people understand democracy in different ways. In many countries, people more commonly define democracy in instrumental terms—focusing on desirable outcomes like economic redistribution and public goods provision—than procedural ones, like fair elections or civil liberties (Canache, 2012; Crow, 2010; Norris, 2011; Teti et al., 2019). What explains whether citizens understand democracy in instrumental or procedural terms?

This question is consequential because citizen understandings of democracy likely influence other key outcomes. Recent scholarship demonstrates citizens who understand democracy in procedural terms are better able to assess their own country’s level of democracy, more supportive of democracy and democratization, and more likely to participate in both institutionalized and noninstitutionalized politics (Ariely, 2015; Canache, 2012; Chapman et al., 2024; Lu and Chu, 2021; Neundorf et al., nd). In contrast, as Pippa Norris (2011, p.150) cautions, “instrumental support for democracy is important and real, but it is also limited and conditional. If governments fail to deliver jobs, prosperity, and social services, then public enthusiasm for democracy may fade.” This cautionary note is important given waning enthusiasm in democracies like South Africa, and growing public support for military intervention against corrupt leaders (Afrobarometer, 2025).

This study evaluates the roles of partisanship and party systems in influencing how citizens understand democracy across Africa. I argue that partisanship influences understandings of democracy, but this relationship depends on the nature of party competition and party-citizen linkages. Across Africa, ruling parties enjoy disproportionate access to state resources and tend to adopt more clientelist messages and strategies than opposition parties, which emphasize democratic procedures. As a result, ruling party supporters are more likely to hold instrumental understandings of democracy than are opposition partisans. However, this relationship is attenuated by party competition: in more competitive party systems, where parties spend time both in power and in opposition, all citizens hold more procedural understandings of democracy.

I support this argument through quantitative analysis of Afrobarometer data. This analysis demonstrates that partisanship has a strong and comparatively large correlation with understandings of democracy, but that the strength of this relationship varies across countries according to the degree of clientelism and strength of party competition. To address concerns of reverse causality, I analyze how partisan understandings of democracy change when their party is in and out of power, using three rounds of World Values Survey (WVS) data covering alternations of power in Ghana and Zimbabwe.

This study advances an burgeoning new literature on democratic understandings (Bryan, 2023; Chapman et al., 2024; Fossati and Coma, 2023; Lu and Chu, 2021; Neundorf et al., nd; Wunsch et al., 2025) by highlighting partisanship and party messaging as key explanatory factors. It demonstrates that, even in weak party systems, political parties may matter more than often supposed. While African political parties are often derided as weak and ineffectual, this study adds to mounting evidence that they still shape citizens’ perceptions of the political realm (Carlson, 2016; Letsa, 2025).

This study proceeds to review the literature on variation in understandings of democracy. Section three elaborates my argument about partisanship and understandings of democracy across Africa. Section four describes the data and methods employed in the main analysis, while section five presents the results. Section six provides suggestive evidence of the causal direction using WVS data. A final section concludes.

Variation in understandings of democracy

Comparativists have long described varied understandings of democracy, often with the aim of linking conceptions of democracy with support for it (Chapman et al., 2024; Cho, 2014; Crow, 2010; Gerber and Chapman, 2018). In line with much of this literature, this study distinguishes between “procedural” and “instrumental” understandings of democracy. Procedural understandings are those that focus on democracy as a set of laws, institutions, and procedures, such as free and fair elections, free speech and association, and free press. This narrow understanding is sometimes referred to as the “procedural minimum” of institutions universally required for a country to be considered democratic (ex. Schmitter and Karl, 1991). With its focus on both free and fair elections and civil liberties, it is closely related to conceptions of liberal democracy, requiring both electoral democracy and the protection of individual rights (Claassen et al., 2025; Coppedge et al., 2011).

In contrast, instrumental understandings focus on outcomes that one might hope a democracy produces. For example, because the poor have greater opportunity to participate in a democracy than an autocracy, one might hope that a democracy would produce pro-poor, redistributive policies. However, democracies are not necessarily more redistributive, however one might hope they are. Instrumental understandings are distinct from egalitarian or other “thick” conceptions of democracy (Coppedge et al., 2011). While both instrumental and egalitarian conceptions of democracy may emphasize socioeconomic equality, egalitarian conceptions include free and fair elections and multiparty competition as equally crucial. Distinctively, instrumental understandings of democracy may not include procedural components and may also include other outcomes, such as security and order. In this study, I thus focus on the distinction between procedural understandings of democracy based on institutions, procedures, and laws and instrumental understandings focused on substantive outcomes.

Scholarship examining how citizens understand democracy describes significant differences across and within countries and regions (Ariely, 2015; Chu and Huang, 2010; Davis, 2022; Gerber and Chapman, 2018; Hale, 2011; Norris, 2011). In Africa, early quantitative analyses found widespread procedural understandings (Bratton and Mattes, 2001; Dalton et al., 2007). However, these understandings were uneven and included instrumental components, such as the expectation that democracy meant economic development (Bratton, 2010; Bratton and Mattes, 2001; Mattes and Shenga, nd). Open-ended survey questions about the meaning of democracy returned eclectic responses, including procedural components like freedom of speech (12%) and electoral choice (3%) alongside substantive outcomes like “peace and unity” (7%) (Bratton et al., 2004: 68). Qualitative investigations indicated that Africans’ understandings of democracy were contextually specific and diverged from the liberal model (Ake, 1993; Karlström, 1996; Schaffer, 2000; Van Binsbergen, 1995). For example, in Senegal, Schaffer (2000) describes how the Wolof understanding of demokaraasi resembles local political traditions based on consensus, community solidarity, and bloc voting rather than individualist liberal democracy, despite being the same word in translation. There continues to be variation in how Africans perceive and understand democracy (Hern, 2020). Summarizing data from the first six rounds of the Afrobarometer, Mattes (2019) shows that African understandings of democracy are sensitive to question wording; while many survey respondents correctly identify procedures associated with democracy, they also regularly invoke material outcomes as being essential democratic characteristics.

