JHEA/RESA Vol. 10, No.1, 2012, pp.139–170
Council for the Development of Social Science Research in Africa 2013
The Roles of Higher Education in theDemocratization of Politics in Africa:Survey Reports from HERANA1
Robert Mattes* & Thierry M. Luescher-Mamashela**
Abstract
Against the theory on the nexus of higher education and citizenship, thisarticle brings together the main findings and conclusions of three relatedstudies with African mass publics, parliamentarians from African legisla-tures, and students from three African flagship universities, conducted bythe Higher Education Research and Advocacy Network in Africa (HERANA). The article shows that higher education provides advantages in variousmeasures of democratic citizenship and leadership. It plays important roleswith regard to access to political information, information gathering skills,and levels of political knowledge; the ability to offer opinions and criticalperspectives on politics and the economy; and levels of democratic valuesand democratic action. Moreover, university-educated MPs seem to makemuch better sense of the unique complexities of legislatures and their mul-tiple competing functions than their less educated peers. This might reflectthe knowledge and analytic skills acquired through higher education, thefact that universities are themselves highly complex institutions that theyneeded to negotiate as students, and the finding that students acquire ex-tensive organisational leadership experience while at university. In light ofthis, the article suggests that higher education can play a crucial role in thedemocratisation of politics in Africa by developing “institution-builders” forstate and civil society.
Democracy in Africa Research Unit, University of Cape Town, South Africa. Email: [email protected]
Centre for the Study of Higher Education, University of the Western Cape,South Africa. Email: [email protected]; [email protected]Résumé
S’appuyant sur la théorie de la relation entre l’enseignement supérieur et lacitoyenneté, cet article rassemble les conclusions principales de trois étudesliées effectuées par le Réseau de Recherche et de Plaidoyer surl’Enseignement supérieur en Afrique (Higher Education Research and Ad-vocacy Network in Africa – HERANA). Ces études avaient pour sujets lepublic de masse africain, des membres de parlements nationaux africains, etfinalement, les étudiants de trois universités africaines de premier ordre. L’article démontre qu’une éducation supérieure offre des avantages de na-ture variée à la citoyenneté et au leadership démocratiques. En effet, elle joueun rôle important en matière d’accès aux informations politiques, lescompétences dans la collecte d’informations et le niveau de connaissancepolitique, la capacité d’offrir une opinion et des points de vue critiques sur lapolitique et l’économie, et différents niveaux de valeurs et d’actionsdémocratiques. De plus, les parlementaires qui ont un diplôme d’universitésemblent se débrouiller beaucoup mieux dans les méandres uniques desassemblées législatives et leurs multiples fonctions opposées que leurscollègues moins éduqués. Il est possible que cela soit du aux connaissanceset aux talents d’analyse acquis dans l’éducation supérieure, ou bien au faitque les universités sont elles-mêmes des institutions très complexes, qu’ilfaut maîtriser lors de ses études, en plus de l’acquisition de grands talentsd’organisation nécessaires aux étudiants pendant qu’ils sont à l’université. Par conséquent, l’article suggère que l’enseignement supérieur peut jouerun rôle crucial dans la démocratisation de la politique en Afrique enencourageant l’émergence d’individus qui vont renforcer les institutionsd’État et de la société civile. Introduction
Because democracy requires democrats, it is said to require a critical massof educated people who believe in and support democracy, and who havethe cognitive skills that enable them to act as critical citizens: sceptical ofgovernment, tolerant and trusting of other citizens, and engaged in the demo-cratic process. These arguments date at least back to Thomas Jefferson,but have since been forcefully articulated by other thinkers such as Alexisde Tocqueville, John Stuart Mill and John Dewey. Indeed, this understand-ing was part and parcel of the movement toward mass public schooling inthe United States and subsequently in Europe. More recently, scholars ofdemocracy, public opinion and political behaviour have continued to echothis basic argument and specified it more explicitly. At minimum, educationhas been argued to increase society-wide literacy levels, enabling largerproportions of the electorate to follow politics by reading newspapers. Edu-cation is also believed to stimulate greater political interest and discussion,
Mattes & Luescher-Mamashela: The Roles of Higher Education
possession of basic facts about government and the internalization of demo-cratic norms, as well as to enable people to acquire and interpret new infor-mation in a more critical manner. Finally, education is hypothesized to trans-late into improved communication and organizational skills that enable peopleto persuade and mobilize others (for an overview see Nie, Junn and Stehlik-Barry 1996). All of this is believed to translate into higher levels of politicalefficacy, which in turn leads to specific actions such as joining civil societyorganizations, contacting elected representatives and other government of-ficials, working with other citizens and participating in community actiongroups, tolerating political opponents, refraining from violent protest, andultimately supporting democracy and defending it if it comes under threat.
Most of these arguments have been corroborated repeatedly by empiri-
cal social science. While some scholars may have concluded that relativelyeducated democratic polities may not always have possessed sufficientamounts of the postulated ‘correct’ information or made the ‘correct’ con-nections between various beliefs and attitudes (Berelson, Lazafeld andMcPhee 1954; Campbell, Converse, Miller and Stokes 1960; Converse andDupeux 1972), or correctly applied abstract democratic principles to spe-cific cases (Prothro and Grigg 1960; McCloskey and Brill 1983; and seeSullivan and Transue, 1999, for an excellent summary), empirical social sci-entists have repeatedly confirmed this basic thesis at the micro-level. Moreo-ver, education has repeatedly been shown to be an important predictor ofvoting in the United States (Verba and Nie 1972; Wolfinger and Rosenstone1980; Verba, Schlozman and Brady 1995; Milligan, Moretti and Oreopoulos2004; Dee 2004) as well as in many other Western countries (Lipset 1960;Norris 2002; Dalton 2008a), though there are exceptions, often in countrieswith strong socialist parties that mobilize less educated working class voters(Powell 1986; Norris 2002; Milligan et al 2004). Education has also beenconfirmed as a strong predictor of other citizen qualities such as interest inpolitics, newspaper readership, political knowledge, interpersonal trust, tol-erance of political opponents, and a wide range of forms of political partici-pation, such as contacting elected leaders, joining community associations,attending community meetings, political internet activism, and protest both inthe United States (Hyman, Wright and Reed 1978; Hyman and Wright 1979;Bobo and Licari 1989; Verba et al 1995; Putnam 1995; Brady, Verba andSchlozman 1996; Nie et al 1996; and Burns, Schlozman and Verba 2001;Dee 2004; Dalton 2008a and 2008b) and other democracies in Europe andLatin America (Verba, Nie and Kim 1978; Milner 2002; Milligan, Moretti
and Oreopoulos 2004; Belucci, Maraffi and Segatti 2006; Magalhaes 2008;Dalton 2008). In a 1996 summary, Nie et al. concluded:
The notion that formal educational attainment is the primary mechanismbehind citizenship characteristics is basically uncontested …. Formal edu-cation is almost without exception the strongest factor in explaining whatcitizens do in politics and how they think about politics. (Nie et al 1996:2)
Education’s impact is so regularly found that political scientist Philip Con-verse (1972) once called it the ‘universal solvent’ of political participation(cited in Dee 2004:1700).
Scholarly arguments about the education-citizenship nexus in the ‘old’
democracies of advanced industrialized societies are really about the linkbetween education and the quality of democracy. In contrast, in new de-mocracies in developing countries – such as in much of sub-Saharan Africa– scholarly arguments about the education-citizenship nexus are less aboutthe link between education and the quality of democracy; rather, argumentsin our context pertain much more directly to questions about the onset andvery survivability of democracy, along with its quality. The existing literatureagain bears out the importance of education. Education has been shown toincrease voter turnout and civic engagement in several different developingworld contexts (Verba, Nie and Kim 1978; Anderson and Dodd 2005; Belucci,Maraffi and Segatti 2006; Magalhaes 2008; Lam and Kuan 2008). Moreimportantly, education has proved to be a very strong predictor of popularsupport for democracy in places like Korea, Chile, Eastern Europe, Russiaand in many countries of sub-Saharan Africa (Rose, Mishler and Haerpfer1998; Shin 1999; Markowski 2005; Bratton, Mattes and Gyimah-Boadi 2005;Rose, Mishler and Munro 2006; Mattes and Bratton 2007).
A less frequently appreciated parallel to the linkage between education
and democratic citizenship is the connection between education and demo-cratic leadership. That is, democracy also requires a smaller, but nonethe-less critical mass of educated elites who not only believe in democracy butalso have the cognitive skills that enable them to run the complex institutionsof modern democracy, primarily elected legislatures and executive depart-ments, as well as regulatory agencies, courts and police, and political partiesand civil society organizations. Institutional leaders and senior staff are re-quired to master complex documents and must often play multiple and com-peting roles. Effective democratic institutions have to establish internal struc-tures and procedures to enable them to perform core functions and managetheir often inherent tensions and contradictions, thus bringing a semblanceof predictability and certainty to the government process. Thus, while it may
Mattes & Luescher-Mamashela: The Roles of Higher Education
not require it, effective institutionalization is certainly assisted by thetypes of cognitive skills acquired through formal education, especiallyhigher education.
