A number of institutions conduct surveys on voter patterns throughout the world to help manage the effectiveness of the electoral process. It is therefore, against this background that the IEC commissioned the Human Sciences Research Council (HSRC) to conduct this survey, the Chairperson of the IEC, Dr Brigalia Bam, said.

Dr Bam said it is important to get objective information about voter participation in South Africa, specific to municipal elections. The survey sought to determine the culture of voting behaviour throughout the country. Other subsidiary objectives included:

 Determining the extent and location of unregistered potential voters in South Africa;
 Assessing voting behaviours and patterns;
 Obtaining information regarding people’s perception of general governance issues that might impact on voting behaviour;
 Determining awareness and knowledge of the forthcoming municipal elections; and
 Assessing perceptions of the IEC.

“It is encouraging to note from the survey report the overwhelming interest by South Africans in voting, after six successive elections. According to the survey report, all registered respondents in the nine provinces indicated that they are registered in the district in which they were currently residing (91%),” said Dr Bam.

Voter registration for the upcoming municipal elections pointed to the fact that South African democracy is maturing. Over 21 million of a possible 27 million eligible voters have now registered to vote in the coming elections. This is indeed encouraging, taking into account that democracy is a new concept in South Africa.

Democracy is an ongoing process and people are starting to internalise multi-partyism. In these municipal elections alone, 97 political parties are participating and 45 189 candidates countrywide will contest the elections. This represents 19% growth in contesting political parties and 50% increase in candidates participating compared to the 2000 municipal elections.

Almost 96, 3% of people polled in the survey, said they had one identity document, while only 3, 7% said they did not have one. These results are consistent for all the provinces with not one scoring lower than 90%.

To check your registration details or confirm where you should vote contact the Call Centre toll free on 0800 11 8000 or visit the IEC website at www.elections.org.za or SMS your ID number to 32810.

ISSUED BY THE ELECTORAL COMMISSION

For media enquiries please contact:

Lydia Young
Media Liaison Officer
082 650 8652

Survey on South African voter participation in elections

Prepared for

The Independent Electoral Commission (IEC)

Prepared by

The Socio-Economic Surveys Unit of the Knowledge System Group

Human Sciences Research Council

The report was compiled by Mbithi wa Kivilu, Yule Davids, Zakes Langa, Kealeboga Maphunye, Bongiwe Mncwango, Susan Sedumedi, & Jare Struwig

Acknowledgement

The survey on “South African voter participation in elections” was conducted by the Socio-Economic Surveys division of the Knowledge Systems group, Human Sciences Research Council, in October and November 2005. The study was financed by the Independent Electoral Commission. The final report is a product of a collective effort involving a number of individuals who played various roles in the whole implementation process. We wish to express our heartfelt gratitude to the following people who in various ways participated and helped to shape the study.

• Michael Kahn and Udesh Pillay for their leadership and support through out the duration of the project.
• The anonymous 4930 South Africans who so generously gave their time to participate in the survey.
• A. Booyens, K. Govender and H. Bolton for project administration and financial management.
• Lavern, Gift, and Dorothy for providing support as project assistants.
• Barbara, Enency, Sara, Peter, Solly, Sydney, Rosina, Joy and their teams of fieldworkers for supervising the collection of the data.
• Jare Struwig, Bongiwe Mncwango and Mbithi wa Kivilu for the development and design of the questionnaire.
• Susan Sedumedi, Yule Davids, Kealeboga Maphunye, Bongiwe Mncwango, Zakes Langa, Jare Struwig, Monica Peret, Marlene Roefs, Annemarie Booyens and Mbithi wa Kivilu for editing the questionnaire.
• Monica Peret, Anneke Joordan, Ann Coetzer, Lynfer consultancy for data management and capturing.
• Prof. Stoker for weighting the data.
• Susan Sedumedi, Yule Davids, Kealeboga Maphunye, Bongiwe Mncwango, Zakes Langa, Jare Struwig, Stephen Rule and Mbithi wa Kivilu for data analysis and report writing.
• Stephen Rule and Annemarie for editing of the report.
• Jare Struwig and Mbithi wa Kivilu for the overall management and coordination of the project activities.
• Stephen Langtry, the IEC project coordinator, for his exemplary work.

Table of Contents

Executive Summary 2
Purpose of the Study 2
Research context 2
Objectives 2
Overall assessments of the last five democratic elections 2
Citizens’ views of Municipalities 2
Methodology 2
Population and Sampling 2
Democracy and Governance Issues 2
What do citizens’ think about our democracy? 2
Democracy and Elections 2
Economic Trends 2
Personal Economic Situation 2
Identity Documents 2
Possession of green-bar coded ID 2
Possession of other old South African ID 2
Intend to apply for a green bar-coded ID 2
Registration as a voter 2
Participation of South Africans in both national and municipal elections 2
Participation in the municipal elections 2
Interest in national and municipal elections 2
General perceptions regarding voting 2
Floor Crossing 2
Voting Behaviour and Voting History 2
Persuasive Issues to Encourage Voting 2
Voting Irregularities 2
Satisfaction with absence of irregularities during the elections 2
Voting Logistics 2
Access of voting stations 2
Satisfaction with time to get to the voting stations 2
Satisfaction with cost to get to the voting stations 2
Satisfaction with time it took to vote 2
Satisfaction with signage and instructions 2
Satisfaction with time spent standing on the queues 2
Conduct of political parties in the voting stations 2
Satisfaction with secrecy of the vote 2
Satisfaction with representatives of the IEC 2
Access to voting stations 2
Satisfaction with length of queues 2
Satisfaction with other services of the IEC in the voting stations 2
Providing for the disabled 2
Perceptions on whether elections were free and fair 2
Perceptions on whether elections get better every time 2
Preference for whether voting should be on a normal working day, public holiday or weekend 2
Media and Information 2
Satisfaction with institutions 2
The Impact of the Social Capital on the Democratic Process 2
General Conclusion 2
Recommendations 2
Identity Documents: 2
Information campaigns of the IEC: 2
Media Usage: 2
Floor Crossing: 2
Impact of lack of service delivery on voting: 2
The role of women in elections: 2
Non-participation and voter turnout in the upcoming elections: 2
Voting irregularities: 2
Race: 2
Youth Issues: 2

Executive Summary
South Africa is now more than ten years into its inclusive democracy. The overall direction and success of a democratic government can be judged with various types of evidence. This executive report focuses on one type, that is, the opinions of South African citizens about the overall direction of their new democracy and in particular the role elections play in consolidating our democracy. We therefore opt not to look at expert judgments or to measures of formal rights, since we believe that the views of ordinary citizens, as the end users of what democratic governments supply, can offer perhaps the most conclusive assessment of the quality of democratic governance including elections.

We begin by outlining the purpose of the study and an overview of the last five democratic elections. This is followed by a summary of the research methodology and sample characteristics of the survey. This brief introduction is provided to give the reader the required background information to the current survey results. Then we turn to look at democracy and governance; an evaluation of personal economic situation of South Africans as well as the overall economic situation of South Africa. We also use public opinion to get a better understanding of voter registration; voter participation; general perceptions on voting; voting behaviour during both the national and municipal elections and voting logistics. We then move to a discussion of the role of the media and information with regards to the IEC. Another section of the executive report highlights satisfaction levels with public institutions such as the IEC. Next we talk about access to basic services and how this relates to voting. The impact of social capital on the democratic process is one of the last sections that we have included in the report. We look in particular at membership affiliation and participation in political events. Finally, we conclude by providing recommendations with regards to elections with the hope that this will further enhance the credibility and legitimacy of elections in South Africa.

This evidence is supplied by the IEC and the Human Sciences Research Council (HSRC) 2005 survey. This survey was conducted in October 2005 and involved a sample of 5000 individuals of ages 18 and above. These individuals were selected from 5000 households across South Africa (see Methodology Section for details).

Ten years into democratic government, South Africa has laid the foundations for the design and implementation of policies to ensure democratic consolidation, competitive multi-party engagement, and citizen participation. The framework created for political representation is laid out in the founding provisions of the Constitution (Chapter 1) which sates that South Africa is one sovereign democratic state founded on:
o Human dignity, the achievement of equality and the advancement of human rights and freedoms,
o Non-racialism and non-sexism
o Supremacy of the Constitution and the rule of law
o Universal adult suffrage, a national voters' roll, regular elections and a multiparty system of democratic government, to ensure accountability, responsiveness and openness.