Scholarship investigating the source of varied understandings have mostly focused on the impact of individual-level characteristics related to socio-economic status (SES). On average, people who experience greater disadvantage (the less educated, the poor, and women) are more likely to hold instrumental understandings of democracy (Gerber and Chapman, 2018; Letsa and Wilfahrt, 2018; Lu and Chu, 2021). While there is little definitive research documenting the mechanisms linking individual characteristics to varied understandings, most proposed pathways focus on access to information and varying preferences related to material need.

Nationally, understandings of democracy correlate with regime type and national income. Citizens of democracies are more likely to hold procedural understandings because political elites and the news media regularly reinforce discourse about the procedural elements of democracy (Cho, 2014), while in in hybrid and authoritarian states, political leaders may strategically manipulate definitions of democracy in order to claim that their country is democratic (despite evidence to the contrary) (Miller et al., 1997; Norris, 2011). For example, facing criticism of his deepening authoritarianism, Rwandan President Paul Kagame has maintained that his country is democratic by highlighting its post-genocide stability and economic growth while downplaying the lack of true political competition.² While the mechanism linking national income to understandings of democracy is unclear, some scholars posit a version of modernization theory, emphasizing the development of post-material values at higher incomes (Lu and Chu, 2021: 4).

There has been limited evaluation of how partisanship and party systems influence understandings of democracy, though there are theoretical and empirical reasons to believe it would. Recent research investigating citizen opinions of democratic violations reveals clear partisan patterns: numerous studies have illustrated that citizens support anti-democratic behavior carried out by their co-partisans under various conditions (Bryan, 2023; Gidengil et al., 2022; Graham and Svolik, 2020; Krishnarajan, 2023; Saikkonen and Christensen, 2023; Simonovits et al., 2022). These studies demonstrate that people’s conceptions of democracy can be flexible, inconsistent, and—importantly—influenced by partisan cues. Elite cues can be powerful in shaping co-partisans’ views and behavior. In Turkey, Windecker et al. (2025) found that partisan cues shaped citizens’ perceptions of election integrity, particularly under electoral autocracy. This finding accords with Moehler’s (2009) earlier work showing that ruling partisans across 12 African countries were far less critical of democratic lapses. In the United States, elite cues contribute to political polarization on issues such as climate change (Van Boven and Sherman, 2021) and immigration (Druckman et al., 2013). In Argentina, they drive polarization over previously non-divisive issues, such as establishing a minimum retirement pension (Yeyati et al., 2020). In the United States, elite cuing by President Trump detracting mail-in voting led some of his supporters, who had voted by mail, to lie about their voting method (Shino et al., 2022). Since elite cues can shape perception, opinion, and behavior, they may also shape the way people understand fundamental political concepts. Uncovering the drivers of these perceptions is particularly important for understanding how and when citizens will resist democratic incursions.

Argument and hypotheses

I contend that party cues inform the way people construct their understandings of the political world and, therefore, the way they understand democracy. Parties make explicit statements to their supporters about the nature and purpose of democracy and convey implicit messages about their role and prospects in the political system. Parties can also shape citizens’ understandings of democracy through the nature of their interactions. Whether parties focus on distributing material goods or checking a powerful executive is consequential for what citizens learn about democratic practice. These foci are context-dependent, and therefore one might expect variation across time, region, regime type, and other institutional factors that shape how party cues influence perceptions. I argue that in the African context, ruling and opposition politicians send different messages to and interact with voters in different ways, with consequences for how voters understand democracy. However, this relationship varies according to the nature of party competition and the degree of clientelism.

Partisanship and understandings of democracy in Africa

Two features of African party systems are theoretically important for shaping citizens’ understandings of democracy: the nature of party competition and party-citizen linkages. Nearly all African countries have electoral systems, though many fall short of democracy. The modal African country has a dominant party system in which the ruling party holds a powerful presidency and a legislative supermajority, and pursues clientelist linkages with voters (Bleck and van de Walle, 2019, Chapter 4). Some ruling parties ensure their dominance through electoral manipulation and coercion. In other cases, elections are reasonably free though typically unfair. Dominant party systems range from thinly-veiled authoritarian regimes (ex. Gabon) to largely democratic ones (ex. Botswana). Ruling parties have disproportionate or near exclusive access to state resources that they can use to bolster their incumbency, while opposition parties tend to be considerably weaker and resource-poor. This differential access to state resources generates different party-citizen linkages and campaign tactics for ruling versus opposition parties.

African political parties typically campaign on valence issues such as service delivery and economic development rather than ideologically distinct platforms (Bleck and Van de Walle, 2013). However, ruling and opposition parties can claim “ownership” over different valence issues in ways that likely influence how partisans understand democracy (Letsa, 2019). Ruling parties, with disproportionate access to state resources, may lean into clientelist linkages emphasizing material benefits like jobs, club goods, or personal assistance (Lindberg, 2010; Van de Walle, 2003). They bolster this expectation through the distribution of goods at campaign rallies (such as bags of rice or wax cloth), which signals to voters that they have the resources to deliver benefits (Jöst and Lust, 2022; Kramon, 2016). While citizens may take advantage of such campaign largesse without voting for the party (Musonda, 2023), ruling party candidates continually reinforce the narrative connecting party support to material gain, through one-shot campaign interactions, emphasis on the development goods they have delivered, or distributive politics that favor supporters with services like farm inputs. Additionally, many long-ruling parties in Africa started as anti-colonial “liberation” parties. While this liberation was rooted in transition to self-rule, it was also often bound up with promises of economic gain (Ake, 1993).