Because developed societies rarely lack a sufficiently large critical mass
of educated elites, scholars in these contexts often fail to address this issue. Or, if they do, they do so for very different purposes. For example, there hasbeen a large amount of descriptive research on the social background ofdemocratic legislators under the rubric of ‘legislative recruitment’, but fewstudies have focused specifically on the impact of formal education per se(for reviews of this literature see Mathews 1985; Norris 1997a). To theextent that scholars ever use education as an explanatory variable, theyhave focused on whether it helps potential candidates get into the legislaturein the first place, not on what they do once they get there (see Norris 1997b).
In a developing society, however, it is an important question whether the
education system is producing a sufficient mass of people able to run andstaff the range of institutions critical to the functioning of modern democ-racy. To our knowledge, there have been few empirical studies in this areasince the 1960s (e.g. Coleman 1965), though Barkan (2009) and his col-leagues have recently collected some initial descriptive data on the educa-tional background of African legislators. Nevertheless, the applicability ofthe ‘education for democracy’ argument in the developing world in generaland across different African polities in particular has received support froma series of macro-level studies that have demonstrated both cross-nationallinkages (countries with higher aggregate levels of education have higherlevels of democracy) and more importantly, over-time linkages (developingnon-democracies with increasing levels of education were much more likelyto become democracies than those within stagnant or declining levels ofeducation), effects that are independent of other factors such as levels ofnational wealth (Lipset 1959; Barro 1999; Glaeser, La Porta, Lopez-de-Silanes and Schleifer 2004). While some analysts argued that the correla-tion was a spurious reflection of larger historic economic and political devel-opment paths that simultaneously increased both education and democracy(Acemoglu, Johnson, Robinson and Yared 2005), further analysis re-affirmedthis link by examining longer-term lags of education on democratization andthrough better statistical controls for the mutual correlations amongst previ-ous levels of education and democracy and their over-time persistence(Glaeser, Ponzetto and Schleifer 2007; Bobba and Coviello 2007; Papaioannouand Siourounis 2008; Castello-Climent 2008).
Micro-level studies conducted in old democracies almost always reflect
attitudinal differences between university graduates and high school drop-outs (after statistically controlling for other relevant factors) and thereforedo not explicitly examine the specific impact of higher education. A fewstudies that have been done so have found that the impact of education oncertain measures of citizenship is not linear but increases once one entersuniversity. There is ample evidence that the university experience has arange of unique social and developmental impacts (see Pascarella andTerenzini 1995). Putnam (1995:667), for example, found that the relation-ship between education and civic engagement was curvilinear with advanc-ing returns as education increased. In particular:
The last two years of college make twice as much difference to trust andgroup membership as the first two years of high school. The four years ofeducation between 14 and 18 total years have ten times more impact on trustand membership than the first four years of formal education. (Putnam1995:667 emphasis in the original)
Dee (2004) also finds similar effects of college entrance on voter registra-tion, voting and volunteering. And Dalton (2008) has shown that universityeducation, when combined with high levels of political interest, makes animportant and unique contribution to the way in which voters relate to politi-cal parties and election campaigns.
Looking at related studies on this nexus in new democracies, it is not at
all clear whether the measured impacts of education reflects the impor-tance of higher education, or more simply, the difference between second-ary versus primary, or no schooling. On one hand, one of the best empiricalstudies of democratic transition in sub-Saharan Africa in the 1990s demon-strates that it was internal protest (rather than pressure from external forces,like the international financial institutions), often driven by university stu-dents, that was the key factor which brought down autocratic leaders acrossthe continent (Bratton and Van de Walle 1997). And Barkan and his col-leagues (2009) found that university-trained MPs formed the core of thecross-party coalitions that have initiated key reforms in some African legis-latures. Read together, the two studies could well be seen as evidence ofimportant role of highly educated Africans in the democratic transitions andconsolidation of democracy on the continent. On the other hand, while Bratton,Mattes and Gyimah-Boadi’s study of public opinion in 12 sub-Saharan Afri-can countries (2005) found a strong impact of education, it did not specifi-cally examine the impacts of different levels of education.
Mattes & Luescher-Mamashela: The Roles of Higher Education
Evans and Rose (2007) attempted to assess the differing impacts of
various levels of education in Malawi by creating a series of ‘dummy’ vari-ables (i.e. variables that take either the value of 0 or 1) for different levels ofschooling (primary, secondary, post-secondary). They found that each levelof education made a statistically significant contribution to popular supportfor democracy in that country. While this is encouraging, the finding is lessthan conclusive since standard dummy variable analysis is designed to com-pare a series of wholly discrete nominal categories with no overlapping orcumulative content to a referent group (in this case, those with no school-ing). But education is different. While each category certainly contains adiscrete set of respondents, the concept is not discrete. The effect of beingin secondary school, when compared to having no schooling, also includesthe effect of having been in primary school; and the effect of post-second-ary and higher education when compared to those with no schooling alsoincludes the impacts of both primary and secondary schooling. Indeed, asEvans and Rose’s (2007b) models become more fully specified, the uniquecontribution of higher education to support for democracy diminishesrapidly. The Higher Education and Democracy Studies of HERANA
It is in this theoretical context of strong and positive findings about the im-pact of education in general, and higher education in particular, on democ-racy in developed societies, but inconsistent and uncertain findings in thedeveloping world, and particularly in sub-Saharan Africa, that the HigherEducation Research and Advocacy Network in Africa (HERANA) con-ducted a series of three studies to examine differing aspects of the role ofhigher education in the democratisation of politics in Africa.2
The first study by Mattes and Mughogho (2009) used Afrobarometer
survey data collected in 2005-2006 in 18 different countries of Central, East,Southern, and West Africa, to assess whether university educated citizensof varying ages were any more engaged with and supportive of the demo-cratic process than less educated citizens.3 The study started by unpackingthe set of factors that Bratton et al (2005) called ‘cognitive awareness’ andattempted to isolate the discrete contribution of formal education in general,and higher education in particular, to each of these other factors. Second, itexamined the impact of education on a much wider range of facets of demo-cratic citizenship than previous studies using Afrobarometer data by lookingat: (1) rates of political participation, (2) the ability to formulate political
opinions, (3) basic democratic values, and (4) the willingness to offer criticalperformance evaluations.
The general applicability of the Afrobarometer findings about higher edu-
cation might be limited if the political attitudes and behaviours of currentuniversity students differ in any systematic way from their older compatri-ots. Thus, the second HERANA study led by Luescher-Mamashela andinvolving teams of researchers and analysts in Kenya, South Africa andTanzania, focused on current students at three premier African universities:the University of Dar es Salaam (UDSM) in Tanzania, the University ofNairobi (UON) in Kenya, and the University of Cape Town (UCT) in SouthAfrica (Luescher-Mamashela, Kiiru, Mattes, Mwollo-ntallima, Ng’ethe andRomo 2011). Surveys were conducted among third-year undergraduate stu-dents and student leaders in all three universities in 2009, producing a sam-ple of 400 weighted responses representative by gender and faculty foreach university.4 In addition, data from the latest round of Afrobarometersurveys (Round 4, 2008) from Kenya, South Africa and Tanzania was usedto compare the students’ responses with those of the general public and therelevant age cohorts of youth without higher education in each country.5Thus, this project explored the current role of universities in the formation ofpolitical attitudes and democratic citizenship among students and studentleaders. Because it used identical (yet indigenised) survey instruments acrossall three universities, with many questions based on Afrobarometer items, itwas possible to do statistical comparisons across universities as well asbetween each set of students, young citizens of the same age but withouthigher education, and the mass publics of each country in general (of anyage and educational level). The student surveys focused on students’ atti-tudes towards democracy, their political behaviour, and their perceptionsand conceptions of politics and governance on campus as well as with re-spect to national government; in addition, they investigated the relationshipbetween students’ political engagement and their attitudes towards democ-racy. The study came to explore, in a wider sense, whether the three univer-sities respectively provided a condition of a ‘political hothouse’ or of a ‘trainingground’ for democratic citizenship and leadership, and thus serving as ‘sitesof citizenship’.