On the basis of these founding provisions, South Africa has run three general national and provincial elections, and two nation-wide local government elections, all of which were declared to be free and fair. The Independent Electoral Commission (IEC) is responsible for organising and managing elections. In order to give legitimacy to elections the Constitution guarantees the institutional independence of the IEC, and restrains other government bodies from interfering with its functions. In addition, no person or organ of state may interfere with the functioning of the IEC. The IEC is also constitutionally accountable to the National Assembly, and must report on its activities and the performance of its functions to the Assembly at least once a year. The IEC is one of six institutions established in terms of Chapter 9 of the Constitution to strengthen our constitutional democracy. These institutions are: the Public Protector; the Human Rights Commission; the Commission for the Promotion and Protection of the Rights of Cultural, Religious and Linguistic Communities; the Auditor-General; and the Commission for Gender Equality.

Purpose of the Study
Research context
In response to the request by the Independent Electoral Commission (IEC ) to conduct a survey prior to the final registration date for the 2006 local government elections, the Socio-Economic Surveys’ division of the Human Sciences Research Council (HSRC) conducted a nationally representative survey in October, 2005. The intention was to determine the state of political culture with respect to voting behaviour in South Africa. Such an assessment would, inter alia, allow for the identification of interventions that would foster an ethos of political participation. The survey would also determine the extent to which the IEC’s primary objective is being achieved, namely to strengthen constitutional democracy through the delivery of free and fair elections in which every voter is able to record his or her informed choice.

Objectives
The main objectives of the survey were thus to evaluate voting behaviour in South Africa and to determine public perceptions of the IEC as the institution responsible for elections in South Africa.

Subsidiary aims were:
o To determine the extent and location of people without identity (ID) documents in South Africa;
o To determine the extent and location of unregistered voters in South Africa;
o To assess reasons for the registration gap;
o To better understand the reasons for voting apathy;
o To assess voting apathy;
o To obtain information regarding people’s perception of general governance issues that might impact on voting behaviour;
o To determine awareness and knowledge of forthcoming elections and subsequent procedures, and
o To determine knowledge and perceptions of, and trust in, the IEC.

Overall assessments of the last five democratic elections
A comparison of overall results of the last five democratic elections clearly reveals that the ANC has enjoyed overwhelming support from the electorate and that if this trend continues the party will dominate other political parties for years to come. This does not pose a major problem for democracy since many established democracies such as the United States, United Kingdom, Germany and India have all endured extended periods of one-party dominance in executive and/or legislative elections . It is however of concern when dominant governing parties see less and less need to respond to public opinion because they are assured of re-election in the face of weak opposition. Moreover, there is always the danger that governing parties with huge majorities in legislatures might become arrogant and less accountable to the electorate, especially in countries with a proportional representation system such as South Africa.

Voter turnout is extremely high for the national and provincial elections but much lower for the municipal elections. Voter turnout in the last two municipal elections was 48% and 49% for 2000 and 1995/6 respectively. The voter turnout for the national and provincial elections was 76% (2004), 88% (1999) and 86% (1994). These results are consistent with many other democratic countries where the electorate seems to favour national and provincial elections over municipal or local elections.

Citizens’ views of Municipalities
Previous studies indicated that even though municipalities (local government) are closest to the citizens and at the heart of service delivery, many citizens feel that they are performing worse than national and provincial government.

To repeat, it will be interesting to see whether the tide will change in favour of the municipalities when people go to vote in the upcoming municipal elections. The results from the 2005 HSRC / IEC survey also indicated that the local sphere of government was viewed less favourably than were the provincial and national spheres of government. This survey asked respondents to “indicate the extent to which you are satisfied or dissatisfied with the following institutions (national, provincial and local government) in South Africa at present”. National government is rated most favourably with 64.0% saying they are "satisfied" or "strongly satisfied", whereas the comparative proportions for provincial and local government were 53.0% and 43.0% respectively.

Methodology
Population and Sampling
The target population of the study was all South African citizens ages 18 and above. This includes people living in households/ structures/ hostels/ but excludes those living in special institutions, hospitals, and prisons.

The HSRC Master Sample, which was developed using the Census 2001 and with the Enumerator Area (EA) as the primary sampling unit, was used as the sampling frame. Explicit and implicit stratification was applied to ensure that the geographic profiles of the targeted population such as province, environment milieu, age category, sex, race education level, Living Standard Measurement (LSM), and current employment status are represented in the sample. The 2001 census database contains descriptive statistics (e.g. total number of people and total number of households) for all EAs in South Africa. The value of using the HSRC Master Sample was that a national representative sample can be drawn and the results of the survey can be properly weighted to the 2001 census population figures.

To ensure that the nine provinces, urban/ rural population and people living in different types of areas (e.g. informal settlements, traditional areas, formal urban, farmlands) were well-represented these variables were used as explicit stratification variables. To ensure that the sample was also representative in terms of the ethnic and cultural diversity of South Africa, the HSRC’s geo-demographic categories were used as the implicit stratification variables. These geo-demographic categories reflect the diversity of the South African population based on rural/urban, income, education, ethnicity and geographic characteristics. Such stratification was also used to ensure that the metropolitan, semi-urban and rural populations of South Africa were covered well in the sample.

A sample of 5000 individuals of ages 18 and above was selected from households in 500 of the Master Sample EAs. Maps for the 500 EAs were developed by the HSRC’s GIS Centre and were used by the team to navigate into the correct areas. Descriptive statistics of the of the sample disaggregated by key demographic variables are provided in Appendix C.

Democracy and Governance Issues
What do citizens’ think about our democracy?

Another key question concerned how satisfied South Africans are with their democracy. The results of the current HSRC survey revealed that 16% of South Africans are “very satisfied with the way democracy works in South Africa”; 40% are “satisfied”; 20% are “dissatisfied”; and 8% are “very dissatisfied”. Analysis of the data shows that provincially, the proportions of people who were very satisfied were highest in North West (25%), Gauteng (22%) and Mpumalanga (20%). At the opposite extreme, less than one in eleven in the Western Cape (10%) and Eastern Cape (10%) were very satisfied with the way democracy is working in South Africa. What we discovered was that environmental milieu plays a very important role in how people assess their satisfaction with democracy. A greater proportion (23%) of the people living in urban informal areas indicated that they were “very satisfied” with the way democracy is working than was the case in urban formal (17%), rural (16%) or tribal (14%) tribal areas. The results indicated that young people are more positive about our democracy than older people. Compare for example the 44% in the age category 18- 24 years that were “satisfied” with the way democracy works in South Africa with the 34% of people over 65 years holding the same view. In racial terms, significantly more Africans (20%) were “very satisfied” with the way democracy is working than was the case amongst Coloureds (6%), Whites (4%) and Indians/Asians (2%). Indians (31%) and Whites (23.4%) also recorded high levels of dissatisfaction with the way democracy is working.

Democracy and Elections
The integrity of the election process and legitimacy of the outcome are important requirements for a stable democracy. The fact that South Africa has conducted five successful and legitimate elections at all levels of government is an indication of well-functioning independent electoral and government institutions. It further attests to the cooperative strength of government institutions and the electoral commission.

Economic Trends
We now turn to examine South Africans evaluations of whether their personal life as well as that of fellow South Africans has improved over the last five years. We first asked respondents to assess the economic situation of most South Africans. The results showed that 56% felt that the lives of most South Africans has improved over the last 5 years, 25% indicated that it stayed about the same while 17% said it got worse. In this survey, in provinces that used to make up the old Cape Province (Western Cape, 40%; Northern Cape, 50%; the Eastern Cape, 49%), sentiment was least favourable in comparison with that expressed about the lives of people in Limpopo (70%); Mpumalanga (65%) and North West (64%) over the last five years.

Whereas more than seven out of ten (72%) people living in urban informal settlements thought that the economic situation of most South Africans had improved over the last five years, this view was less prevalent in tribal (65%), rural (56%) and urban formal (49%) contexts, where the opposite perception was proportionately higher. Younger people were more likely than their older counterparts to hold the view that the lives of South Africans have improved over the last five years. Compare for example the 63% in the age category 18–24 years that were positive in this respect, with the 47% of the age category 55–64 years. Race still appears to play a very important part in the lives of people of South Africa. Whereas almost two-thirds (65%) of Africans felt that the life of South Africans over the last five years has improved, more than half of Whites (50%) and Indians/Asians (59%) indicated that South Africans’ lives over the last five years have gotten worse. About two-thirds of South Africans with low LMS levels (64%) or medium LSM levels (62%) were of the view that the lives of South Africans hade improved over the last five years. At 40%, however, this positive sentiment was far less prevalent amongst high LSM people.