Alternatively, opposition parties lack access to the state resources to support such campaigning, and lack the track record in office to claim credit for development projects or economic progress. As Letsa (2019) has documented, opposition parties are best placed to “own” the issue of democracy, campaigning against the authoritarian abuses of dominant parties and for procedural democratic improvements. Focusing on procedural elements like fair electoral competition, free speech, and freedom of association is strategically important for opposition parties operating on uneven playing fields. Letsa (2025) shows that in authoritarian Cameroon, opposition partisans are thus much more likely to recognize that the country is not democratic. Additionally, Bertrand (2021: 605-6) argues that a primary role of opposition parties in hybrid regimes is to denounce the anti-democratic actions of the incumbent, such as attempted term limit contravention. Opposition partisans are therefore more likely to internalize a narrative of democracy that emphasizes the procedures that protect fair electoral competition.

H₁: Supporters of opposition parties [the incumbent] are more likely to hold procedural [instrumental] understandings of democracy.

Qualitative accounts of campaigning across African dominant-party systems bears out this distinction. Paget (2024) describes some African opposition parties as “anti-authoritarian populists”: in Tanzania, the opposition Chadema characterizes the ruling CCM as anti-democratic, accusing the government under Magufuli of an “unrelenting onslaught on democracy” (877). Similarly, he notes that the opposition in Zimbabwe framed their campaign as a “democratic struggle” against authoritarianism (882). Conversely, Letsa (2019:5) illustrates how dominant ruling parties campaign on the benefits they have purportedly delivered: in Zimbabwe, ruling ZANU-PF in 2011 campaigned on having delivered “peace,” Cameroon’s ruling RDPC campaigned on the “great achievements” of its development projects such as a new deep water port, and Uganda’s ruling NRM campaigned in 2016 on job creation and economic development.

If understandings of democracy are in part transmitted through party cues, then nonpartisans may be agnostic about what democracy is. In surveys, this orientation might manifest through “don’t know” responses or refusal to respond to questions about the nature of democracy.

H₂: Nonpartisans are more likely to fail to respond to questions about the meaning of democracy.

Variation in party systems and linkages should moderate the partisan gap in understandings of democracy. First, the expectation that ruling partisans are more instrumental is predicated on the expectation of clientelist linkages. In countries with less clientelism, ruling parties are less likely to reinforce such narratives through their campaign strategies. Thus, the partisan gap in understandings of democracy should be smaller in less clientelist countries. Similarly, the ability of opposition parties to “own” the issue of democracy diminishes with stronger multipartyism. If there is regular alternation of power and stronger institutions protecting party competition, the opposition will struggle to claim ownership over democracy as their key issue (particularly if they have recently held the presidency). Rather, if all citizens have greater exposure to procedural rules that protect multiparty competition, then all citizens should be more likely to hold procedural understandings of democracy, regardless of their partisanship. For example, in an analysis of four elections campaigns across Kenya and Ghana, where there had been alternation in power, Taylor (2017) found that campaigns included programmatic appeals, distributive promises, and ethnic mobilization, but not specific content about democracy. While the incumbent in clientelist multiparty countries still has an advantage in campaigning on distributive “pork” and offering one-time material incentives (Brierley and Kramon, 2020), alternation of power means that such clientelist practices are not associated with one party, as in dominant party systems. I thus expect clientelism to intensify the partisan gap in understandings of democracy, while multipartyism tempers it.

H₃: The partisan gap in understandings of democracy is larger in more clientelist countries.

H₄: The partisan gap in understandings of democracy is smaller in countries with stronger multipartyism.

Alternative explanation

While my argument presumes that the causal direction runs from party cues to partisan beliefs, this is not the only plausible causal story. In countries where democracy is threatened by pervasive clientelism or a lower level of multipartyism, procedurally-minded individuals may sort themselves into opposition partisanship. This interpretation reverses the causal arrow, suggesting that beliefs around democracy drive partisan attachments, rather than the inverse. It is difficult to test the causal direction of this relationship without panel data covering individuals in countries where the ruling party has changed over time. However, I address this issue with suggestive data from the WVS in section six.

Data and methods

The data to test these hypotheses predominantly come from Round five of the nationally representative Afrobarometer survey, carried out from 2011-2013.³ The main variable of interest in this study is “Instrumental,” measures of how often respondents selected instrumental rather than procedural definitions of democracy. It is derived from a series of four questions asking respondents to identify what they think of as an “essential characteristic of democracy.” Each time, the respondent must select one of four options, divided between two instrumental and two procedural responses. Procedural responses come in two forms: civil liberties (freedom of media, freedom of speech, freedom of association) or democratic institutions (elections, multiple parties, legislative oversight). Similarly, instrumental responses come in two forms: government performance (governing efficiency, law & order, provision of public goods), or provision of welfare or personal benefits (closing the gap between the rich and poor, provision of jobs, provision of basic needs). The variable “Instrumental” is a count of how many times each respondent selected an instrumental rather than a procedural response as an “essential characteristic of democracy,” and thus ranges from 0 to four where 0 represents no instrumental responses and four represents all instrumental responses. The wording and structure of these questions, alongside alternative operationalizations, are included in the appendix.⁴ “Agnostic” is a binary variable coded as one if respondents failed to answer any of the questions about components of democracy.

Approximately 10% of the sample selected no instrumental responses, 23% selected one, 27% selected two, 24% selected three, and 15% selected four. The mean number of instrumental responses is 2.1 Figure 1 depicts country-level variation in the mean, ranging from a high of 2.8 in Swaziland (now Eswatini) to a low of 1.5 in Tanzania. Rates of non-response also vary greatly, ranging from a high of 34% in Madagascar to a low of 1.4% in Benin, with a sample mean of 13%.

Table 1.

Partisanship and instrumental understandings over time in Ghana.

	Instrumental understandings (full model)	Instrumental understandings (2007)	Instrumental understandings (2012)
Ruling partisans	−0.238
Ruling partisans	(0.185)
NDC partisans		0.856**	−0.302
(vs.NPP partisans)		(0.308)	(0.293)
Controls?	YES	YES	YES
N	2972	1097	1132

Coefficients with standard deviations in parentheses. ***p < 0.001; **p < 0.05; *p < 0.1.

Figure 1.

Mean of instrumental responses by country.