In the third HERANA study, known as the African Legislatures Project
(ALP), Mattes and Mozaffar (2011) examined the extent to which Africanparliaments are composed of university educated members of parliaments(MPs), and whether university educated legislators approached their jobdifferently to other legislators. This project therefore explored the ability of
Mattes & Luescher-Mamashela: The Roles of Higher Education
national systems of higher education to supply the human capital to run thenational legislatures in selected African countries. The study used a combi-nation of primary and secondary sources, including interviews with MPs. Intheir report, Mattes and Mozaffar attempted to subject these broad proposi-tions to more systematic test, using a unique and original set of data fromsurveys of randomly chosen, representative samples national legislators ineleven countries carried out by ALP between 2008 and 2010.6 The research-ers asked three broad sets of questions for this data. The first was a simpledescriptive question: What is the overall level of educational attainment (par-ticularly higher education) amongst African MPs, and how does this varyacross countries? They then move on to two larger explanatory questions. They investigate, briefly, what national level characteristics might accountfor these cross-national variations? They then move on to examine, at greaterdepth, the actual political consequences of education amongst African legis-lators. Do more educated MPs possess different social and political charac-teristics than less educated ones? And are highly educated MPs, in fact,more likely to adopt the types of attitudes and behaviours that might expandthe role of representative assemblies in their national political systems, andthus contribute to the process of democratization (Fish 2006)? Specifically,they ask whether highly educated MPs are more likely to bring with themmore professional skills and government experience, adopt an ‘institutional-ist’ role orientation, devote greater proportions of their time to activities thatbuild legislative institutions, show greater independence from party bosses,and support legislative reform measures. Higher Education in Sub-Saharan Africa
The very first thing one needs to take into account when considering theroles of higher education in the development and consolidation of democ-racy in Africa is the relative scarcity of highly educated people on the con-tinent. This is reflected in the Afrobarometer data. Thus, in response to thequestion, ‘What is the highest level of education you have completed?’ theAfrobarometer data reveal that less than one in ten (9%) of the 21,600adults interviewed across 18 countries in 2005-2006 had advanced beyondhigh school, either obtaining post-secondary non-university qualifications orattending a university. Just 2 per cent said they had actually completed anundergraduate degree. At the same time, it is also important to consider thatlevels of post-secondary education vary widely across countries, rangingfrom around one-in-five Nigerians (23%), Zambians (19%), and South Afri-cans (17%) to less than one in every twenty adults in Lesotho, Madagascar
and Benin (4% each), Mali and Mozambique (3%) and Malawi and Tanza-nia (1%). Possession of actual university degrees ranged from 4 per cent inNigeria to less than 1 per cent in many countries according to the surveydata. Given the rapid expansion of education systems across the continentin less than fifty years, the possession of a formal education – as of 2005-2006 – decreases sharply with age, as well as amongst rural people andamongst women. (Mattes and Mughogho 2010)
The ALP surveys also looked at highest education attainment but regis-
tered extremely different results. Across 11 of the same countries includedin the Afrobarometer study, voters elected substantially larger proportionsof educated and highly educated legislators: the majority of the randomlyselected samples of MPs (59%) had at least completed an undergraduateuniversity degree; 19 per cent had completed an undergraduate degree, and15 per cent had either a post-graduate diploma or honours degree. One-in-five had a Masters degree (20%) and 3 per cent had doctorates. As illus-trated in Figure 1, these levels of higher education attainment amongst MPsare anywhere from 15 to 80 times higher than amongst ordinary citizens intheir respective countries (for details see Mattes and Mozaffar 2011). Thisfinding in itself tells us something about the ability of political parties to re-
Figure 1: Educational Attainment amongst African MPs and Citizens
Mattes & Luescher-Mamashela: The Roles of Higher Education
cruit candidates out of the highest social strata in these countries. Further-more, that so many voters are prepared to vote for candidates with farhigher levels of education than themselves tells us something about the po-litical culture in these countries and the huge responsibility this political cul-ture bestows on higher education and its role in the democratisation of poli-tics in Africa.
Cross-nationally, the highest levels of education were found in Ghana
(where 65% had some sort of post-undergraduate diploma or degree andalmost half had a Masters degree or higher), followed by South Africa (55%),Nigeria (51%), Uganda (50%), and Kenya (49%). On the other side of thespectrum, less than one-in-five MPs had postgraduate experience in Ma-lawi (19%), Mozambique (12%) and Lesotho (7%). Unfortunately, we haveno data with which to assess whether these levels of educational attainmentconstitute an advance over time. However, we do know that since 1990 –the beginning of Africa’s recent wave of democratisation – the proportionof higher education enrolments has increased in most African countries forwhich data is available. It is reasonable to assume that the ALP results alsoreflect an upward curve.
Perhaps surprisingly, given the various advantages enjoyed by incum-
bent political parties in Africa, governing parties were no more likely to havemore highly educated MPs than were opposition parties. Moreover, com-pared to the gender differences in educational attainment amongst the masspublic, we find no statistically significant differences in the level of educa-tion attained by male and female MPs, neither is there any evidence thatyounger or urban-based MPs are any more likely to be highly educated. There is fairly strong evidence, however, that highly educated MPs are morelikely to come from a professional background, and that they are less likelyto represent geographically far flung constituencies distant from the nationallegislature. Highly educated MPs are also more likely to have held a previ-ous position at some level of national government before they were electedto parliament, usually by working in a ministry, agency or parastatal. How-ever, they do not differ with respect to experience in local government. Finally, highly educated MPs are much less likely to have held any type ofleadership position within their political party than less educated ones. Thisbegins to set up an interesting paradox. Beyond the cognitive skills devel-oped through the educational process, highly educated MPs bring with themgreater familiarity with the types of organizational and business skills thatcome with higher education and a professional background, as well as greaterfamiliarity with various facets of national government. However, they have
no significant advantage in terms of knowledge of, or familiarity with locallevel politics, and are at a significant disadvantage when it comes to partypolitics. Given the powerful role of political parties in structuring and organ-izing legislative activities, the very advantages brought by highly educatedMPs in terms of policy relevant expertise might be negated by their com-parative lack of rootedness in the political parties to which they belong (seeMattes and Mozaffar 2011). Higher Education, Information and Information Gathering
To recall, the main arguments about the contribution of education to democ-racy revolve in the first instance around citizens’ and leaders’ grasp of cru-cial information, cognitive engagement with the democratic process, and theability to gather and process new information, through for example, the newsmedia. At the mass level, the Afrobarometer study revealed that citizenswith higher levels of formal education did indeed possess far higher levels ofpolitical information (which is defined as the extent to which respondentswere able to correctly answer three questions about the identity of politicalleaders, and three questions about their constitutional and governmental sys-tem). They were also more likely to use news media, thus constantly acquir-ing new factual information. These impacts remained even after Mattesand Mughogho (2010) controlled for a range of demographic covariates ofeducation (such as age, rural-urban status, gender, and levels of householdpoverty). However, when these covariates were taken into consideration,they found that formal education offered no real advantage in terms of in-creasing citizens’ levels of cognitive engagement (measured as the combi-nation of political discussion and political interest) or in terms of politicalefficacy (i.e. the extent to which people felt they could influence other citi-zens).
Thus, the most direct impact of formal education in Africa on citizenship
is through the stimulation of news media use and by giving citizens the skillsto accumulate basic facts about the political system through the news me-dia. This means that formal education may also have an indirect impact onother elements of democratic citizenship flowing through greater newsmedia use and higher level of factual knowledge about politics. Conse-quently, Mattes and Mughogho (2010) examined all subsequent ‘downstream’effects of education by looking at its combined direct and indirect effects,calculating a block adjusted R2 estimate for only formal education, newsmedia use, and political information (which strips out the overall explanatorycontributions of non-cognitive elements of age, gender, urban residence, or
Mattes & Luescher-Mamashela: The Roles of Higher Education
poverty). They initially found that average rates of news media consump-tion increased sharply and steadily across all increments of education, in-cluding higher education. However, they saw the first hints of a limited im-pact of higher education, with regard to the possession of political informationwhich also increased sharply with education but levelled off amongst uni-versity attendees.
And as discussed above, Mattes and Mughogho attempted to isolate the
specific impact of higher education by analyzing only those respondentswith at least a high school education, and then comparing those who neverwent beyond high school with those who possessed some form of post sec-ondary qualification and with those who actually went to university (holdingconstant for other covariates of higher education). Once this was done, theyfound very minimal effects: those who went on to university education wereonly slightly more likely to use news media or know basic facts about thepolitical system than ordinary school-leavers. And because these weak re-lationships meant that higher education was likely to have little indirect im-pact on democratic citizenship through increased news media use or politi-cal information (or for that matter, through increased cognitive engagementor efficacy), all subsequent analyses focused only on the direct impacts ofhigher education (holding constant news media use, political information,cognitive engagement and efficacy, as well as age, rural/urban location, gen-der and poverty).