Personal Economic Situation
More than two in five of the people (44%) felt that their personal economic situations have improved over the last five years. About 37% said that these have stayed the same while 18% indicated that they had got worse. Provincially, the extremes were in Limpopo, where almost two-thirds (64%) expressed the view that their personal economic situations have improved and in the Northern and Western Cape, where the equivalent proportions were only 39% and 33% respectively. More than half (52%) of the people living in tribal areas indicated that their lives have improved, as opposed to smaller proportions elsewhere. Conversely, almost a quarter (23%) of those living in urban formal settlements were of the view that their personal economic situations have worsened, more than was the case amongst residents of other environmental milieu. Greater proportions of young people than older people felt that their personal economic situations had improved over the last five years. Gender differences on this issue were insignificant. When we asked respondents to indicate whether their lives have “improved, stayed the same or gotten worse over the last five years”, Africans again demonstrated a much more positive evaluation than did the other race groups. Almost half (50%) of Africans indicated that their lives have improved over the last five years, as opposed to the only 19% of Whites who expressed this view.

Identity Documents
Possession of green-bar coded ID
We asked people to indicate whether they have a green-bar coded identity document. Almost all (96%) of the people said they have one, while only four percent said they do not. These results are consistent for all the provinces with not one scoring lower than 90%. It is interesting to note that in the least wealthy provinces, Limpopo and Eastern Cape, more than five percent indicated that they do not posses a green-bar coded identity document. In relation to age, the results clearly showed that a large proportion (11%) of the youth is not in possession of a green-bar coded identity document. Although differences were very small, Africans (4%), males (4%) and people with low LSMs (4%) were slightly more likely not to be in possession of a green-bar coded ID book than were people of other races, females and high LSM people.

Possession of other old South African ID
Relatively few people possess an identity document other than the green-bar coded variety. Peak levels occur in the Northern Cape (16%), Western Cape (12%) and Gauteng (11%), where the proportions are one in six or fewer. In urban formal areas more than 10% indicated that they have an identity document other than the green-bar coded, far more than in other environmental milieu. The possession of identity document other than the green-bar coded is more common among the older population than in the other age categories with 50% in the 45 – 54 age category indicated that they have such a document. Males emerged as more likely to have older forms of ID than did females. In terms of race, greater proportions of Whites (33%) and Indians/Asians (24%) said that they are in possession of an identity document other than the green-bar coded document than was the case amongst the other races. Similarly, people with a high LSM (12%) are more likely to be in possession of an identity document other than the green-bar coded document than are those with a low LSM.

Intend to apply for a green bar-coded ID
We asked those people not in possession of a green-bar coded identity document to indicate whether they would apply for one. Most (92%) said that they would and seven percent said “no”. The proportions of people who said “no” were highest in North West and Limpopo and in tribal areas (12%) and urban informal areas (7%). Males (11%) appeared more reluctant than females to apply for the green-bar coded identity document. Greater proportions of Indians/Asians (19%) and Whites (8%) than other races indicated that they would not apply for the green-bar coded identity document. It was also found that the older generation was less likely to see the need to apply for the green-bar coded identity document; with 27% of those aged 65 years or older saying they do not intend to apply.

The most frequent reason provided for not intending to apply for a green-bar coded identity was affordability (27%). Others said that the Home Affairs Office is too inaccessible (11%) and they do not know where or how to apply (3%).

Registration as a voter
With respect to voter registration, there is a general positive trend in people’s perceptions of registering as a voter; the convenience of registering and the technical aspects involved including time taken to register, obtaining proof of registration and knowledge about registering in own districts; as well as intention to vote after registering. There was a general increase in the percentage of respondents who were registered as South African voters irrespective of demographics characteristics such as province, environment milieu, age category, sex, race, education level, and Living Standard Measure (LSM). This was confirmed by results of statistics from various surveys across the years on the question of whether people are registered as voters.

Most of the registration happened before 1998 or in the inter-election period of 2002 – 2003 with a high percentage of recent registration in Eastern Cape (48%) and Northern Cape (40%). Rural informal (21%) and tribal areas (25%) accounted for most of the registration in 2004 – 2005. Factors that seemed to have played a pivotal role in registration include residing in the Northern Cape (88%); living in tribal environmental milieu (83%); being in age category 65 years and above (97%); being a female (84%); being African or White (82% respectively); having no schooling or Grade 0 (94%); and having a medium LSM (85%). However, differences in year of registration and sex of the respondents as well as level of education did not yield any significant pattern over these years in term of registration as a voter.

Almost all the registered respondents in the nine provinces indicated that they have registered in district in which they were currently residing (91%). The immobility of respondents from one district to another seemed to have played a role in most of them having registered in their respective districts. About 89% o f the respondents indicated that they have not moved home in the last 18 months.

Throughout the nine provinces, four environmental milieus, age and sex categories, race, educational level and LSM, a significant percentage of respondents seemed to be fairly satisfied with the ease/ convenient of the registration process with 97% indicating that they found it easy/ convenient to register based in all the provinces. Differences among the various demographic variables were minor. The time taken to register was not a problem either as the majority of the respondents indicated that it took them approximately 5 – 10 minutes to register as voters 32% of which were distributed throughout the provinces; various environmental milieus, with different ages, sex, race, educational level and LSM respectively. Another piece of evidence showing the ease of going through the registration process is the proof of registration which the almost all respondents (ranging from 99% in North West to 94% in Free State) who received the proof of registration. Among those who stated that they did not receive proof of registration in the nine provinces 68% said that they were not given any reason for not receiving proof.

It should be emphasized that people’s perceptions and attitudes differ in the reporting of their actual behaviour and the reporting of intent to doing something. As a way of establishing the overall percentage of those who are likely to register to vote and to participate in elections, it was important to also consider those who intended to register. A very high proportion of residents in Limpopo (92%) indicated that they intend registering to vote soon. The equivalent proportion for Western Cape and Free Sate, at the other extreme was 51%. Overall, 70% intended to register soon. Young people aged 18-24 were most likely (75%) to be of the intention to register soon, in contrast to their older counterparts. The question here is whether the reported intention of the young people to vote will turn into actual registration to vote. The main reason for being uncertain or not intending to register is that they were not interested in voting (64%). This ranged from as high as 80% in the Western Cape to a low of 27% in the Free State. In the latter province the main reason provided was that facilities to register were not available (28%). The lack of interest in voting emerged as the main reason for non-registration across all age groups. Only amongst 55-64 year olds was “have not yet got around to it” a more frequently stated reason.

From the 83% of the respondents who have re-registered to vote in all the provinces, 92% indicated that they plan to vote. The likelihood of having a high volume of voter turnout in the 2005 local elections might be 60% which is higher than the 49% turnout in the 1995/96 local elections.

As a disincentive to voting, lack of interest in politics emerged as most important in urban formal (31%) and rural informal areas (39%). Dissatisfaction with municipal service delivery was most frequently mentioned in urban informal areas (46%) and tribal areas (38%).

Based on the positive results from the section on registration as a voter, we can state with optimism that the percentage of people who will register to vote for the upcoming municipality election would be higher than previous years.

Participation of South Africans in both national and municipal elections
Information on participation in the previous national and municipal elections was gathered by asking the respondents to indicate whether they voted in the last elections. Generally there was a decline in the proportions of people participating in the national elections irrespective of how they were grouped. Three provinces reported a decrease in voter participation in the three previous national elections with Limpopo reporting a decline of 15% and KwaZulu-Natal and the Western a decline of 13% each from 1994 to 2004. There was a slight increase from 1999 to 2004 of three percent in Mpumalanga and one percent in the Eastern Cape. Males showed a gradual decrease over the years but female participation remained constant at the 75% reached in 1999. When compared by racial groups, Indians/Asians had the largest decrease of six percent from 1999 to 2004 elections, followed by Africans with a three percent decline in the same period. Coloured respondents remained steady at the 62% reached in 1999. There was an increase in the participation levels among the 25-34 year olds of 11% from 1999 to 2004, a slight decrease of one percent among the 55-64 years olds and no change among the 65+ category. Participation increased by four percent among those with no schooling and stayed the same for those with education qualification levels of Grades 1 to 6. A decrease was evident among the rest, notably amongst the post-matric respondents.

Participation in the municipal elections
Participation levels in the municipal elections were varied with four of the nine provinces, namely, the Eastern Cape, Northern Cape, KwaZulu-Natal and Mpumalanga showing an increase in participation, with Eastern Cape indicating the highest increase of about five percent. The rest of the provinces showed a general decrease, the largest of about six percent occurring in Limpopo. Although a slight increase in participation was identified amongst the low LSM group, there was a decline in municipal voting amongst medium and high LSM people. The participation of the three smaller race groups increased, whereas that of African participation decreased slightly by about two percent. While those with no schooling did not show much change in participation, there was an increase among those with Grade 1-9 levels of education. There was a gradual decrease among those with at Grade 10 level of education.