The primary independent variable is partisanship, operationalized using the question “Is there any party you feel close to? (If yes) Which one?” The resulting categorical variable is coded as 0 if respondents report they do not feel close to any party, one if they report feeling close to a party that does not hold the executive branch, and two if they report feeling close to the party that holds the executive branch. In more authoritarian regimes, opposition party support is likely to be underreported—some opposition partisans may declare themselves non-partisan, thus muddying the “nonpartisan” category. As such, using non-partisan as a base category may minimize the difference between opposition party partisans and nonpartisans, as discussed in the results section.

Each analysis also includes a suite of individual- and country-level control variables that proved important for understandings of democracy in previous research. Individual-level controls include poverty, education, political interest, civil society membership, sex, age, internet access, and urban residence. Country-level controls include measures of national income (HDI), duration of democracy, and regime components (multipartyism, media freedom, repression) measured during the survey year.⁵

To predict instrumental understandings of democracy, I used multilevel mixed effects regression, nesting individuals within countries, with random effects for country. This modeling strategy allows both the slope and intercept to vary by country. The results presented in the main text are from linear regression to facilitate direct comparison across the models examining instrumental understandings and agnosticism. However, as “instrumental” is a count variable and “agnostic” is a binary variable, the appendix includes additional estimation strategies (Poisson, ordered logit, and logit) that account for the nonlinear nature of these dependent variables. The results are robust to these alternative estimation strategies.

Results

Main models

Figure 2 shows the relationship between partisanship and instrumental understandings of democracy (triangles) and agnostic responses (squares). These results demonstrate support for H1 and H2. First, compared to non-partisans, opposition partisans are far less likely to hold instrumental understandings of democracy, while ruling partisans are more likely. As noted above, the nonpartisan category also likely includes quiet opposition partisans in the more authoritarian countries, which may reduce the magnitude of the difference between opposition partisans and nonpartisans. Collapsing the variable to compare opposition and ruling partisans directly confirms that ruling partisans have considerably more instrumental understandings than opposition partisans (appendix table A4). H2 posited that non-partisans should be more agnostic about democracy. The results support this contention: both opposition and ruling partisans are less likely to be agnostic about democracy than non-partisans (and opposition and ruling partisans do not differ significantly from each other).

Figure 2.

Partisanship and understandings of democracy.

Interaction models

Next, I hypothesized that party cues and behavior would differ across countries according to the degree of clientelism and multipartyism As measured by V-Dem on a 0-1 scale, clientelism scores range from 0.2 to .82, with a mean of 0.57. Measured on a 0-4 scale, multipartyism ranges from 0.65 to 3.97, with a mean of 3.67 (alternative measures in the appendix, Table A5). If incumbents foster instrumental understandings of democracy because of their tendency to campaign on club goods and patronage, then the partisan gap in understandings of democracy should be higher in countries with more clientelism. Alternatively, in countries with more robust party competition, one would expect incumbent and opposition parties to behave more similarly and the partisan gap in understanding to be smaller. To test this claim, I dropped non-partisans from the sample and ran models interacting ruling versus opposition partisanship with a measure of clientelism and multiparty competition.⁶ Figure 3 displays the linear predictions of understandings of democracy for opposition partisans (solid line) and ruling party partisans (dashed line) over varying degrees of clientelism and multipartyism.⁷ Panel A shows that, as clientelism increases, the partisan gap in understandings of democracy increases. At the lowest levels of clientelism observed in the sample, ruling and opposition partisans are indistinguishable in how instrumentally they understand democracy, but a sizeable gap appears as clientelism increases. Conversely, Panel B illustrates how the partisan gap collapses when multipartyism increases. These models are robust to Poisson and ordered logit specification (appendix table A3) and lend credence to the mechanisms underlying the relationship between ruling versus opposition party partisanship and understandings of democracy in Africa.

Figure 3.

Interaction models of partisanship and instrumental understandings of democracy.

Substantively, the size of the relationship rivals the impact of other individual-level factors. In the full sample, the marginal effect of opposition versus ruling partisanship is a 0.15 point decrease on the four-point measure of instrumental understandings. In the interaction models, the partisan gap is zero at the lowest observed levels of clientelism but 0.26 points at the highest. The conditional impact of multipartyism is even larger: the partisan gap is zero under the highest observed levels of multipartyism, but the gap grows to 0.31 points once multipartyism drops beneath the median score.

The control variables behave as expected, in line with existing literature: instrumental understandings are concentrated among people who are likely to face political marginalization along other lines: the poor, the less educated, and women. Education and gender are the other two individual-level characteristics that rival or exceed partisanship in their magnitude. This result is consistent with previous research that linked procedural understandings to civic education and democratic socialization that can take place at higher levels of education (Croke et al., 2016) and instrumental understandings to deprivation (women tend to have less access to wealth and power) (Lu and Chu, 2021). Characteristics that correlate with political connectedness—internet access, group membership, and urban residence—are associated with less instrumental understandings. Country-level variables are omitted from the plot because their larger confidence intervals distort the x-axis. The country-level variables are all insignificant, with the exception that increased multipartyism is associated with more procedural understandings of democracy.

Alternative explanation: Self-sorting partisans

An alternative explanation is that people with more procedural understandings of democracy select into opposition partisanship, while people with more instrumental views select into ruling partisanship, and that this phenomenon predominantly occurs in countries with higher levels of clientelism and limited multipartyism. One way to test this possibility is to examine what happens to partisan understandings of democracy when there is a rotation of power: when a long-time opposition party gains power, do its partisans’ understandings become more instrumental? Conversely, if a party loses power, do its partisans’ understandings become less instrumental?

Data and methods

Unfortunately, there are very limited data to test this idea, as (a) there is only one cross-national survey—the WVS—that asks about understandings of democracy in procedural and instrumental terms over multiple rounds, and (b) rotation of power in countries with lower levels of multipartyism is rare. Fortunately, the WVS fielded rounds asking about characteristics of democracy before and after a rotation of power in two African countries: Ghana and Zimbabwe. These countries offer important variation along the key country-level factors identified above. Ghana has the strongest degree of multipartyism in the sample (3.97 out of 4) and has clientelism scores near the mean (0.56). Zimbabwe has weaker multipartyism (3.65) and much higher clientelism (0.72). Because Ghana experiences regular rotation of power, one would expect minimal partisan differences in understandings of democracy. In Zimbabwe, the dominant party system and comparatively high levels of clientelism suggest larger partisan differences in understandings of democracy.