The Afrobarometer study also examined whether education facilitated
increased citizen ‘articulateness’ (defined simply as the ability to provideanswers to survey questions). Mattes and Mughogho (2010) found, aftercontrolling for associated variables, that formal education indeed substan-tially increased respondents’ ability to offer opinions to fieldresearchers on arange of issues of political and economic performance. Taken together, higherlevels of education, news media use, and political information enhanced citi-zens’ ability to offer (positive or negative) opinions about the performanceof the political system, preferences about democracy versus alternative re-gimes, and a range of social and political values, as well as provide somemeaning to the word ‘democracy’. However, the authors again observed acurvilinear trend: that is, respondents’ ability to provide fieldresearchers withevaluations of political and economic performance rose sharply across in-creasing levels of education until one graduated from high school, or ob-tained some form of non-university post-secondary qualification, but lev-elled off after that. Indeed, multivariate analysis demonstrated that higher
education made no significant contribution to respondents’ ability to offeropinions on these matters (see Table 1). Table 1: Formal Education, Cognitive Awareness and Articulateness
At Least Some University Total Adjusted R2 “Block R2“ N
Able to Offer Opinions on Government Performance
Controlling for age, rural/urban location, gender and lived poverty, as well as cognitive en-
NS p=>.05, * p=<.05, ** p=<.01 *** p=<.001 Source : Mattes and Mughogho (2010:17).
In contrast to the Afrobarometer study which examined respondents of vary-ing ages who might have attended university several years ago, the surveysby Luescher-Mamashela et al (2011) focused on current students, many ofwhom have grown up in a context of democratic governance, a moderneducational system and with the advantages of access to new informationand communication technologies. On the one hand, Luescher-Mamashelaand his colleagues found that the students at the universities of Cape Town,Dar es Salaam and Nairobi were not necessarily more interested in politicsthan their fellow citizens (as measured by Afrobarometer Round 4 results);on the other hand, the three student surveys clearly indicate that the stu-dents in all three universities discussed politics far more frequently thaneither their fellow citizens in general, or youth of the same age cohort butwithout higher education of their respective country. Students also use agreat diversity of news media (radio, television, newspapers and internet) ata level at least equal to or above that of their respective mass publics(Luescher-Mamashela et al. 2011).
While UCT students read newspapers at about the same rate as South
Africans in general, or their cohort of 20-23 year olds with no higher educa-tion, newspaper readership at Dar Es Salaam and Nairobi universities faroutpaces that of their respective national publics or less educated agecohorts. Moreover, despite the mushrooming of internet cafes across thecountries, internet use as news media remains almost entirely a studentprivilege: eight-in-ten students at all the three universities say they haveaccess to and use the internet daily or several times a week, compared tojust one-in-ten ordinary citizens in their respective country. Even among the
Mattes & Luescher-Mamashela: The Roles of Higher Education
relevant age cohort without higher education, internet use is not much higherthan among the three mass publics in general. Thus, on all the three cam-puses, access to and acquisition of information about public affairs and poli-tics is considerably better and more frequent than among the relevant pub-lics in general and the less educated same age peer groups in their respectivecountries.
In terms of actual knowledge about politics, UON students are highly
knowledgeable about the identity of political incumbents at national level(i.e. the President of the country is correctly identified by 98% of the stu-dents, the Minister of Finance by 96%, and their local MP by 87%) and to alesser extent of student representatives and campus officials (e.g. the Vice-Chancellor, 83% correct responses, the President of the student body 82%,and the Dean of Students, 52%) followed in rank order by UDSM and UCTstudents. Looking closely at responses to the question about the minister offinance in comparative perspective, it shows that the UON and UDSMstudents are much more knowledgeable about this specific political incum-bent than their compatriots. Compared to only 12 per cent of Tanzanians, 60per cent of UDSM students correctly name their Minister of Finance; inKenya only 44 per cent of Afrobarometer respondents get the Minister’sname right as against 96 per cent of the UON students. In South Africa incontrast, 76 per cent of South Africans identified the longstanding and popu-lar minister of finance correctly in the 2008 Afrobarometer survey, but only55 per cent of the UCT students correctly named the former minister or hissuccessor.7 A similar pattern emerges with regard to students’ knowledgeabout key political institutions at national level (such as term limits for presi-dents) and knowledge about university governing bodies. The number ofyears the president can constitutionally hold office is known by 63 per centof students at UCT, 72 per cent at UDSM and 84 per cent at UON. Therole of the courts in determining the constitutionality of a law is known tohalf of the students at UCT, but to less than a third of UDSM and UONstudents.
Initial analyses of Afrobarometer data (e.g. Bratton et al 2005) placed a
great deal of emphasis on the role of respondents’ own understandings ofdemocracy. The HERANA student surveys found that more than nine-in-ten students could provide a valid definition of democracy in their own words,and that almost all definitions carried a positive connotation. In much thesame patterns as the overall population, close to half of the students (47%)defined democracy in terms of political rights and freedoms; and one-third(34%) as popular participation and deliberation in politics. Conversely, less
than one-in-ten provided definitions mentioning ‘equality’, ‘fairness’, ‘jus-tice’, ‘rule of law’ or ‘good governance’. Moreover, conceptions of democ-racy as ‘socio-economic development’ or ‘access to basic services’ werealmost completely absent from students’ definitions (1%). In other words,while the stability and very survivability of multi-party democracy in manyAfrican countries is often under question (and local variations are usuallyassessed in the literature in terms of ‘deficiencies’ of political culture), Afri-can mass publics and the students at the elite universities surveyed byHERANA can be trusted to know very well what democracy is, and is not(see Luescher-Mamashela et al. 2011).
Cognitive engagement through access to information and news media
use were among the starting points of both Mattes and Mughogho’sAfrobarometer and Luescher-Mamashela et al’s student studies. Conversely,the starting point for the ALP legislators study was to assess MPs’ ‘roleorientations’ or how elected representatives in African legislatures under-stand and define their job. Combining answers to several open and close-ended relevant questions, the researchers created valid and reliable com-posite indices that measure three distinct role orientations: ‘institutionalists’,‘constituency servants’ and ‘partisans’ (Mattes and Mozaffar 2011).8 Theyfound that the highly educated MPs of African parliaments were more likelyto see themselves as having an institutionalist role than MPs who had notcompleted high school. However, they were no different from other MPs interms of the extent to which they saw themselves as constituent servants. Reflecting the fact that they were less likely to have had any experience inparty organizations prior to entering the legislature, it turned out that highlyeducated MPs were far less likely to see themselves as strong partisansdoing the bidding of the political party. These initial bivariate associationsremained even after the application of controls for a range of other poten-tially important variables such as party membership, age, gender, constitu-ency characteristics, background experiences, and experience within thelegislature. In fact, after controlling for these variables, formal educationconstituted the single strongest explanation of why MPs adopt a more insti-tutionalist or less partisan orientation.
While the ALP study did not include any test of political information or
news media use, it did measure the extent to which MPs used various mecha-nisms to gather information necessary to their job. What Mattes and Mozaffarfound was that highly educated MPs were indeed more likely to use theinternet on a frequent basis, use various resources internal to the legislaturesuch as the parliamentary library, research staff, committee staff or legal
Mattes & Luescher-Mamashela: The Roles of Higher Education
counsel, and would more likely turn to information resources external to thelegislature, such as civil society organizations or universities, than their lesseducated peers. Highly educated MPs were also more likely to have trav-elled outside of Africa on fact-finding trips. With one exception (using inter-nal resources), these correlations remained even after the application of therelevant statistical controls for demographic and experiential factors androle orientations. The authors also noted that education had an additionalindirect effect via role orientations: that is, education promoted an institu-tionalist role definition, which in turn also separately and independently pro-moted the use of the internet and other internal parliamentary resources;and education reduced partisan role orientations, which in turn promote internetuse but substantially detract from the use of internal resources. Therefore,
Figure 2: Committed Democrats Source: Luescher-Mamashela et al. (2011:57)
the study showed that, as with students, better educated MPs not only startout with increased factual knowledge, but are also more motivated and bet-ter able to gather additional information than their less educated peers. Higher Education and Democratic Values
Replicating findings based on earlier rounds of the Afrobarometer, Mattesand Mughogho (2010) confirmed that education had a positive and sizableimpact on commitment to democracy (measured as consistent preferencefor democracy and rejection of presidential dictatorship, military rule andone-party rule), both directly and indirectly, through news media use and
political information, each of which also had a positive impact. They alsoexamined the link between education and a range of other democratic val-ues, measured as support for various democratic qualities such as criticalcitizenship, the rule of law, freedom of expression, political equality, genderequality, bureaucratic and electoral accountability, as well as opposition tocorruption. Again, they found that education had a notable impact on peo-ple’s stated willingness to demand accountability from intransigent bureau-crats. Yet, across the rest of these values, the total impact of education (aswell as of news media use and political information) was negligible to non-existent in statistical terms. While formal education generally increased theextent to which citizens supported these democratic principles, the overallsize of the impact was almost always very small. Indeed, more educatedrespondents were less likely to favour a universal, equal franchise, or tosupport the rule of law.