Interest in national and municipal elections
Provinces with the largest proportions of eligible voters who indicated no interested in the national elections were the Northern Cape (19%), Western Cape (18%) and Gauteng (18%). The pattern was slightly different with the municipal elections where Mpumalanga (29%) and Northern Cape (29%) had the highest proportions while KwaZulu-Natal (18%) had the lowest proportion of people who indicated no interest in the municipal elections.

It makes a big difference which political party wins the national elections to a large proportion of eligible voters in the North West (83%) and in the Free State (82%) compared to 66% in the Eastern Cape. With regard to the municipal elections, it makes a big difference which political party wins to 69% of eligible voters in Gauteng and to 67% in Mpumalanga, but only to 55% in the Free State.

A higher proportion of males than females were interested in both national and municipal elections. These results are corroborated by the higher proportion of males than females who participated in both the national and municipal elections.

It makes a big difference to slightly more male than female voters which party wins either the national or the municipal elections. Racial group differences were evident in the extent of interest South Africans shown in both national and municipal elections. Respectively, 57% and 41% of Africans had interest in national and municipal elections while this was the case with only 22% and 16% of Indians/Asians, respectively in national and municipal elections. Of interest to the IEC would be the groups that showed no interest in the elections. Leading among those who were not interested in the national elections were the Indians/Asians followed by Coloureds while in the municipal elections were the Indians/Asians followed by the Whites.

While it makes a difference which political party wins the national elections to 78% of Africans and 65% of Whites, this is the case with only 57% of Coloureds and 55% of Indians/Asians. These proportions emerged as even lower for municipal elections: 64% for Africans, 61% for Whites, 49% for Coloureds and 46% for Indians/Asians.

When compared by environmental milieu and LSM, the urban formal group (18%) and the high LSM group (25%) had the highest proportion that were not interested in the national elections while the urban informal (8%), tribal area (8%) and low LSM (8%) groups had the lowest proportions of those not interested. With regard to the municipal elections urban (26%) and rural (24%) formal and the high LSM group had the highest proportions that were not interested while tribal area residents and the low LSM group had the lowest. Interest in national elections emerged as highest amongst residents of urban informal areas and people in the lowest LSM category.

Age of the participant was important in identifying those who were interested or not interested in both the national and the municipal elections. Lack of interest in national elections was highest among the 18-24 years and 55-64 years while 35-44 years had the least proportion of those not interested. The young, 18-44 years had the highest proportion not interested in municipal elections while the old 65+ years old had the least.

It makes a difference most to those aged 35-44 years which political party wins the national elections (77%) and municipal elections (67%) while to those aged 65+ years it makes no difference to 68% and 53% respectively.

Education level had an influence on interest people had in elections. In both national and the municipal elections, the largest proportion that was not interested was among the post-matric with 17% and 31% respectively and among those with no schooling with 7% and 13% respectively.
It makes a big difference which political party won the national elections among those with no schooling (78.0%) and least among Grades4-9 and post-matric. The municipal elections were slightly different with those with low education showing the least proportion agreeing while those with at least Grade 10 indicating the largest proportion.

General perceptions regarding voting
Analysis of questions pertaining to general perceptions around voting issues indicated that South Africans embrace the principles of a true democracy and portray a high level of civic conscience, in that 92% of respondents indicated that they agree that each person can freely choose who to vote for without feeling forced by others. About 80% of the respondents stated that it was the duty of all citizens to vote. Just more than a quarter (27%) felt that voting was pointless since all parties are the same after being elected. Interestingly half of the respondents indicated that voting should be made compulsory.

Some discontent with the electoral system was noted with 37% of the respondents stating that the party that they voted for did not protect their interest. Almost half (49%) of the respondents indicated that they felt that politics were too complicated these days for them to understand. Discontent in terms of floor crossing was also recorded in that 58% stated that elected politicians should resign from offices when they cross the floor. Significant differences amongst LSM groups were also found for the various statements. Low and medium LSM groups tended to agree more than the high LSM group that voting makes a difference, that it is the duty of all citizens to vote, that politics are rather complicated, that voting should be made compulsory and that quotas for women should be increased. The high LSM group tended to agree more than the other groups that after being elected all parties are the same so voting is pointless, that elected politicians who cross the floor should resign and that young people are losing interest in politics.

Just more than a quarter (27%) agreed to the statement: “I will not vote because of a lack of delivery of services”. When this statement was analysed by age it was found that more people in the youngest age group (18-24 years) agreed most with the statement.

When the age groups were further compared for responses to various variables it was found that the younger age group was less likely than the other groups to agree that politics was too difficult to understand and less likely to think that the party they voted for did not protect their interests. This age group was also more likely than the other age groups to agree that young people are losing interest in elections.

People in urban formal areas, Asian and African as well as high LSM respondents were more likely not to vote because of the lack of delivery of services.

When questions relating to women in politics were posed to respondents a clear gender bias was found. Female respondents campaigned much more for women than did their male counterparts for the reason that most (63.0%) of women felt that their needs would be better addressed if there were more women in politics.

Floor Crossing
Half of the respondents stated that floor crossing discourages people from voting and about 46% stated that it means that they have wasted their vote. Less than a third (31%) indicated that it is a true reflection of a working democracy. Floor crossing was clearly a bigger concern in the Western Cape where 72% of respondents stated that floor crossing discouraged people to vote, followed by the Free State and Northern Cape provinces. A higher proportion of respondents in urban formal areas were of the opinion that floor crossing discourages voting. This was also true for Coloured, Indian/Asian and White respondents. High LSM respondents were more inclined than the lower LSM respondents to agree that floor crossing discourages people from voting.

Exactly three-fifths (60%) of respondents know what floor crossing is and a third (33%) do not know what floor crossing is. It was in the Eastern Cape and the Limpopo that such knowledge appeared to be glaringly lacking. Though all races seem to know what floor crossing is, only just over half (53%) of Africans seem to know what floor crossing is. This lack of knowledge is prevalent in the lower LSM groups, where it is generally known that most Africans are classified.

Asked whether floor crossing should be continued or not, nearly six out of ten (57%) of respondents did not think that floor crossing should be continued. This was particularly pronounced in the formal urban areas and much less evident in the tribal areas. More males than females seem to have a particular aversion to floor crossing and think it should not be continued. It is also respondents with higher levels of education who think that floor crossing should not be continued.

Hence, most respondents tended to disagree or strongly disagree with the statement that floor crossing is a true reflection of a working democracy (question 38 of the IEC questionnaire). However, for rural or tribal communities this statement seemed confusing and they preferred to choose a ‘don’t know’ response rather than commit to it. Also most female respondents tended to choose this option as well whilst the formal urban and male respondents disagreed with the statement that floor crossing is a true reflection of a working democracy.

In the same vein, respondents from lower grades did not commit themselves either way whilst those with higher qualifications tended to disagree with the statement that floor crossing is a true reflection of a working democracy.

In terms of the statement that floor crossing discourages people from voting about half of the respondents agreed. However, it was more in provinces like the Western Cape, the Northern Cape and the Free State that respondents agreed whilst those from Limpopo, the North West and the Free State tended to disagree with that statement. In terms of race, Indians/Asians and Coloureds tended to strongly agree with the statement that floor crossing discourages people from voting.

In terms of the statement that says ‘floor crossing means I wasted my vote’, nearly half (46%) agreed. Just over a fifth (22%) disagreed or disagreed strongly with the statement. However, an equal proportion did not know either way. People who had an opinion about this were from the formal or informal urban areas and they tended to agree with this statement. These were also largely male respondents and many of them have higher levels of education.

Voting Behaviour and Voting History
Almost three quarters (72%) of the respondents indicated that they had voted in the 2004 elections and 60.0% indicated that they had voted in the 2000 municipal elections. When compared with actual voter turnout an undercount was found for national elections and an over-count for local elections. With regards to voting intention, tested by the question: “If there were an election tomorrow, would you vote?” a definite indication of social desirable responses was found in that 86% of respondents indicated that they would vote in forthcoming national elections and 81% in forthcoming municipal elections.

Of those that said they would not vote if a municipal election were held tomorrow, the main reasons sited were: not interested (46%), not in possession of necessary documents to register (14%), not registered (13%), disillusioned with politics (11%) and the vote would not make a difference.