In Ghana, the two main political parties are the New Patriotic Party (NPP) and the National Democratic Congress (NDC). Since Ghana’s adoption of multiparty democracy in 1992, it has reliably alternated power every 8 years, in 2000, 2008, 2016, and 2024 (Frempong and Amankwah, 2024). The WVS conducted surveys in 2007, when the NPP held the presidency, and in 2012, when the NDC did. In Zimbabwe, the ZANU-PF has dominated party politics since the country gained majority rule in 1980. The Movement for Democratic Change (MDC), the main opposition party, has never won an election outright. However, after an escalating political crisis and pressure to democratize, the MDC and ZANU-PF entered a power-sharing agreement from 2009 to 2013, after which the ZANU-PF reasserted its electoral dominance (LeBas, 2014).⁸ The WVS fielded surveys in 2012, during the power-sharing agreement, and in 2020, after its conclusion.

The following analysis evaluates the relationship between co-partisanship with the executive and instrumental understandings of democracy over two time points in Ghana and Zimbabwe by country-year using OLS. Control variables include age, sex, and political interest. The other variables included in the analysis above (rural residence, education, poverty, and group membership) were unfortunately unavailable for all four country-years, so had to be omitted from the analysis. “Instrumental” is an additive index of the importance that respondents afforded to aid and redistribution as essential characteristics of democracy, generating a variable that ranges from 2-20. Full rationale for this operationalization is in the appendix.

Results

In Ghana, there was no relationship between partisanship and instrumental understandings of democracy. The only significant variable predicting instrumental understandings between 2007 and 2012 was the year (β = −2.42, p = 0.00), indicating a considerable decline in instrumentalism between 2007 (mean of 12.6) to 2012 (mean of 10.3). Examining the relative instrumentalism of NPP and NDC partisans each year reveals that, in 2007, NDC partisans were marginally more instrumental than their NPP counterparts (p = 0.006), despite being out of power at the time. In 2012, when NDC was in power, instrumental views between the two groups were indistinguishable (p = 0.303). In Ghana, where these two parties regularly alternate power, instrumental understandings of democracy have decreased over time, and partisanship does not have predictive power.

While Ghana has moderate levels of clientelism, it impact on understandings of democracy is likely attenuated by the regular rotation of power and its highly institutionalized party system (Frempong and Amankwah, 2024). As Lindberg (2010) has illustrated, patron-client dynamics in Ghana occur across party lines. Differential incentives to campaign on either patronage or pro-democracy sentiments are absent, as neither party has undisputed control over government resources nor can claim the mantle of “democratic agitator.” While neither party tends to campaign on programmatic platforms, they are each highly organized with stable support bases (Osei, 2013). Approximately 20% of the country’s voters are swing voters, whose votes tend to be determined by retrospective evaluation of government performance, reliably delivering a rotation of power every other election (Lindberg and Morrison, 2005). In this context, instrumental understandings of democracy are evenly distributed across party lines and declining over time.

In Zimbabwe, where multipartyism is weaker and clientelism is more pronounced, partisanship predicts instrumental understandings both in general and over time. While ZANU-PF has always been a ruling party, the MDC was in the ruling coalition during the 2012 survey, but not the 2020 survey. As Table 1 displays, across both years, ruling party partisans had more instrumental understandings than opposition partisans. Instrumental understandings increased by about 2.3 points between 2012 and 2020. In 2012, during the power-sharing agreement, MDC partisans were more instrumental than even ZANU-PF, by about one point on the 20-point scale. Once they were back out of power in 2020, however, MDC partisans became less instrumental than ZANU-PF partisans, by about 1.4 points. This pattern is consistent with MDC partisans conceptualizing democracy in terms of the spoils they expected once MDC gained power in 2009, but not after their crushing defeat.

Prior to the power-sharing arrangement, MDC and ZANU-PF campaigned in terms that would have reinforced different understandings of democracy: MDC as pro-democracy activists, and ZANU-PF as the only party with the legitimacy to rule due to its origins as a liberation movement, alongside a hefty dose of patronage (LeBas, 2014: 60–63). However, the period of power-sharing changed the MDC in two ways: first, as members of the ruling coalition, the party lost its claim as the pro-democratic government outsider, and “party leaders abandoned the consciously polarizing prodemocracy rhetoric” (LeBas, 2014: 60). Second, ZANU-PF ran roughshod over the MDC in government, preventing them from achieving policy goals (Aeby, 2018; Bratton, 2014). Indeed, the MDC attempted to position itself as the party that would deliver improved public services, but its inability to do so in the context of the coalition government led to declining support over time (Bratton, 2014: 135–6). In contrast, LeBas notes that ZANU-PF exited the arrangement having found “new means of deploying patronage,” which may explain the increasing instrumentalism—particularly among ZANU-PF partisans—over time (2014: 62). These differences have persisted, with Freisen (2022) finding that MDC partisans are still much more likely to prioritize multipartyism and term limits than their ZANU-PF counterparts.

Conclusion

What people understand as “democracy” varies widely. Alongside SES and national-level characteristics, this study indicates that partisanship is another likely source of variation in how people understand democracy. With a focus on electoral regimes in Africa, this study has illustrated that opposition partisans hold more procedural understandings of democracy, while ruling partisans hold more instrumental understandings. This pattern is likely due to the nature of party competition across much of Africa, where dominant incumbent parties monopolize state resources and campaign on clientelism, while perpetual opposition parties claim the issue of “democracy.” Accordingly, partisan gaps in understandings of democracy are much larger in countries that have higher degrees of clientelism and less multiparty competition. Further analysis of partisan beliefs about democracy over time in Ghana and Zimbabwe affirms that partisan understandings of democracy change over time depending on whether their party is in power, but that the partisan gap in understandings disappears in countries with regular alternation of power. Crucially, this study demonstrates that even in a context where parties are considered weak, they still shape the cognitive process through which people understand the political world.