In general, cognitive factors seem to have very little to do with the ques-
tion whether or not Africans held democratic values or predispositions. Com-pared to school leavers, those who had attended university were very slightlymore likely to support freedom of expression. But they were less likely tobelieve that everyone should have an equal vote, or to say that elected lead-ers should be governed by public opinion (rather than their own beliefs). Overall, it turned out that there was virtually no difference between highschool graduates and those who had attended African universities across abroad range of democratic values.
Consistent with the Afrobarometer findings, the student surveys found
that only a minority of students at UDSM (36%) and Nairobi (45%) couldbe described as fully committed democrats (e.g. who always prefer democ-racy and reject each non-democratic alternative), and in the national com-parison the students from these two universities also emerged as less com-mitted to democracy than their respective national age cohorts of youth withno higher education, as well as their mass publics in general (see Figure 2). Only at UCT and in South Africa was the picture somewhat different. 54per cent of UCT students qualified as committed democrats who consist-ently demand democracy; this figure is considerably higher than the SouthAfrican mass public (where only 35% were fully committed to democracy)or their age peers without HE (32%). Notwithstanding this, demand for keyfreedoms such as free speech, press freedom, and freedom of association,is high among students of all campuses (and highest at UCT), albeit not asunfettered freedoms: on the one hand, the majority of students reject or
Mattes & Luescher-Mamashela: The Roles of Higher Education
even strongly reject all statements that suggest government should be ableto curtail free speech, freedom of association or press freedom; on the otherhand, the corresponding positive statements that suggest unfettered freedomsreceive varied levels of support. Provided that the students more stronglydisagree with statements that limit political freedoms than they agree withunfettered rights to free speech, free association and press freedom, theirresponses suggests a rather nuanced demand for these freedoms compat-ible with students’ conceptions of democracy (as noted above), wherebydemocracy is not simply conceived as a political system of rights and freedomsbut one which also involves citizen participation in decision-making (andperhaps other kinds of citizen responsibilities and citizenship duties)(Luescher-Mamashela et al. 2011).9
Higher Education and Critical Citizenship
If highly educated African citizens are not especially likely to hold more pro-democratic dispositions, does higher education at least contribute to the like-lihood of citizens giving critical evaluations of the political and economicperformance of their country? The analysis of Afrobarometer data con-ducted by Mattes and Mughogho examined the impact of formal education,along with news media use and political information, on a wide range ofpolitical and economic performance evaluations. Three important findingsemerged. First, taken together, formal education, media use and politicalinformation had sizeable impacts on how those people with opinions evalu-ated their personal economic conditions, the status of political rights in thecountry, and the degree of trust they placed in government institutions andstate media. Education had an especially strong direct effect relative to thatof information and news media. Indeed, with one exception, formal educa-tion consistently had a statistically significant and negative impact on allperformance evaluations. Thus, increased schooling not only enables Afri-cans to offer more opinions, it also allows them to offer more critical opin-ions. At the same time, the contributions that formal schooling made towardenabling a more critical citizenship could potentially be mitigated by the con-founding effects of higher levels of political information and, sometimes, theeffects of news media use. That is, while education (holding constant mediause and information) made people more critical of performance, formal edu-cation simultaneously led people to acquire greater amounts of political in-formation, which in turn (holding constant education and news media use)
made people consistently less critical, and thus more forgiving of govern-ment performance. Moreover, while higher levels of news media consump-tion sometime induced greater criticalness, it more often had the oppositeeffect of making people less critical (possibly either because high mediausers were consuming news produced largely by state-owned media or be-cause increased levels of information decreased mistrust). The most con-sistent impacts of higher education (over and above the effect of going tohigh school) were to be seen in this area. Controlling the other elements ofcognitive awareness and demographic factors, Mattes and Mughogho found(with one exception) that university attendees were consistently more criti-cal of the performance of their economies, governments and political re-gimes. At the same time, the size of the impact was quite limited, however.
Consistent with these findings, the student surveys found that the major-
ity of third year students at all three institutions felt that their countries wereeither ‘not a democracy’ or a ‘democracy with major problems’ (UON,86%; UDSM, 66%; and UCT, 52%); results indicating that far more stu-dents were critical of regime performance than their age peers without HEor the respective mass publics. The majority of students at the two EastAfrican universities were also dissatisfied with the way democracy workedin their country (UON 87%; UDSM 70%). As against this, 58 per cent ofKenyans in general and 57 per cent of the Kenyan age cohort without HEwere not satisfied with regime performance (i.e. the way democracy workedin their country), and only 29 per cent of Tanzanians and 30 per cent of therespective age cohort without HE in Tanzania. At UCT, a majority of stu-dents was ‘fairly’ or ‘very satisfied’ (57%), a figure that was slightly morepositive than South Africans in general (49%) and their South African peerswithout HE (44%). Furthermore, Luescher-Mamashela and colleagues makea useful contribution by developing the concept of the potentially‘transformative democratic citizen’: that is, someone who prefers democ-racy, but is critical or highly critical of the current extent of democracy andimpatient to see regime change. As may be expected from the political de-velopments in Kenya at the time of the survey, a majority of third-yearstudents at UON (61%) emerge as potentially transformative democrats,compared to just under half at UDSM (47%) and four-in-ten at UCT (40%). In each case, the students were significantly more likely to be critical andimpatient transformative democrats than their respective fellow citizens ortheir age peers without HE (see Figure 3).
Mattes & Luescher-Mamashela: The Roles of Higher Education
Figure 3: Tranformative Democrats Source: Luescher-Mamashela et al. (2011: 80).
In quite concrete terms, the ALP study adds to the finding that highly edu-cated MPs tend to be more critical of the performance of the legislature,that they are also more likely to demand various kinds of reforms. Thesurvey asked MPs whether they would support or oppose a wide range ofpotential reform bills, ranging from enabling the legislature to initiate legisla-tion, amend executive bills, raise and lower taxes, pass bills over the objec-tion of the executive, and increase the amount of executive oversight. Whilehigher education made no direct impact, Mattes and Mozaffar detected astrong indirect impact because MPs with institutional orientations, as well asthose MPs who actively use parliamentary resources to gather information,were much more supportive of reforms to strengthen the legislative arm ofgovernment over that of the executive. Higher Education and Democratic Action
In response to more participatory conceptions of democracy and variousconceptualisations of ‘active citizenship’, each study examined in some waythe degree to which higher education translated into relevant forms of demo-cratic action. The Afrobarometer study demonstrated that formal educationmade positive, statistically significant but small contributions to membership
in civic groups, as well as contacting formal and informal leaders by citi-zens. Moreover, any positive contribution that education made to demo-cratic participation appeared to level off after high school. Furthermore,when it comes to partisan identification, higher education had negative ef-fects: African citizens with a university degree were less likely than highschool graduates to identify with a political party. And while they were morelikely to take part in protest and contact formal officials, the absolute size ofthe difference was relatively small. Lastly, higher education played no rolein encouraging people to join civil society organizations, become involved incommunity affairs, or vote.
Active organisational membership, formal and informal political action
and voting were also investigated for those being in higher education in theHERANA student surveys. In terms of voting, the students at Nairobi (79%)and UCT (62%) both reported having voted in the most recent nationalelection at rates about equal to their less educated age cohort, while UDSMstudents reported voting at lower rates. However, active organisationalmembership in an off-campus, non-religious association was much higheramong students at UCT (43%) and UDSM (53%) and at least slightly higheramong UON students (48%) than amongst their respective national agecohorts (11% in South Africa; 29% in Tanzania; 43% Kenya). Moreover,students were more likely to be leaders of off-campus voluntary secularorganisations than their respective age cohort without HE (29% of UONstudents as against 12%; 15% of UDSM students versus 1% only; and 13%of UCT students compared to 4% of the South African age cohort). Inaddition to that, students are of course extensively involved in campus-basedstudent organisations whereby 63 per cent of students at UON, 71 per centat UDSM and 57 per cent at UCT reported active membership or leader-ship of an officially recognised student organisation on campus. And exceptfor UCT students, who participated in national demonstrations (17%) aboutas frequently as South Africans in general, students from the East Africanuniversities were around twice as likely to take part in an off-campus pro-test as their respective compatriots (Nairobi: 28%; UDSM: 36%).