When reasons for non-voting intention were aggregated by province, the Western Cape and the Northern Cape respondents were much less interested in politics than people from the other provinces. Most respondents not in possession of the necessary documents to register for voting were found in the Eastern Cape and North West.

Residents in urban formal areas tended to be less interested in voting than respondents in other areas. A fairly large number of respondents in the youngest age group (18-24 years) stated that they would not vote due to the fact that they did not possess the necessary documents to vote.

Clearly females, Coloured, Indians/Asians and White respondents as well as high LSM respondents were less interested to vote than the other groups.

A regression analysis was done regarding factors that might influence or predict voting in the coming municipal elections. It was found that factors that predicted whether a person will vote or not were: satisfaction with democracy, issues of service delivery, floor crossing issues, signage at voting stations, the time it took to vote, pleased with performance during IEC elections, satisfaction with national government, race and geographical area of residence. The more satisfied respondents were with democracy the more likely they were to vote in the coming municipal elections. Those respondents that agreed with the statements: “I will not vote because of a lack of services” and “Floor crossing discourages people to vote” were less likely to vote. Actual issues pertaining to voting on the day, i.e. the amount of money it cost to get to the voting station, the time it takes to queue, the time it takes at the voting station also have a bearing on whether people will vote or not. The more dissatisfied the respondents were with the time and money spent during voting process, the less likely they were to vote.

Persuasive Issues to Encourage Voting
When asked what might encourage voters to vote, about a third indicated that they would be encouraged to vote if they felt that their vote would make a difference. Just under a quarter (24%) indicated that shorter queues would encourage them to vote, followed by if there were a political party that they believed in (20%), they were more convinced that it was important to vote (15%), the polling station was closer and parties tried harder to get their votes (10%).

When responses were disaggregated by age it was found that not having to register and parties trying harder to get their votes were stated as potential pull factors by the younger (18-24 years) age group while shorter queues and closer polling stations were cited by the older group (65 years and over). KwaZulu-Natal and Gauteng in particular cited shorter queues as a pull factor. People from KwaZulu-Natal and North West would be more encouraged to vote if they felt their vote would make a difference. Free State people wanted to be convinced of the importance of voting. African respondents wanted to have shorter queues and to be closer to polling stations. Coloured people said they would be enticed to vote if there was a party that they really believed in. When analysed by environmental milieu, urban informal respondents wanted shorter queues and closer polling stations. Respondents in tribal areas said they would be more encouraged to vote if they felt it would make a difference and if they understood why it was important.

Ninety one percent of respondents said they ultimately decided who to vote for and were not influenced by any other factors. Three percent of respondents further indicated that a history of struggle influences their ultimate decision of who to vote for. Asked what respondents would do if they felt the party did not meet their expectations the younger respondents tended to be less loyal and less amicable to giving the party another chance. Older respondents were more likely to give the party a second chance.

Clearly Indians/Asians and White respondents are less tolerant of parties that do not meet their expectations while African respondents are more likely to give political parties a second chance. High LSM respondents are also less likely to give parties a second chance and would switch to another political party more easily than low LSM respondents.

When asked what respondents would do when they felt that they could not vote for the political party they normally support, almost half (49%) said they would abstain from voting, 33% would support another party and one percent would spoil a ballot paper. Africans were most likely (55%) to indicate that they would rather abstain from voting than support another party, while the other race groups were more willing to support other parties. Low LSM respondents preferred abstaining from voting while high LSM respondents indicated that they would shift allegiance more easily.

Voting Irregularities
Almost two thirds of the respondents were of the opinion that the counting and reporting of votes during the last national election was accurate. African respondents were more likely to feel that counting and reporting during the last national election was accurate while high LSM respondents were less likely to agree with this sentiment. A fairly large proportion (24.0%) did not know whether the counting was accurate or not. Only a mere six percent of the people indicated that they had experienced or observed irregularities during elections and a further 13.0% stated that they thought voting irregularities had occurred during elections since 1994.

Asked if the incidence of voting irregularities has increased or decreased since the 1994 election, six percent said they had increased a lot, 13% said it had increased somewhat, 19% were of the opinion that irregularities had stayed the same, eight percent said it had decreased somewhat and 15% said it had decreased a lot. A large proportion (46%) indicated that they did not know whether it has increased or decreased. White and high LSM respondents tended to believe that voting irregularities have increased since 1994.

Satisfaction with absence of irregularities during the elections
Almost 80% of those who had once voted felt that there were no irregularities during the elections in which they participated.

Voting Logistics
Access of voting stations
In order to determine the accessibility of voting stations, respondents were asked to indicate the extent to which they were satisfied with: the time it took to get to the voting station, the amount of money spent, the time spent in the queue to vote, the overall time it took to complete the voting process. In addition, respondents were also requested to indicate whether they felt that there were adequate instructions and signage in the voting stations. It should be noted that these questions were only posed to people who indicated that they had once participated in the national or municipal elections. The discussion is therefore solely based on the experiences and perceptions of those who have once voted.

Satisfaction with time to get to the voting stations
The results indicated that almost every person who once voted (91%) was contented with the time it took to get to the voting station. This is a positive indication that voting stations were well positioned and could be accessed with ease. More than nine out of ten people (90%) in all provinces indicated satisfaction in this regard, with Limpopo (95%) and Western Cape (93%) having the highest proportions. Conversely, although not high, Gauteng (9%) and Free State (8%) had the highest proportionate levels of dissatisfaction with time it took to get to the voting station. More than 90% in all different geotypes with the exception of people in tribal areas indicated satisfaction with time it took them to get to the voting stations. Satisfaction was slightly lower (87%) amongst residents in tribal areas. Satisfaction levels were highest amongst people in high LSMs (98%).

Satisfaction with cost to get to the voting stations
The vast majority (85%) of respondents indicated that they were satisfied with costs involved. Again, this illustrates that polling stations were easily accessible to the majority of the population, or that free transport was made available to the public to improve the accessibility of the voting stations. Even though satisfaction levels were evidently high in all provinces (satisfaction levels ranged from 94% in Northern Cape to 75% in KwaZulu Natal), people in KwaZulu Natal appeared to be the least satisfied (75%). Likewise, even though there were no significant differences in terms of the breakdown by geographic area, people in tribal areas appeared to be the most dissatisfied in this regard (27%). A slightly higher proportion of women (87%) indicated satisfaction compared to their male counterparts (84%).

Satisfaction with time it took to vote
South Africans were generally satisfied (84%) with the overall time it took to complete the voting process. Only nine percent indicated dissatisfaction. High satisfaction levels were evident in Limpopo (93%); conversely, people in Northern Cape seemed to be the most dissatisfied (77%). Urban informal areas also had the highest number of people (7%) who indicated dissatisfaction. Dissatisfaction was also more common (13%) amongst people in the low LSMs.

Satisfaction with signage and instructions
More than eight out of ten people (80%) indicated satisfaction as far as the signage and instructions in the voting stations in general are concerned. The highest proportion of those satisfied with signage and instructions was from the North West (98%) while the least was from the Eastern Cape (83%). It is interesting to note that highest proportion of those satisfied was found in rural formal areas (93%), where more than nine out of ten people indicated that they were satisfied. This was followed by 89% in tribal areas, 88% in urban formal areas and urban informal (87%) residents being the least satisfied. There were minor differences in terms of the satisfaction with instructions and signage by race (Asian (93%), Whites (86%), Coloured (89%), Africans (89%).

Satisfaction with time spent standing on the queues
With regard to the time spent standing on the queues, there were significant differences in terms of satisfaction across different provinces. A very high proportion of people in Western Cape (76%) indicated satisfaction with time spent on the queue. People in Mpumalanga (49%) were least satisfied compared to other provinces.

Conduct of political parties in the voting stations
Respondents were asked to indicate their overall assessment of the conduct of political party representatives, inside and outside of the voting stations. Nearly eight out of ten people (79%) felt that representatives of different political parties behaved well inside the voting stations. High proportion of respondents who indicated satisfaction with the conduct of political parties were reported in all the provinces ranging from 88% in Limpopo to a low of 67% in North West.

When asked to rank their satisfaction with regard to the conduct of political parties outside the voting stations, an overwhelming 80% expressed satisfaction. This was however much less in North West (65%) compared to other provinces.