How people understand democracy influences the degree and depth of their support for it. As Norris (2011) notes, support for democracy on instrumental terms is fickle, and likely to evaporate when countries face economic crises. A growing body of research indicates that support for the institutions and procedures underlying democracy is critical for resistance to democratic backsliding, making citizens less tolerant of democratic violations (Claassen, 2020; Jacob, 2025). Understanding democracy specifically in terms of institutions and procedures reduces the likelihood of supporting anti-democratic co-partisans (Eroglu et al., 2025). Conversely, more instrumental understandings may lead citizens to overestimate the supply of democracy in their country (Mattes and Teka, 2016).

This study demonstrates the importance of investigating how elite cues from political parties may also shape understandings of democracy. While the findings described here may be limited to clientelist dominant party systems, the underlying theory has broader applicability. It suggests that opposition parties in hybrid regimes offer important counter-programming if the dominant ruling party consistently describes its authoritarian actions as “democratic.” Extrapolating beyond this study, political parties broadly may influence the way their partisans understand democracy. Malign actors in declining democracies may actively manipulate these understandings, eroding public resistance to the democratic decline. In such cases, opposition parties may also be critical messengers to reinforce what democracy is.

This study has important limitations that future research might address. First, this study is not causally identified. While examining the trends in Zimbabwe over time lend some credence to the causal direction of the argument, it is still plausible that people in dominant party contexts self-sort into ruling and opposition partisanship due to their pre-existing understandings of democracy. Because I expect party messaging on democracy to be influential through repeated exposure over time, it is challenging to design an experimental intervention that would have external validity. Further research could track party messaging on democracy with co-partisan understandings of democracy over time to determine whether the former drives the latter.

Relatedly, it is not clear how mutable people’s understandings of democracy are. The WVS survey illustrates that, at the national level, understandings of democracy shifted over time in both Ghana and Zimbabwe—becoming less instrumental in the former and more instrumental in the latter. New experimental research suggests that simple civic education interventions may shift the way people understand democracy, but it is unclear how durable such an intervention might be (Eroglu et al., 2025). Additionally, co-partisan messaging may be more impactful than a neutral messenger, and the duration and consistency of the messaging likely influences its durability. Despite these open questions, the role of political parties in shaping the way populations conceptualize democracy is critical for understanding when people are more supportive of democracy, and when they may resist attempts to undermine it.

Supplemental material

Supplemental Material - Partisanship, party systems, and understandings of democracy across Africa

Supplemental Material for Partisanship, party systems, and understandings of democracy across Africa by Erin Hern in Party Politics

Footnotes

Acknowledgements

This manuscript has greatly benefitted from input from Yael Zeira, Jaimie Bleck, Lauren Honig and audiences at APSA, Syracuse University, and the Institute for African Development at Cornell University.

ORCID iD

Erin Hern

Ethical considerations

Ethical approval was not required for this study.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

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

Data Availability Statement

The data used in this study are publicly available. Replication files available upon request*.

Supplemental material

Supplemental material for this article is available online.

Notes

Author biography

Erin Hern is an Associate Professor of Political Science at the Maxwell School of Citizenship and Public Affairs at Syracuse University. She has published widely on African politics, including on the politics of service delivery, gender, and political behavior.

References

Aeby

(2018) Inside the inclusive government: interparty dynamics in Zimbabwe’s power-sharing executive. Journal of Southern African Studies 44(5): 855–877. https://doi.org/10.1080/03057070.2018.1497122

Afrobarometer (2025) African Insights: Democracy at Risk - The People’s Perspective. Afrobarometer Flagship Report. Afrobarometer.

Ake

(1993) The unique case of African democracy. International Affairs 69(2): 239–244. https://doi.org/10.2307/2621592

Ariely

(2015) Democracy-assessment in cross-national surveys: a critical examination of how people evaluate their regime. Social Indicators Research 121(3): 621–635. https://doi.org/10.1007/s11205-014-0666-y

Bertrand

(2021) Opposition in a hybrid regime: the functions of opposition parties in Burkina Faso and Uganda. African Affairs 120(481): 591–610. https://doi.org/10.1093/afraf/adab023

Bleck

van de Walle

(2013) Valence issues in African elections: navigating uncertainty and the weight of the past. Comparative Political Studies 46(11): 1394–1421, Sage Publications Inc. https://doi.org/10.1177/0010414012453448

Bleck

van de Walle

(2019) Electoral Politics in Africa since 1990: Continuity in Change. New York: Cambridge University Press.

Bratton

(2004) Public Opinion, Democracy, and Market Reform in Africa. Cambridge University Press.

Bratton

(2010) The meanings of democracy: anchoring the ‘D-Word’ in Africa. Journal of Democracy 21(4): 106–113. https://doi.org/10.1353/jod.2010.0006

10.

Bratton

(2014) Power Politics in Zimbabwe. Lynne Rienner Publishers.

11.

Bratton

Mattes

(2001) Support for Democracy in Africa: intrinsic or instrumental? British Journal of Political Science 31(3): 447–474. https://doi.org/10.1017/s0007123401000175

12.

Brierley

Kramon

(2020) Party campaign strategies in Ghana: rallies, canvassing and handouts. African Affairs 119(477): 587–603. https://doi.org/10.1093/afraf/adaa024

13.

Bryan

(2023) What kind of democracy do we all support? How partisan interest impacts a citizen’s conceptualization of democracy. Comparative Political Studies 56(10): 1597–1627. https://doi.org/10.1177/00104140231152784

14.

Canache

(2012) Citizens’ conceptualizations of democracy: structural complexity, substantive content, and political significance. Comparative Political Studies 45(9): 1132–1158. https://doi.org/10.1177/0010414011434009

15.