On the basis of these and related findings, Luescher-Mamashela and
colleagues concluded their investigation by developing the concept of the‘active democratic citizen’ for the purpose of the student surveys: that is,someone who always prefers democracy and either participates in protestsor demonstrations on or off campus or acts in a formal capacity as an offi-cial leader/leader of an association on or off campus. By this definition, theactive citizens represent just over one-third of the final year student body at
Mattes & Luescher-Mamashela: The Roles of Higher Education
UDSM (35%), one-quarter at Nairobi (27%) and one-fifth at UCT (22%). The surveys showed that, compared to their respective national age cohortsof youth without HE, students were considerably more likely to be activecitizens at an early age (Tanzanian cohort: 16%; Kenyan cohort: 14%; SouthAfrican cohort: 8%). Therefore, while students do not necessarily preferdemocracy more than the mass publics or their less educated age peers, thebig difference that emerges from the surveys is that those students who doso are much more likely to protest and take leadership positions in organisa-tions on or off campus.
The relevant forms of participation and pro-democratic action for MPs
are different than ordinary citizens, of course. Thus, the ALP study askedMPs rather about a range of relevant legislative behaviours. First of all, theyexamined how much time MPs allocated amongst the activities that under-pin competing key legislative functions: committee work (oversight), con-stituency work (constituency service and representation), plenary work (law-making) and party work (see Table 3). They found few linkages betweenhigher education and time allocation with regard to committee work, con-stituency work, or plenary work. However, controlling for role orientationsand other demographic and experiential factors, highly educated MPs –reflecting their reduced party backgrounds and partisan role orientations –were far less likely to devote time to party work. But they also found animportant indirect impact of formal education through institutionalist andpartisan role orientations. MPs who see themselves as institutionalists weremore likely to devote time to committee and plenary work, and less likely todevote time to party work. In contrast, those who see themselves foremostas representatives of their party were less likely to devote time to constitu-ency work and far more likely to devote time to party work. Lastly, becauseof their central importance to democratic politics and to neo-patrimonialismin African politics, the ALP survey devoted another set of questions to MPactivities in their constituency. Focusing only on the frequency with whichMPs travelled to the constituency when the legislature was in session, andhow long they stayed, they found that highly educated MPs (controlling forother factors) travelled home substantially less often, and stayed fewer dayswhen they did travel home. Table 3: Explaining MPs’ Time Allocation
Controlling for age, gender, and party. Source: Mattes and Mozaffar (2011: 12). Conclusions
Taken together, the findings of the three HERANA studies suggest thatAfrica’s schools and universities have paid democratic dividends. Based onattitudes and values measured across 18 countries in 2005-2006, the analy-sis of Afrobarometer data demonstrates several important effects of formaleducation on citizens. Its findings are in parts confirmed, in parts nuancedand put into perspective by the student surveys conducted at three premierAfrican universities and the ALP studies with MPs of eleven national legis-latures in Africa that completed the set of HERNANA democracy studies. Mattes and Mughogho’s study (2010) has shown that formal education bothenables and stimulates Africans to make greater use of the media to getnews about politics. It facilitates citizens’ acquisition of the basic informa-tion that allows them to make sense of the larger political system. Citizens inAfrica with higher levels of schooling are generally also more likely to de-velop preferences and adopt more readily critical stances toward regimeand government performance. Finally, higher levels of schooling also leadAfricans to demand democracy. However, beyond a preference for democ-racy over other regimes, educated Africans are not any more likely to holdother democratic orientations. Education also makes only a limited contribu-
Mattes & Luescher-Mamashela: The Roles of Higher Education
tion to political participation – except in the case of students who are cur-rently in higher education, as the student surveys show.
The Afrobarometer study also shows that most of these impacts are
largely a reflection of the great differences between those with little or noformal education and those who have finished high school. African citizenswho go on to university show few advances in various measures of demo-cratic citizenship beyond that displayed by high school graduates. They dis-play few statistically significant, and even fewer substantively importantdifferences with high school graduates in terms of political information, newsmedia consumption, political participation, articulateness or pro-democraticvalues. It is only in the area of evaluations of the performance of the economy,the government and the larger democratic regime that African citizens withhigher education are significantly more critical, though the size of the differ-ences are small in the mass public survey (but substantial in the comparisonbetween students and non-students of the same age cohorts).
Having started the HERANA investigations into the nexus of higher edu-
cation and democracy with the Afrobarometer study, Mattes and Mughoghoconcluded in 2010 with an observation and a question. The observation isthat even with the enormous challenges faced by Africa’s schools, studentswho move up the educational ladder and complete high school manage toacquire more political knowledge, develop firmer opinions, and adopt morecritical perspectives. The question that emerged, however, was why do wesee so little further democratic dividends amongst those who have managedto get a higher education? The results of the student surveys gain somepurchase toward addressing these questions. While current students at theuniversities of Cape Town, Dar es Salaam and Nairobi are no more inter-ested in politics than their fellow citizens, nor more demanding of democ-racy, they do exhibit a range of other important democratic advantages. They display far higher levels of political discussion and news media use(with huge advantages in terms of internet use). They display high levels ofpolitical knowledge. The students also tended to have higher rates of activemembership in off-campus secular organization (over and above associationalinvolvement on campus), as well as far higher self-reported rates of organi-zational leadership. Lastly, students are also far more critical of the per-formance of democracy in their respective country. Indeed, Luescher-Mamashela et al (2011) show that students at these universities are farmore likely to be dissatisfied democrats eager to see pro-democratic regimechange (or what the authors call ‘potentially transformative democrats’), as
well as active democrats, who combine support for democracy with formaland/or informal political participation.
These effects are not only visible through a comparison of the students
with their respective national publics, but also by a comparison of studentswith fellow nationals of the same age cohorts but who have not been ex-posed to higher education. This second comparison is important because itshows that it is not youthfulness that accounts for the more critical andactivist orientation of students in politics, and suggests that it is rather whatthey bring to these universities, or what occurs at these universities, thataccounts for these differences. Hence the authors concluded by posing twoalternative interpretations of the data, neither of which can be fully corrobo-rated with the existing data. On one hand, the increased availability of newsmedia, especially internet, increased frequency of political discussion andample opportunity for participation and leadership in campus organizationsmay mean that these universities effectively function as political ‘hothouses’whereby the high levels of intense involvement in university and public af-fairs may wither and disappear once students leave the peculiar environ-ment offered by the university. On the other hand, the findings also indicatethe potential of universities acting as effective ‘training grounds’ for demo-cratic leadership with potential ‘spill-over’ effects into off-campus politicalactivity and a more critical outlook on politics.
The broader interpretation of these findings, that current students at these
three African elite universities display a number of democratic advantagesover fellow nationals in general and those of the same age cohort but with-out higher education, confronts a range of questions. That is, should weemphasize the ‘recency’ aspect of these findings and conclude that the roleof higher education in citizenship development is improving in Africa (e.g. interms of students’ exposure to key democratic values, skills, and practices;increased access to information through the internet and other new tech-nologies)? Or should we focus on the ‘elite’ angle and conclude that theseeffects are peculiar to either the unique characteristics of the students re-cruited into these universities and/or the type of teaching that these institu-tions have always offered? The available evidence does not permit us tomake a conclusive decision at this point but it opens up exciting new leadsfor further enquiry. For instance, the HERANA findings suggest furtherinvestigations into the practical ways in which African universities alreadyrealise their potential as ‘training grounds for democracy’ using a broadrange of methodologies.
Mattes & Luescher-Mamashela: The Roles of Higher Education
Mattes and Mozaffar (2011) of the ALP surveys uncover perhaps the
strongest impacts of higher education of the three studies. Across membersdrawn from eleven legislatures in sub-Saharan Africa, they demonstratethat highly educated MPs bring with them important social and political char-acteristics and experiences that enhance their performance as effectivelegislators (though their lack of partisan background may reduce their ef-fectiveness by isolating them politically). University educated MPs are morelikely to see themselves as representatives of the interests of their legisla-tive institutions, rather than their constituents or political party. In turn, bothformal education and its consequent role orientations combine to lead highlyeducated MPs to devote more time to activities that advance the interest ofthe institution, to some degree in plenary session but much more importantlyin the committee, and less time toward their party and constituency. Theyare also more likely to use the resources of the institution to gather addi-tional information about bills and budgets, and use that information to departfrom under the yolk of their party leaders on occasion. Finally, these twoforces also combine to make highly educated MPs the prime constituencyof legislative reform and legislative strengthening.
The university experience seems to enable these MPs to make sense of
the unique complexities of the legislature, which require MPs to balancemultiple competing legislative functions. This might reflect the knowledgeand analytic skills acquired through a university degree, and especially apostgraduate degree. Or, it may reflect the fact that universities are them-selves highly complex organizations and that negotiating undergraduate andpostgraduate degrees provides graduates with key advantages in handlinglife in the institutions where they work, including legislatures. Or it may be,in effect, of the extensive organisational membership and leadership experi-ence that students acquire on and off campus reported in the student sur-veys. Again, current evidence cannot answer these questions conclusively;we can just assume that it is most likely a combination of all the above.