Satisfaction with secrecy of the vote
The majority (92%) of South Africans (who had once voted) were satisfied that their vote was secret. Proportions of those who indicated satisfaction were high in all provinces; ranging from a high of 99% in Limpopo to a low of 82% in Gauteng. Almost every nine out of ten people in tribal areas were convinced that their vote was secret. People with the least satisfaction levels to this regard were largely found in urban formal areas. It is also evident that about a tenth (9%) of residents in urban formal areas was unsure. Africans and Indians/Asians had higher proportions (94%) of people who held this view, followed by the Coloureds (93%). White respondents (84%) were the least convinced that the vote was secret.

Satisfaction with representatives of the IEC
Respondents were further asked to indicate their level of satisfaction regarding the efficiency and impartiality of the IEC, friendliness of the IEC representatives, professionalism and helpfulness of the IEC representatives. In general, South Africans appear to be satisfied with the efficiency (94%), of the IEC, friendliness (94%), impartiality (85%) professionalism (91%) and helpfulness (94%) of staff of the IEC. Notably, Northern Cape appeared to be the least satisfied province with the efficiency of IEC (85.7%) compared to other provinces.

Access to voting stations
When asked to indicate the extent to which people were satisfied with the IEC in terms of access to the voting station, the majority (over 90%) indicated that they were satisfied. There were no significant differences between the different population groups and geographic areas. It should however be noted that dissatisfaction levels were highest amongst people in rural formal areas (8%). The overall low level of dissatisfaction in this respect was highest amongst people in the lowest LSM groups (7%).

Satisfaction with length of queues
Of all provinces, Mpumalanga (45%) had the highest proportions of people who indicated dissatisfaction with the length of the queues. Proportions of people who indicated dissatisfaction with queues were also largely from the rural informal areas (21%).

Satisfaction with other services of the IEC in the voting stations
More than nine out of every ten people (90%) indicated satisfaction with regard to the signposts in the voting stations, level of neatness and cleanliness, and the IEC’s ability to organise all necessary facilities.

Providing for the disabled
More than three quarters (78.0%) felt that the IEC does cater for the needs of the disabled during the elections. High levels of dissatisfaction were evident in Northern Cape (16%), followed by North West (11%).

Perceptions on whether elections were free and fair
A vast majority (89%) perceived that elections in South Africa were free and fair. This perception was highest in urban informal areas (95%) and lowest in urban formal areas (87%). Significant differences between the population groups were found when asked about ‘freeness or fairness’ of the elections. Dissatisfaction was most common amongst the Indian/Asians (14%), Whites (12%), Coloureds (9%) and least common amongst Africans. Uncertainty levels in this regard were also higher amongst Whites (20%).

Perceptions on whether elections get better every time
When asked whether they thought the organisation of elections had improved since the first elections in South Africa, just over three quarters (77%) of those who once voted thought that the organisation of elections has improved since the first elections in South Africa. These perceptions were most prevalent in Limpopo (84%) and least in the Eastern Cape (61%). These perceptions were also very common amongst Africans (82%). Perceptions that there had not been any improvements were highest in urban formal areas (12%). People in the highest LSMs were very pessimistic in this regard, as over a tenth (11%) indicated that said there had not been any improvements and about 13% indicated that the way elections were organised had remained the same. These perceptions were also common amongst Whites with 17% of them indicating that the way elections were organised in South Africa has stayed the same, and over a tenth indicating that there had not been any change.

Preference for whether voting should be on a normal working day, public holiday or weekend
Just over a third (36%) preferred voting to take place during the weekend. Only about 28% indicated preference for the normal working day. About 31% indicated preference for a public holiday. People who made a preference for a weekend were largely people in Mpumalanga (51%) and Gauteng (45%). Those who made a preference for a normal working day were largely people in Cape Town (52%). The majority of people aged 55 to 64 years (38%) and those aged (65%) and above were also in favour voting during the normal working day. People who were in favour of the weekends were largely people aged 18 to 24 years (42%) and 25 to 34 years (42%).

Media and Information
Asked to indicate whether they had heard of the Independent Electoral Commission (IEC) or not, eight out of ten respondents indicated that they had heard of the Independent Electoral Commission (IEC). Only just over a fifth (21%) of respondents in Limpopo province indicated that they had not heard of the IEC. The same proportion of respondents from the rural areas (21%) indicated that they had not heard of the IEC. These respondents who had never heard of the IEC tended to be 65 years and older.

A slightly greater proportion of females (14%) than males (12%) indicated that they had not heard of the IEC. Of the four population groups, Coloured people (15%) surpassed Africans (13%) in saying that they had not heard of the IEC.

Respondents with little or no schooling (25.7%) were more likely not to have heard of the IEC than those respondents with higher schooling like Matric and above. A higher proportion (16.0%) of those in the lowest LSM group than in the higher LSM groups had not heard of the IEC. It would seem that respondents, who were currently unemployed, whether looking for work or not, were more likely to say that they had not heard of the IEC.

In terms of sources of information about the IEC, radio, television and newspapers, in that order, were the most popular sources of information about the IEC. Posters were also popular sources of information about the IEC.

Respondents from Limpopo, Mpumalanga and the North-West seemed to use radio as the most important source of information about the IEC. Respondents from the Western Cape, Gauteng and Mpumalanga seemed to depend more on television as a source of information. Newspapers are largely utilised in this regard in Gauteng, KwaZulu-Natal and the Western Cape.

Very young respondents between the ages of 18 and 24 as well as those between 45 and 54 years tend to use television as a source of information. Slightly older respondents between 55 and 64 years tend to read newspapers as a source of information on the IEC. Also respondents who rely on radio as a source of information on the IEC tended to be 65 years and older.

In terms of race, Whites used television as their source of information on the IEC, with just under a third (31%) stating that this was the case. Africans, at a little over a quarter (30%) preferred radio as a source of information on the IEC. The Coloured respondents (26%) tended to rely on newspapers as a source of information on the IEC. The higher the LSM measure, the higher the reliance on television and newspapers as sources of information. Conversely, the lower the LSM measure, the higher the reliance on radio as a source of information.

It was only in the North-West and Limpopo that respondents felt that there was too little or far too little information on the IEC. Indeed, it was respondents from the tribal areas who felt that there was too little or far too little information on the IEC or the services provided by the IEC. There were slightly greater proportions of females than males who felt that there was enough information about the services provided by the IEC. It was respondents at lower educational levels, especially in grades 1 to 3 who felt that there was little or far too little information on the services provided by the IEC. This could be attributed to the low level of literacy and therefore the inability to access available information.

Upon being asked where they generally get information about voting, television, radio, newspapers and posters, in that order, were the most frequently cited media. These account for three-quarters (75%) of all sources of information about voting.

Urban dwellers tend to get information about voting from television and newspapers. Tribal and rural inhabitants tend to get this information from radio. The preference for posters on information about voters seem to be largely confined to urban formal areas.

Slightly more females (31%) than males (29%) used television as a source of information about voting. However, more males than females relied on newspapers as a source of information about voting. Females tend to use more contacts like friends, family and acquaintances than males, though the differences were not great.

Respondents with high education qualifications get information about voting from television as well as newspapers. In contrast, respondents with lower qualifications tend to get information on voting from radio, contacts and IEC voter education officials.

Asked about where they generally find information about voting procedures, most indicated television as their source of information. However, radio is the second most important medium to inform on voting procedures.

Television is a popular source of information on voting procedures in the Western Cape and Gauteng. The popularity of radio is noticeable in KwaZulu Natal, the Eastern Cape and Mpumalanga. Television is also popular among the young generation (aged 18-24) as well as those in the 45-54 age category. In contrast, radio as a source of information about voting procedures is common among the 65 years and older citizens. The Coloured, Indian and White population groups tend to get information about voting procedures from television whereas Africans get the same from radios. In the same vein, higher LSM groups get information about voting procedures from television whilst lower LSM groups rely on radio.

Asked to indicate whether they had too little or too much information about how to vote, most respondents thought they had enough information. Also, regardless of educational level, most respondents thought that they had enough information about how to vote, especially those with higher educational qualifications.

Respondents were also asked about how pleased or displeased they were with the performance of the IEC in previous elections. More than three-quarters of all respondents appeared to be pleased or very pleased with the performance of the IEC. This was shown across provinces, age categories, race groups and lifestyle measures (LSM).

In conclusion, it does seem that the IEC and its services are well known and that a variety of media are utilized to obtain this information. The only emphasis coming from these findings is that the IEC should target its interventions in the sense that lower LSM, Africans, and older respondents seem to prefer radio as their source of information about the IEC whereas higher LSM, largely White and relatively young respondents seem to prefer television, newspapers, and posters as their source of information.

