Uganda’s National Resistance Movement has always claimed to be the party of the rural poor. It is a claim that has its roots in Yoweri Museveni’s early radical politics; and since the 1990s, it has been the standard riposte of the NRM to political challenges — from Paul Ssemogerere’s presidential bid in 1996 to the current election campaigns.

Results from a recent survey designed by us suggest that there is some truth to this claim. Of course, most people in Uganda are classified as (rural) according to the country’s population census (the categorisation is based on population density), and so the majority of support for all the candidates is from rural voters.

Nevertheless, the survey shows that rural Ugandans are significantly more likely to describe themselves as NRM voters; conversely, they are less likely to describe themselves as supporters of Kizza Besigye, the candidate of the Forum for Democratic Change (FDC) — who does better in urban areas. And they are even less likely to say that they will vote for Amama Mbabazi, the new challenger, who split off from the ruling party last year.

Some may be inclined to shrug off these results; and it’s true we cannot ever be entirely sure of every individual answer in this kind of survey. Confidentiality is always promised, but people may nonetheless fear to say that they support an opposition candidate — especially if they live in an area where the ruling party has strong support.

Yet the recent survey was nationally representative and carefully conducted through face-to-face interviews, and draws together information from 2,600 respondents.

The margin of error is relatively small, at +/-2 per cent, and so, even if we accept that fear may skew some responses, the pattern is clear: according to the survey, rural Ugandans are 22 per cent more likely to vote for the NRM than urban voters, which seems to bear out the NRM’s sense of its own support.

Of course, this still leaves the question of why and how the NRM generates this support. To some extent it is because the party has made a conscious effort to deliver services and patronage rural areas.

However, it also has other sources. Opposition parties tend to do better in urban areas precisely because there is higher population density and so they can reach more people. Mr Besigye may be able to hold rallies all over the country, but he lacks a functioning political machine in many rural areas.

At the same time, rural voters are less likely to have access to diverse media sources, and are more likely to be dependent on government services than their urban counterparts. They are, therefore, easier to keep in the dark about government failures and to persuade about government success.

Thus, although the NRM knows its support base well, there is a question mark over how well its support base knows the NRM. It is also more difficult for civil society organisations and others to monitor and prevent intimidation and violence in rural areas.

The survey also suggests that women are 15 per cent more likely than men to support Museveni. The difference in this case is even more striking. A substantial majority of those who say they intend to vote for NRM are women; but the majority of those who say they would vote for Mr Besigye and Mr Mbabazi are men.

One of the challenges of survey data of this kind is that while the numbers can show patterns, they don’t always explain them.

The NRM is likely to argue that its emphasis on stability and order explains the party’s apparent strength among female voters.

There may be some truth in this: Surveys in other countries have often found that women hold more “conservative” (with a small c) views when it comes to voting — in other words, that other things being equal, they are more likely to vote for continuity than change. Other explanations are also possible; but whatever the cause, the pattern is very clear in the survey data.

There are other interesting patterns too, that both ruling party and opposition candidates may want to think about. The first is that both major opposition candidates draw a significant amount of support from younger voters.

Those who say that they support the NRM are spread across every age category; and, broadly speaking, someone over 50 is more likely to say they will vote for Museveni. The figures are quite striking: 75-year-olds are 28 per cent more likely to vote for the NRM than 18-year-olds.

This implies, of course, that younger voters are more attracted to the opposition: Overall, 18 year olds are 17 per cent more likely to vote for the FDC than 75 year olds. This tells us that the more radical messages about change that are Mr Besigye’s stock in trade are resonating with the youth.

This group also did not experience the conflicts of the 1970s and 1980s, and so are less responsive to Museveni’s claim to be the only candidate who can deliver peace and security.

Of course, this is good news for the opposition as the Ugandan population is getting younger and younger, but it will only translate into electoral success if these voters can be registered and persuaded to vote. Historically, this has proved to be a major challenge.

The survey has some other lessons, and not only for the candidates. The responses show a strong public preference for democracy, and for free and fair polls.

Despite having suffered many problematic elections, 72 per cent of Ugandans said that democracy is preferable to any other form of government, and an overwhelming majority of the population disapproves of one-man and one-party rule. This support extends to free and fair elections: over 80 per cent of the population believe that the use of election gangs, ballot box stuffing and buying of ID cards is not only wrong, but should be punished.

Election monitors are also looked upon favourably. More than 80 per cent of respondents agreed that domestic election observers were a good thing, while some 78 per cent of Ugandans feel that domestic election monitors have helped to improve the quality of elections.

This means that the NRM has a strong incentive not just to win the polls, but to win them fairly and — which is not quite the same thing — to be seen to win them fairly.

Another finding of the survey is that voters who believe that Uganda is more of a democracy are more likely to say that they plan to vote for the ruling party. Those who see the country as less of a democracy, break more heavily for the opposition.

With Museveni still in charge these issues may not necessarily undermine the NRM’s hold on power in this election, but they will become increasingly important when Museveni steps down, especially if this triggers further splits within the government.

The way that the NRM handles the transition to come will therefore be critical. A recent Afrobarometer survey in Uganda found that 85 per cent of people supported presidential term-limits.

President Museveni initially built Uganda’s “no-party” democracy on the principle of individual merit, and the notion that the capacity of the candidate mattered more than party identity.

If he tries to remove the age limit from the Constitution to stand for yet another term, and then seeks to impose a candidate whose qualities are questionable, the impact on public sympathy for the NRM is likely to be significant.

Nic Cheeseman, University of Oxford, Gabrielle Lynch, University of Warwick and Justin Willis, University of Durham; [email protected]; @GabrielleLynch6