The HERANA surveys have therefore provided important evidence on
the roles of higher education in the deepening of democracy in Africa. Onthe one hand, highly educated citizens in Africa are significantly more criti-cal of the performance of their economy, government and larger democraticregime, and they are better informed and obtain their information about poli-tics from a greater variety of news media than less educated citizens. Thesame democratic advantages are already evident among students at univer-sity level. On the other hand, higher education has seemingly no positiveimpact on support for democracy per seand higher levels of political partici-
pation observed among students disappear once the graduates leave theuniversity. In our view, if political participation can be sustained and demo-cratic values are more successfully inculcated at university level, Africangraduates may come to play the crucial role we observe already amonguniversity educated parliamentarians in African legislatures: namely, thatthrough their commitment to democracy and knowledge of politics, the ana-lytic skills acquired through a university degree and insight gained into theoperation of complex institutions, African graduates may come to act aseffective democratic ‘institution builders’ in state and civil society and thusplay a crucial role in the democratisation of politics in Africa.
1. We would like to acknowledge the support of the US Partnership for Higher Education
in Africa - in particular the Ford Foundation, the Carnegie Corporation of New York,the Rockefeller Foundation and the Kresge Foundation – which contributed fundingand expertise for the HERANA studies. We are also grateful to the Afrobarometer foravailing data and certain research instruments to the researchers of the studies onwhich this article is based. Merci to Ms Magalie Bertrand for the translation of theabstract. Professor Mattes was a Fellow of the Stellenbosch Institute for AdvancedStudies (STIAS) when the article was written.
2. The surveyed countries (Afrobarometer Round 3) were: Benin, Botswana, Cape Verde,
Ghana, Kenya, Lesotho, Madagascar, Malawi, Mali, Mozambique, Namibia, Nigeria,Senegal, South Africa, Tanzania, Uganda, Zambia, and Zimbabwe (for more informa-tion about the Afrobarometer, see www.afrobarometer.org).
3. The sample for each university survey was stratified by faculty, whereby proportion-
ate to the size of third-year student enrolment in each faculty, a number of third yearcourses was randomly selected. Questionnaires were administered in class by theresearchers. In addition, a subsample of student leaders was constructed and inde-pendently surveyed. A total of 1, 411 students completed the questionnaires; statisti-cally re-weighted to 400 responses per university (Total N=1200). By the time Luescher-Mamashela et al. (2011) completed their report, a related study at the University ofBotswana was still underway (see Kgosithebe forthcoming)
4. The relevant age cohorts were for Kenyans the ages 22-25 years, South Africans 20-
23 years, and Tanzanians 22-26 years (Luescher-Mamashela et al. 2011: 19).
5. The countries included in the ALP study were Ghana (2010), Kenya (2008-2009),
Lesotho (2010), Malawi (2008), Mozambique (2008), Namibia (2009), Nigeria (2009),South Africa (2009), Tanzania (2009), Uganda (2009) and Zambia (2007). The in-tended sample size in each country was n=50. The realized sample sizes differed inKenya (n=47), Malawi (n=49), Namibia (n=37), Nigeria (n=57), and Zambia (n=51). By the time Mattes and Mozaffar completed their analysis, surveys were still ongoing
Mattes & Luescher-Mamashela: The Roles of Higher Education
in Lesotho and South Africa; their study reported the result from 27 and 32 interviewsrespectively.
6. While the survey was conducted at UCT, a new cabinet was under formation and thus
some students named the longstanding outgoing minister while others named thenew incoming minister as incumbent; moreover, many answered ‘know but can’tremember’.
7. ‘Constituency servants’ is measured on a four-point scale: They see constituency
service or representation as most important part of job; as the most rewarding part ofjob; as the most important influence on position taking; and say MPs should followwishes of constituency over those of party. ‘Partisans’, as measured on a 0 to 4 scale,see their party as most the important influence on position taking; say MPs shouldfollow wishes of party over those of their constituency; say MPs should followwishes of party over national interest; and follow wishes of party over personalconviction. ‘Institutionalists’, measured on a scale from 0 to 5, see law-making, debat-ing or oversight as the most important part of job; and as the most rewarding part ofjob; they see the national interest or own knowledge as most important influence ontheir positions; say MPs should follow the national interest rather than their party;and say MPs should follow their personal conviction rather than the party (compareMattes and Mozaffar 2011:10).
8. The ALP study did not ask MPs for their attitudes towards democracy per se, assu-
ming (perhaps without good reason) that elected representatives would uniformlyprovide pro-democratic answers. References
Acemoglu, D., Johnson, S., Robinson, J. and Yared, P., 2005, ‘From Education to Democ-
racy?’ American Economic Review 95(2), pp. 44-49.
Anderson, L., and Dodd, L., 2005, Learning Democracy: Citizen Engagement and ElectoralChoice in Nicaragua, 1990-2001. Chicago: University of Chicago Press.
Barkan, J.D., ed., 2009, Legislative Power in Emerging African Democracies. Boulder, Col:
Barro, R., 1999, ‘Determinants of Democracy’, Journal of Political Economy 107(S6),
Bellucci, P., Maraffi, M., and Segatti, P., 2006, ‘Intermediation through Secondary Associa-
tions: The Organizational Context of Electoral Behaviour’, in Gunther, R., Montero,J.R. and Puhle, H.J., eds., Democracy, Intermediation and Voting on Four Continents. Oxford: Oxford University Press.
Berelson, B., Lazafeld, P. and McPhee, W., 1954, Voting: A Study of Opinion Formation in aPresidential Campaign. Chicago: University of Chicago Press.
Bobo, L. and Licari, F., 1989, ‘Education and Tolerance: Testing the Effects of Cogni-
tive Sophistication and Target Group Affect’, Public Opinion Quarterly 53(3),pp. 285-308.
Bobba, M. and Coviello, D., 2007, ‘Weak Instruments and Weak Identification In Estimating
the Effects of Education on Democracy’, Economics Letters 96, pp.301-306.
Brady, H., Verba, S. and Schlozman, K., 1995, ‘Beyond SES: A Resource Model of Political
Participation’, American Political Science Review 89(2), pp. 271-294.
Bratton, M., Mattes, R. and Gyimah-Boadi, E., 2005, Public Opinion, Democracy, andMarket Reform in Africa. Cambridge, UK: Cambridge University Press.
Bratton, M. and Van de Walle, N., 1997, Democratic Experiments in Africa: Regime Transi-tions in Comparative Perspective. New York: Cambridge University Press.
Burns, N. K., Schlozman, L. and Verba, S., 2001, The Private Roots of Public Action Cam-
bridge, MA: Cambridge University Press.
Butler, D. and Stokes, D., 1969, Political Change in Britain: Forces Shaping ElectoralChoice. New York: St. Martins Pres.
Campbell, A., Converse, P.E., Miller, W.E. and Stokes, D.E., 1960, The American Voter. New
Castello-Climent, A., 2008, ‘On the Distribution of Education and Democracy’, Journal ofDevelopment Economics 87, pp. 179-190.
Coleman, J., ed., 1965, Education and Political Development. Princeton: Princeton Univer-
Converse, P. and Dupeux, G., 1972, ‘Politicization of the Electorate in France and the United
States’, Public Opinion Quarterly 26, pp. 1-23.
Dalton, R., 2008a, Citizen Politics: Public Opinion and Political Parties in Advanced Indus-trial Democracies, 5th Ed. Washington DC: CQ Press.
Dalton, R., 2008b, The Good Citizen: How A Younger Generation Is Reshaping AmericanPolitics. Washington DC: CQ Press.
Dee, T., 2004, ‘Are There Civic Returns to Education?’ Journal of Public Economics
Evans, G. and Rose, P., 2007a, Support for Democracy in Malawi: Does Schooling Matter?
World Development 35(5), pp. 904-919.
Evans, G. and Rose, P., 2007b, ‘Education and Support for Democracy in Sub-Saharan
Africa: Testing Mechanisms of Influence’ Afrobarometer Working Paper No. 75. EastLansing/Cape Town/Accra: Afrobarometer.
Fish, S.M., 2006, ‘Stronger Legislatures, Stronger Democracies’, Journal of Democracy.
Glaeser, E., La Porta, R., Lopez-de-Silanes, F. and Schleifer, A., 2004, ‘Do Institutions Cause
Growth?’ Journal of Economic Growth 9(3), pp. 271-303.
Glaeser, E., Ponzetto, G. and Schliefer, A., 2007, ‘Why Does Democracy Need Education?’
Journal of Economic Growth 12, pp. 77-99.
Hyman, H. and Wright, C., 1979, Education’s Lasting Influence on Values. Chicago: Univer-
Hyman, H., Wright, C., and Reed, J., 1978, The Enduring Effects of Education, Chicago:
Mattes & Luescher-Mamashela: The Roles of Higher Education
Keulder C., and Wiese, T., 2005, ‘Democracy Without Democrats? Results from the 2003
Afrobarometer Survey in Namibia’, Afrobarometer Working Paper, No. 47, CapeTown: IDASA.