Satisfaction with institutions
Public satisfaction may loosely be explained as the extent to which the populace in a particular country are generally satisfied about the political process, e.g. how elections are conducted. Public satisfaction about the institutions that facilitate the democratic process in a country is one of the greatest expectations political rulers often have about the electorate and their prospective supporters, especially before an election. One way of understanding how public perceptions about satisfaction (or otherwise) with public institutions are formed is to examine their expectations. According to political scientist Heather Deegan , “At times, public expectations of government can be high, in fact, over exaggerated, especially with regard to the delivery of certain ‘economic goods’. A newly enfranchised polity may hold unrealistic, almost utopian views of governmental capabilities which, if unfulfilled, render the electorate increasingly vulnerable to disappointment and disenchantment.” There are signs that as in other newly democratising societies in the world the South African electorate, especially at local government level, have experienced such disappointment and disenchantment whenever their expectations about certain public institutions were not realised. Thus, the electorate’s perceptions about these institutions might be linked to their delivery of services in their various areas of responsibility.

The results with regards to level satisfaction with public institutions revealed that there is a striking contrast between satisfaction levels for national and provincial levels. Unlike the high levels of public satisfaction recorded for national government, the results of the IEC/HSRC survey indicate that the respondents were almost evenly divided between those who were satisfied or highly satisfied with local government, in comparison to those who said they were dissatisfied or highly dissatisfied with it. Here again, there might be different variables at play that influenced the respondents’ perceptions.

Another crucial element regarding the respondents’ perceived satisfaction with local government is difference by age category. There appears to be no major difference among those who said that they were ‘strongly satisfied’ about local government, save for those aged 65 and above whose perceived level of satisfaction is lower than that of the other age groups. (Amongst the respondents who reported that they were satisfied with local government, only the 18-24 age group scored low (at 33%). This suggests that the youth in this age group were less positive about the performance of local government in comparison with the other age groups. Similarly, the respondents in the 18-24 age category who stated that they were ‘strongly dissatisfied’ with local government recorded the highest response rate (22%).

The level of satisfaction with public institutions might also be seen as an indicator of their role in a country’s democracy. One way of showing this is to examine the extent to which voters are satisfied with political parties, given that the parties’ role in the democratic process would be expected to be much higher compared to that of other institutions. The South African public across the different age groups are on the whole satisfied with the role of political parties in the country, even though satisfaction levels are below 50.0%.

The results further indicated that the respondents across all age groups were satisfied with political parties, although this sharply contrasts with the low figures for the ‘strongly satisfied’ category (which was well below 10%), except in the 25-34 age group. Further analysis indicates that those who were satisfied with political parties were mostly in the first four age categories, that is, 18-54 years. Thereafter, satisfaction with political parties amongst the respondents in the older age groups declined gradually.

The role of the IEC as an independent and impartial facilitator and monitor of all elections in the country was also recognised by the respondents. The results showed that the respondents were happy with the IEC’s role in running elections, given the high number (52%) of those who reported that they were satisfied, compared to the four percent who were ‘dissatisfied’ or the 11% who were ‘neither satisfied nor dissatisfied’. Such satisfaction clearly differs by age as those in the 18-24 age group scored the lowest level of satisfaction amongst those who were ‘highly satisfied’ with the work of the IEC.

Access to services

In order to get an in-depth understanding of some of the reasons why people might chose to vote or not to vote, respondents were presented with statements outlined below on which to indicate whether they agree or disagree.

Just under three quarters of South Africans (74%) agreed that it is important to vote whether one has access to basic services or not. Males were generally in support of this statement (77%). Proportions of people who held similar sentiments were mostly residents of the urban formal (79%) or rural formal (75%) areas. Agreement levels were also high amongst Whites (77%) and very low amongst Indians/Asians (25%) and Africans (20%). Provincially, more than seven out of ten respondents in Limpopo (80%), North West (77%) and Western Cape (77%) were mostly in agreement with the statement. Conversely, Mpumalanga (28%), Free State and KwaZulu-Natal (21%) had higher proportions of people who disagreed.

Slightly lower levels of agreement are evident (69%) with over 20% of the population indicating disagreement with the statement that it is important to vote even if you do not have access to basic services. Agreement levels were higher amongst people in Western Cape (74%). Proportions of people who did not believe this were higher amongst people in Mpumalanga (32%) and in the tribal areas (71%). The majority of males (71%) also felt that it is necessary to vote even if you do not have access to services. Whites were generally supportive of the statement (70%) compared to other race groups.

A vast majority (76%) of the respondents agreed to the statement that it is important to vote in order to have access to basic services. This was almost exclusively the case in Limpopo (91%). High proportions of people who shared a similar sentiment were also found in North West (82%) and Northern Cape (83%). People in support of the statement were also largely from tribal areas (87%) or urban informal areas (82%). Likewise, the majority of males concurred that it is important to vote in order to have access to basic services. Africans (81%) were also mostly in support of this statement.

More than half (55%) of the respondents disagreed that it is not important to vote because they do not have access to services, with just a third of the population concurring with the statement. Proportions of people who felt that it is not important to vote because they do not have access to services were largely from Mpumalanga (46%), Eastern Cape (41%), Gauteng (46%), or North West (35%). The Eastern Cape had an equal split of those who agreed (41%) and those who disagreed (46%). People in support of this statement were also mostly from the tribal (40%) and rural formal (30%) areas.

The highest proportions of those who felt that it is still important to vote even if one does not have access to services were largely people in Free State (70%), Western Cape (64%) or KwaZulu-Natal (66%). Urban formal areas (57%) and rural informal areas (55%) also had larger proportions of people sharing a similar sentiment. High proportions were also found amongst Indians (71%) and Whites (63%).

The results on the link between willingness to vote and delivery of basic services were very interesting when disaggregated by some of the key demographic variables. While the majority of the respondents who were from Limpopo, North West and Western Cape, or were White, and from urban formal areas felt it was important to vote whether one has access to basic services or not, respondents who were Africans, hailed from Limpopo, tribal areas and urban informal felt it was important to vote in order to have access to basic services.

The Impact of the Social Capital on the Democratic Process
The term ‘social capital’ increasingly appears in the contemporary literature of several disciplines, e.g. development studies, politics, sociology, law, social anthropology and public administration. The World Bank (1999) states that ‘social capital’ “refers to the institutions, relationships, and norms that shape the quality and quantity of a society’s social interactions. Social capital is not just the sum total of institutions which underpin a society – it is the glue that holds them together.” The notion of social capital introduces a useful way of understanding the role of civil society in the governance process especially at local government level. On the one hand, social capital is an important concept for developmental local government. On the other, “Local government has an essential role to play in building social capital and a sense of common purpose so that local solutions to problems may be found…”. One way of identifying such common purpose might be through the participation of local communities and individuals in various groups such as religious groups, trade unions, political parties, business, sports and others associations. These institutions are often regarded as providing regular opportunities for consolidating social relations, which might enhance social solidarity, cohesiveness or national unity.

Thus, many respondents in the current IEC/HSRC data indicated that they were ‘active members’ of a religious group, especially those aged 25 and above. The highest proportion for those aged 65 and above who belonged to this category was 73%. In general, the proportions for the different age groups that confirmed active membership in a religious body ranged from 52% to 73% from the youngest to the oldest age groups, respectively. These figures contrast sharply with the respondents in these age categories that indicated that they were not members of a religious group, whose percentage scores ranged from 17% to 34%.

In terms of membership of a political party, there was a striking difference between those who said they were active members of a political party and those who said they were not members across the different age groups. Most important, about 84% in the 18-24 age group, stated that they were ‘not a member’, a significant number given the active participation of people in this age group shortly before the 1994 democratic elections. Overall, the highest proportion of respondents stated that they were not members of political parties, and this ties in closely to the high number of respondents who never attended election campaign rallies.

Attendance of community meetings called by a local councillor, public marches and government imbizos; such gatherings are expected to contribute to the consolidation of the social capital in a country. However, given the low attendance figures revealed by the survey, it is not clear how far such gatherings contribute towards the consolidation of the social capital in this country. Of the respondents who answered questions on these issues, almost 52.0% stated that they never attend community meetings called by a local councillor; whereas an even higher percentage stated that they never attend government imbizos (78%) and public marches. Such figures are in striking contrast to the fact that barely a decade ago many residents in local communities were very often seen to be active in such gatherings, especially in the run-up to the 1994 democratic elections. Another striking feature of the data is that a surprisingly high percentage of respondents reported that they had never attended a voter education workshop.