Carlson

(2016) Finding partisanship where we least expect it: evidence of Partisan bias in a New African democracy. Political Behavior 38(1): 129–154. https://doi.org/10.1007/s11109-015-9309-5

16.

Chapman

Hanson

Dzutsati

, et al. (2024) Under the veil of democracy: what do people mean when they say they support democracy? Perspectives on Politics 22(1): 97–115. https://doi.org/10.1017/s1537592722004157

17.

Cho

(2014) To know democracy is to love it: a cross-national analysis of democratic understanding and political support for democracy. Political Research Quarterly 67(3): 478–488. https://doi.org/10.1177/1065912914532721

18.

Chu

Huang

(2010) The meanings of democracy: solving an Asian puzzle. Journal of Democracy 21(4): 114–122. https://doi.org/10.1353/jod.2010.0009

19.

Claassen

(2020) Does public support help democracy survive? American Journal of Political Science 64(1): 118–134. https://doi.org/10.1111/ajps.12452

20.

Claassen

Ackermann

Bertsou

, et al. (2025) Conceptualizing and measuring support for democracy: a new approach. Comparative Political Studies 58(6): 1171–1198, Sage Publications Inc. https://doi.org/10.1177/00104140241259458

21.

Coppedge

Gerring

Altman

, et al. (2011) Conceptualizing and measuring democracy: a new approach. Perspectives on Politics 9(2): 247–267. https://doi.org/10.1017/s1537592711000880

22.

Croke

Grossman

Larreguy

, et al. (2016) Deliberate disengagement: how education can decrease political participation in electoral authoritarian regimes. American Political Science Review 110(3): 579–600. https://doi.org/10.1017/s0003055416000253

23.

Crow

(2010) The party’s over: citizen conceptions of democracy and political dissatisfaction in Mexico. Comparative Politics 43(1): 41–61. https://doi.org/10.5129/001041510x12911363510358

24.

Dalton

Jou

Shin

(2007) Understanding democracy: data from unlikely places. Journal of Democracy 18(4): 142–156. https://doi.org/10.1353/jod.2007.a223229

25.

Davis

(2022) Democracy’s Meanings: How the Public Understands Democracy and Why It Matters. University of Michigan Press.

26.

Druckman

Peterson

Slothuus

(2013) How elite partisan polarization affects public opinion formation. American Political Science Review 107(1): 57–79. https://doi.org/10.1017/s0003055412000500

27.

Eroglu

Finkel

Neundorf

, et al. (2025) Choosing democracy over party? How civic education can mitigate the anti-democratic effects of partisan polarization. British Journal of Political Science 55: e65. https://doi.org/10.1017/s000712342500016x

28.

Fossati

Coma

FMI

(2023) The meaning of democracy in Southeast Asia: liberalism, Egalitarianism and participation. In: Elements in Politics and Society in Southeast Asia. Cambridge University Press.

29.

Frempong

AKD

Amankwah

(2024) Electoral and party politics in Ghana’s Fourth Republic: ramifications of two-party dominance. In: Political Institutions, Party Politics and Communication in Ghana: Three Decades of the Fourth Republic, pp. 101–119.

30.

Friesen

(2022) Democratic Enculturation: Explaining Inaugural Party Resilience in sub-Saharan Africa. Doctoral. University of Notre Dame.

31.

Gerber

Chapman

(2018) Familiarity breeds contempt? Knowledge and understanding of democracy, support for democratization, and global city residence in Russia. Democratization 25(3): 481–503. https://doi.org/10.1080/13510347.2017.1387537

32.

Gibney

Haschke

Arnon

Pisanò

Barrett

Park

Barnes

(2021) The Political Terror Scale 1976-2021. Retrieved 2021, from the Political Terror Scale website: https://www.politicalterrorscale.org

33.

Gidengil

Stolle

Bergeron‐Boutin

(2022) The partisan nature of support for democratic backsliding: a comparative perspective. European Journal of Political Research 61(4): 901–929. https://doi.org/10.1111/1475-6765.12502

34.

Graham

Svolik

(2020) Democracy in America? Partisanship, polarization, and the robustness of support for democracy in the United States. American Political Science Review 114(2): 392–409. https://doi.org/10.1017/s0003055420000052

35.

Hale

(2011) The myth of mass Russian support for autocracy: the public opinion foundations of a Hybrid Regime. Europe-Asia Studies 63(8): 1357–1375. https://doi.org/10.1080/09668136.2011.601106

36.

Hern

(2020) Infrastructure and perceptions of democracy in Zambia: democracy off the rails. African Affairs 119(477): 604–632. https://doi.org/10.1093/afraf/adaa023

37.

Jacob

(2025) Citizen support for democracy, anti-pluralist parties in power and democratic backsliding. European Journal of Political Research 64(1): 348–373. https://doi.org/10.1111/1475-6765.12703

38.

Jöst

Lust

(2022) Receiving more, expecting less? Social ties, clientelism, and the poor’s expectations of future service provision. World Development 158: 1060008.

39.

Karlström

(1996) Imagining democracy: political culture and democratisation in Buganda. Africa 66(4): 485–505. https://doi.org/10.2307/1160933

40.

Kramon

(2016) Electoral handouts as information: explaining unmonitored vote buying. World Politics 68(3): 454–498. https://doi.org/10.1017/s0043887115000453

41.

Krishnarajan

(2023) Rationalizing democracy: the perceptual bias and (Un) democratic behavior. American Political Science Review 117(2): 474–496. https://doi.org/10.1017/s0003055422000806

42.

LeBas

(2014) A new twilight in Zimbabwe? The perils of power sharing. Journal of Democracy 25(2): 52–66. https://doi.org/10.1353/jod.2014.0035

43.

Letsa

(2019) The political geography of electoral autocracies: the influence of party strongholds on political beliefs in Africa. Electoral Studies 60: 102047. https://doi.org/10.1016/j.electstud.2019.05.001

44.

Letsa

(2025) The Autocratic Voter: Partisanship and Political Socialization Under Dictatorship. Cambridge University Press.

45.