Kgosithebe, L., forthcoming,‘Higher Education and Democracy in Botswana: A Study of
Students’ Attitudes towards Democracy, draft M.Ed. Dissertation, University of theWestern Cape, Cape Town.
Lam, W.M, and Kuan, H.C., 2008, ‘Democratic Transition Frustrated’, In Chu, Y.H, Dia-
mond, L., Nathan A., and Shin, D.C., eds., How East Asians View Democracy. NewYork: Columbia University Press.
Lipset, S.M., 1959, ‘Some Prerequisites of Democracy: Economic Development and Politi-
cal Development’, American Political Science Review 53: 69-105.
Lipset, S.M., 1960, Political Man, New York: Doubleday.
Luescher-Mamashela, T.M. with Kiiru, S., Mattes, R., Mwollo-ntallima, A.M., Ng’ethe, N.,
and Romo, M., 2011, The University in Africa and Democratic Citizenship: Hothouseor Training Ground? Wynberg: CHET.
Magalhaes, P., 2008, ‘Voting and Intermediation: Informational Biases and Electoral Choices
in Comparative Perspective’, In Gunther, R., Montero, J.R., and Puhle, H.J., eds.,Democracy, Intermediation and Voting On Four Continents. Oxford: Oxford Univer-sity Press.
Markowski, R., 2005, ‘Support for Democracy and Its Consolidation in Fragile Polities’, in
Van Beek, U., ed., Democracy Under Construction: Patterns From Four Continents,Bloomfield Hills & Opladen: Barbara Budrich Publishers.
Mathews, D., 1985, ‘Legislative Recruitment and Legislative Careers’, In Loewenberg, G.,
and Patterson, S.C., eds., Handbook of Legislative Research, Cambridge MA: HarvardUniversity Press.
Mattes, R., Davids, Y.D., and Africa, C., 1999, ‘Views of Democracy in South Africa and the
Region: Trends and Comparisons’, South African Democracy Barometer. Cape Town:Idasa.
Mattes, R., and Bratton, M., 2007, ‘Learning about Democracy in Africa: Awareness, Per-
formance, and Experience’, American Journal of Political Science 51(1), pp. 192–217.
Mattes, R., and Mozaffar, S., 2011, ‘Education, Legislators and Legislatures in Africa’,
Paper prepared for the Higher Education Research and Advocacy Network in Africa(HERANA), Wynberg: CHET.
Mattes, R., and Mughogho, D., 2010, The Limited Impacts of Formal Education on Demo-cratic Citizenship in Africa, Paper prepared for the Higher Education Research andAdvocacy Network in Africa (HERANA). Wynberg: CHET.
McCloskey, H., 1964, ‘Consensus and Ideology in American Politics’, American PoliticalScience Review 58(2): 361-382.
McCloskey, H., and Brill, A., 1983, Dimensions of Tolerance, New York: Russell Sage.
Milner, H., 2002, Civic Literacy: How Informed Citizens Make Democracy Work, Hanover,
N.H.: University Press of New England.
Milligan, K., Moretti, E., and Oreopoulos, P., 2004, ‘Does Education Improve Citizenship?
Evidence from the United States and the United Kingdom’, Journal of PublicEconomics 88, pp. 1667-1695.
Nie, N., Junn, J. and Stehlik-Barry, K., 1996, Education and Democratic Citizenship inAmerica, Chicago: University of Chicago Press.
Norris, P., 1997a, ‘Introduction’, in Norris, P., ed., Passages to Power: Legislative Recruit-ment in Advanced Democracies, Cambridge: Cambridge University Press.
Norris, P., 1997b, ‘Conclusions: Comparing Passages to Power’, in Norris, P., ed., Passagesto Power: Legislative Recruitment in Advanced Democracies. Cambridge: CambridgeUniversity Press.
Norris, P., 2002, Democratic Phoenix: Reinventing Political Activism. Cambridge: Cambridge
Papaioannou, E. and Siourounis, G., 2008, ‘Economic and Social Factors Driving the Third
Wave of Democratization’, Journal of Comparative Economics 36, pp. 365-387.
Pascarella, E. and Terenzini, P., 1995, ‘The Impact of College on Students: Myths, Rational
Myths, and Some Other Things That May Not Be True,’ NACADA Journal 15(2), pp. 26-33.
Powell, G.B., 11986, ‘American Voter Turnout in Comparative Perspective’, American Po-litical Science Review 80(1): 17-43.
Prothro, J. and Grigg, C., 1960, ‘Fundamental Principles of Democracy: bases of Agreement
and Disagreement’, Journal of Politics 22, pp. 276-294.
Putnam, R., 1995, ‘Tuning In, Tuning Out: The Strange Disappearance of Social Capital in
America’, PS: Political Science and Politics 28(4), pp. 664-683.
Rose, R., Mishler, W. and Haerpfer, C., 1998, Democracy and Its Alternatives: Understand-ing Post Communist Societies. Baltimore: Johns Hopkins University Press.
Rose, R., Mishler, W. and Munro, W., 2006, Russia Transformed: Developing PopularSupport for A New Regime. Cambridge: Cambridge University Press.
Shin, D.C., 1999, Mass Politics and Culture in Democratizing Korea. Cambridge: Cambridge
Sullivan, J. and Transue, J., 1999, ‘The Psychological Underpinnings of Democracy: A
Selective Review of Research on Political Tolerance, Interpersonal Trust and SocialCapital’, American Review of Psychology 50(1), pp. 625-50.
Texeira, R., 1992, The Disappearing American Voter. Washington: Brookings Institution.
Verba, S. and Nie, N., 1972, Participation in America: Political Democracy and Social Equal-ity. New York: Harper and Row.
Verba, S., Nie, N. and Kim, J., 1978, Participation and Political Equality. Chicago: University
Verba, S., Schlozman, K. and Brady, H., 1995, Voice and Equality: Civic Voluntarism inAmerican Politics. Cambridge, MA: Harvard University Press.
Wolfinger, R. and Rosenstone, S., 1980, Who Votes? New Haven: Yale University Press.
Media Release - 11 October 1999 MALARIA AND THE KRUGER NATIONAL PARK A combination of factors has resulted that the malaria areas of South Africa are experiencing unusually high numbers of malaria cases, and expectations are that the situation will worsen in the coming months. Visitors are advised that the Kruger National Park and adjoining areas fall within the malaria area, and prec
JHEA/RESA Vol. 10, No.1, 2012, pp.139–170
Council for the Development of Social Science Research in Africa 2013
The Roles of Higher Education in theDemocratization of Politics in Africa:Survey Reports from HERANA1
Robert Mattes* & Thierry M. Luescher-Mamashela**
Abstract
and Benin (4% each), Mali and Mozambique (3%) and Malawi and Tanza-nia (1%). Possession of actual university degrees ranged from 4 per cent inNigeria to less than 1 per cent in many countries according to the surveydata. Given the rapid expansion of education systems across the continentin less than fifty years, the possession of a formal education – as of 2005-2006 – decreases sharply with age, as well as amongst rural people andamongst women. (Mattes and Mughogho 2010)
The ALP surveys also looked at highest education attainment but regis-
tered extremely different results. Across 11 of the same countries includedin the Afrobarometer study, voters elected substantially larger proportionsof educated and highly educated legislators: the majority of the randomlyselected samples of MPs (59%) had at least completed an undergraduateuniversity degree; 19 per cent had completed an undergraduate degree, and15 per cent had either a post-graduate diploma or honours degree. One-in-five had a Masters degree (20%) and 3 per cent had doctorates. As illus-trated in Figure 1, these levels of higher education attainment amongst MPsare anywhere from 15 to 80 times higher than amongst ordinary citizens intheir respective countries (for details see Mattes and Mozaffar 2011). Thisfinding in itself tells us something about the ability of political parties to re-
Figure 1: Educational Attainment amongst African MPs and Citizens
Mattes & Luescher-Mamashela: The Roles of Higher Education
counsel, and would more likely turn to information resources external to thelegislature, such as civil society organizations or universities, than their lesseducated peers. Highly educated MPs were also more likely to have trav-elled outside of Africa on fact-finding trips. With one exception (using inter-nal resources), these correlations remained even after the application of therelevant statistical controls for demographic and experiential factors androle orientations. The authors also noted that education had an additionalindirect effect via role orientations: that is, education promoted an institu-tionalist role definition, which in turn also separately and independently pro-moted the use of the internet and other internal parliamentary resources;and education reduced partisan role orientations, which in turn promote internetuse but substantially detract from the use of internal resources. Therefore,
Figure 2: Committed Democrats
Mattes & Luescher-Mamashela: The Roles of Higher Education
Figure 3: Tranformative Democrats