Clearly, the small percentage (4%) admitting to attending a voter education workshop suggests that the respondents either did not see the importance of these workshops or felt that they played a minimal role in the preparations towards elections. However, it may well be that many respondents were now aware of the role of the IEC and party election officials to the extent that they felt they were adequately informed of election-related processes. But how far this contributes to enhancing the social capital, given that less than 15% across all age categories stated that they ‘sometimes’, ‘often’ and ‘sometimes’ attend voter education workshops, is not clear and perhaps indicates the limited usefulness of variables like attendance at voter education workshops to assess the consolidation of the social capital in a country.

Overall, respondents’ satisfaction with public institutions generally indicates their confidence that such institutions are important to democracy and the improvement of their overall situation. However, such perceptions are also influenced by various factors including the nearness of an election, which often sees intense political campaigns and rallies raising various issues for debate and inclusion into the national public discourse. Similarly, the importance of the social capital (i.e. ties that bind people in the different associations) also has to be seen in the context of a country that has just completed its first ten years of democracy. Given the short period of its new dispensation, and the fact that South Africa still struggles to encourage the development of national unity, citizenship and social solidarity across its different communities, the current data suggests a slight improvement in this regard given that socio-political transformation often takes a long time before visible signs of the significance of the social capital may be seen. As McEwan (2003: 480) indicates, “Processes of democratic transformation are still unfolding in South Africa. Only time will tell whether experiences here will point towards workable alternative notions of citizenship, towards locally-rooted and participatory democracy that can affect “good governance”.

General Conclusion
The overall results of the 2005 HSRC / IEC is positive since most of the respondents indicated that they satisfied with the way democracy works in South Africa. Race still played a major role in people’s evaluations whether they satisfied with the way democracy works in South Africa. More Africans (20%) were “very satisfied” with the way democracy is working than was the case amongst coloureds (6%), whites (4%) and Indians/Asians (2%). Indians/Asians (31%) and Whites (23%) also recorded high levels of dissatisfaction with the way democracy is working. Although unemployment is high and poverty seem to be growing most South Africans indicated that their personal life as well as that of fellow South Africans has improved over the last five years. Based on the positive results from the survey and particularly the section on registration as a voter, we can state with optimism that the percentage of people who will register to vote for the upcoming municipality elections would be higher than previous years.

Recommendations
Identity Documents:
• Although differences were very small, Africans (4%), males (4%) and people with low LSMs (4%) were slightly more likely not to be in possession of a green-bar coded ID book than were people of other races, females and high LSM people. Government should therefore strongly consider giving financial assistance to those who are unable to afford the costs involved in the application for identity documents. The lack of a valid identity document not only excludes one from elections but also from obtaining other social benefits provided by the state.
• The possession of identity document other than the green-bar coded is more common among the older population since 50% in the 45 – 54 age category indicated that they have such a document.
• In terms of race, greater proportions of Whites (33%) and Indians/Asians (24%) said that they are in possession of an identity document other than the green-bar coded document than was the case amongst the other races.

Information campaigns of the IEC:
• Should be concentrated in rural areas, especially Limpopo and the North-West, and among elderly respondents, more specifically, pensioners.
• Should emphasise female, lower education, unemployed, Coloured, and more poverty-stricken respondents, in no particular order.
• Should emphasize younger respondents who should be targeted through television whilst older respondents in their middle ages should be targeted through newspapers. Senior citizens should be targeted through radio media.
• Should target Whites through television; Coloureds should be targeted through newspapers. Africans should be targeted through radio.
• With regards to people with higher standards of living should be targeted through television whilst those with lower living standards should be targeted through radio.

Media Usage:
• Television, radio, and newspapers, as the most popular sources of information on voting should get about three-quarters of advertising spend.
• On information about voting, urban areas should be targeted through newspapers, posters and television whilst rural areas should be reached through radio.
• Information about voting in terms of gender should be in such a way that males who prefer newspapers are catered for whilst female, who prefer either television or more informal contacts like friends, relatives and neighbours are catered for as well.
• Since higher status respondents tend to prefer television and newspapers and lower status respondents tend to prefer radio and informal contacts with friends, relatives and neighbours, higher status respondents should be reached through more formal media like television and newspapers whilst others should be reached through Imbizo’s and informal channels like traditional leaders.
• Since television and radio are the most popular media used to find out about voting procedures, these two media should surpass every other media when conveying information about voting procedures. Respondents who are White, Coloured or Indian/Asian, young or middle aged, live in Gauteng or the Western Cape, are more likely to find out about voting procedures from television. On the other hand respondents who are African, older, live in Kwa-Zulu Natal, the Eastern Cape or Mpumalanga, are more likely to listen to the radio.

Floor Crossing:
• More people from the Eastern Cape and Limpopo, more specifically Africans, should be taught what floor-crossing is because they do not know what it is.
• Floor crossing seem to be a voting deterrent and the majority of respondents were against floor crossing. In particular, the Western Cape, Free State and Northern Cape respondents as well as the high LSM respondents and the Coloured, White and Andian/Asian were opposed to floor crossing.

Impact of lack of service delivery on voting:
• The government and IEC should be mindful of the fact that the lack of service delivery might impact on voting behaviour. This was particularly prevalent amongst the younger age group where a large percentage agreed that they would not vote because of a lack of service delivery. This issue needs to be monitored since it could escalate and impact on voter turnout.

The role of women in elections:
• Women tended to be more sceptical, less interested and eager to participate in political issues than their male counterparts. This could be related to a feeling that their needs are not sufficiently addressed by the male dominated party political structures.

Non-participation and voter turnout in the upcoming elections:
• The main reasons given for non-participation in the forthcoming elections were no interest, no documentation, not registered. When disaggregated by some of the demographic variables, it was found that most of the people who were not interested in politics can be profiled as being from the Western Cape; from formal areas; female, Coloured, followed by Indian/Asian and White respondents. Particular effort should be made to encourage these groups to vote and to address specific concerns that discourage them from voting. A large proportion of respondents that do not possess necessary document to register were found in the 18-24 year old category. Special campaigns should be initiated to furnish these respondents with the necessary documentation. It was also evident that almost a third of respondent in KZN indicated that they would not vote since they were not registered. Registration campaigns should be increased for KwaZulu Natal.
• Analysis showed that IEC performance does impact on voting behaviour. Good and efficient signage at voting stations, shorter queues, less time and money spend during the voting procedure and feelings of satisfaction with IEC performance encourages people to vote. Professional and efficient conduct during elections by the IEC is crucial for voter participation.
• When responses to the question what would encourage people to vote were disaggregated by age it was found that not having to register and parties trying harder to get their votes were stated as potential pull factors by the younger (18-24 years) age group while shorter queues and closer polling stations were cited by the older group (65 years and over). KwaZulu-Natal and Gauteng in particular cited shorter queues as a pull factor. People from KwaZulu-Natal and North West would be more encouraged to vote if they felt their vote would make a difference. Respondents from the Free State wanted to be convinced of the importance of voting. African respondents wanted to have shorter queues and be closer to polling stations. Coloured people said they would be enticed to vote if there were a party that they really believed in. When analysed by environmental milieu, urban informal respondents wanted shorter queues and closer polling stations. Respondents in tribal areas said they would be more encouraged to vote if they felt it would make a difference and if they understood why it was important. The IEC should take all the issues above into consideration when campaigning towards the elections. Although some generic approaches to media briefs could be followed, specific approaches addressing specific issues relating to specific sub populations might have a more desired effect.
• The Coloured, Indian/Asian and White respondents indicated that they would support another party if they could not voter for their preferred party while a large contingent of African respondents indicated that they would rather not vote than vote for another party. Disillusionment with preferred parties would therefore lead to a lower voter turnout rather than a voter shift amongst a large proportion of specifically the African population. These findings should be taken into account when low voter turnout is recorded.

Voting irregularities:
• Almost a quarter of respondents recorded a “did not” when asked whether the counting and reporting of the April 2004 elections was accurate. Almost half stated that they did not know if voting irregularities had increased or decreased since 1994. Since perceptions regarding efficiency do influence voter participation the IEC should make a special effort to report in successes in these domains.
• Since voting irregularities seem to be perceived more by Eastern and Northern Cape respondents IEC officials should be especially proactive in these provinces.

Race:
• Race still appears to play a very important role in the lives of people of South Africa.

Youth Issues:
• The concerns of young people should be address with great care since they are the future custodians of our well earned democracy. Although the majority of the youth indicated that their lives have improved over the last five years, an equally big proportion is unhappy with the level of service delivery. The recent municipal demonstrations has showed that the youth and in particular the unemployed youth have participated with great vigour in these protest actions.
• Large proportions of the youth are still not in possession of a green-bar coded identity document.