Letsa

Wilfahrt

(2018) Popular support for democracy in autocratic regimes: a micro-level analysis of preferences. Comparative Politics 50(2): 231–273. https://doi.org/10.5129/001041518822263674

46.

Levy Yeyati

Moscovich

Abuin

(2020) Leader over policy? The scope of elite influence on policy preferences. Political Communication 37(3): 398–422. https://doi.org/10.1080/10584609.2019.1698681

47.

Lindberg

(2010) What accountability pressures do MPs in Africa face and how do they respond? Evidence from Ghana. The Journal of Modern African Studies 48(1): 117–142. https://doi.org/10.1017/s0022278x09990243

48.

Lindberg

Morrison

MKC

(2005) Exploring voter alignments in Africa: core and swing voters in Ghana. The Journal of Modern African Studies 43(4): 565–586. https://doi.org/10.1017/s0022278x05001229

49.

Chu

(2021) Understandings of Democracy: Origins and Consequences Beyond Western Democracies. Oxford University Press.

50.

Mattes

(2019) Public opinion and democratic legitimacy in Africa Routleedge Handbook of Democratization in Africa. Routledge, pp. 345–363.

51.

Mattes

Shenga

(nd) ‘Uncritical Citizenship’ in a ‘low-information’ Society: Mozambicans in Comparative Perspective. Afrobarometer Working Paper 91.

52.

Mattes

Teka

(2016) Ethiopians’ Views of Democratic Government: Fear, Ignorance, or Unique Understanding of Democracy? Afrobarometer Working Paper 164.

53.

Miller

Hesli

Reisinger

(1997) Conceptions of democracy among mass and elite in Post-soviet societies. British Journal of Political Science 27(2): 157–190. https://doi.org/10.1017/s0007123497000100

54.

Moehler

(2009) Critical citizens and submissive subjects: election losers and winners in Africa. British Journal of Political Science 39(2): 345–366. https://doi.org/10.1017/s0007123408000513

55.

Musonda

(2023) He who laughs last laughs the loudest: the 2021 donchi-kubeba (don’t tell) elections in Zambia. Review of African Political Economy 50(175): 71–89, Routledge. https://doi.org/10.1080/03056244.2023.2190452

56.

Neundorf

Öztürk

Finkel

, et al. (nd) Promoting Democracy Online: Evidence from a Cross-National Experiment. OSF.

57.

Norris

(2011) Democratic Deficit: Critical Citizens Revisited. Cambridge University Press.

58.

Osei

(2013) Party System institutionalization in Ghana and Senegal. Journal of Asian and African Studies 48(5): 577–593. https://doi.org/10.1177/0021909612465720

59.

Paget

(2024) The anti-authoritarian populisms: ideologies of democratic struggle in Tanzania, Zimbabwe and worldwide. Government and Opposition 59(3): 866–887. https://doi.org/10.1017/gov.2023.42

60.

Saikkonen

IA-L

Christensen

(2023) Guardians of democracy or passive bystanders? A conjoint experiment on elite transgressions of democratic norms. Political Research Quarterly 76(1): 127–142. https://doi.org/10.1177/10659129211073592

61.

Schaffer

(2000) Democracy in Translation: Understanding Politics in an Unfamiliar Culture. Cornell University Press.

62.

Schmitter

Karl

(1991) What Democracy Is...and Is Not. Journal of Democracy 2(3): 75–88.

63.

Shino

Smith

Uribe

(2022) Lying for Trump? Elite Cue-taking and expressive responding on vote method. Public Opinion Quarterly 86(4): 837–861. https://doi.org/10.1093/poq/nfac045

64.

Simonovits

McCoy

Littvay

(2022) Democratic hypocrisy and out-group threat: explaining citizen support for democratic erosion. The Journal of Politics 84(3): 1806–1811. https://doi.org/10.1086/719009

65.

Taylor

(2017) Ethnic politics and election campaigns in contemporary Africa: evidence from Ghana and Kenya. Democratization 24(6): 951–969. https://doi.org/10.1080/13510347.2016.1253557

66.

Teti

Abbott

Cavatorta

(2019) Beyond elections: perceptions of democracy in four Arab countries. Democratization 26(4): 645–665. https://doi.org/10.1080/13510347.2019.1566903

67.

van Binsbergen

(1995) Aspects of democracy and democratisation in Zambia and Botswana: exploring African political culture at the Grassroots. Journal of Contemporary African Studies 13(1): 3–33. https://doi.org/10.1080/02589009508729561

68.

Van Boven

Sherman

(2021) Elite influence on public attitudes about climate policy. Current Opinion in Behavioral Sciences 42: 83–88. https://doi.org/10.1016/j.cobeha.2021.03.023

69.

van de Walle

(2003) Presidentialism and Clientelism in Africa’s emerging party systems. The Journal of Modern African Studies 41(2): 297–321. https://doi.org/10.1017/s0022278x03004269

70.

Windecker

Vergioglou

Jacob

(2025) Living in different worlds: electoral authoritarianism and partisan gaps in perceptions of electoral integrity. British Journal of Political Science 55: e44. https://doi.org/10.1017/s0007123424000863

71.

Wunsch

Jacob

Derksen

(2025) The demand side of democratic backsliding: how divergent understandings of democracy shape political choice. British Journal of Political Science 55: e39. https://doi.org/10.1017/s0007123424000711

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.33 MB

Partisanship,party systems,and understandings of democracy across Africa

Abstract

Keywords

Introduction

Variation in understandings of democracy

Argument and hypotheses

Partisanship and understandings of democracy in Africa

Alternative explanation

Data and methods

Results

Main models

Interaction models

Alternative explanation: Self-sorting partisans

Data and methods

Results

Conclusion

Supplemental material

Supplemental Material - Partisanship, party systems, and understandings of democracy across Africa

Footnotes

Acknowledgements

ORCID iD

Ethical considerations

Funding

Declaration of conflicting interests

Data Availability Statement

Supplemental material

Notes

Author biography

References

Supplementary Material