[Ugnet] Uganda's incredible shrinking voters register

[musamize]

Ssemakula <[EMAIL PROTECTED]> wrote:

Uganda's incredible shrinking voters register

By OGOLLA DICKSON
Special Correspondent

One of the riddles of the Ugandan elections is the fact that the number of registered voters has reduced over the past five years.

In 2001, there were 10.7 million registered voters: 29 per cent from the western region, 29 per cent from central region, 27 per cent from the east and 15% from the northern region.

Yet the register used for last month's elections had 10.4 million people reflecting a reduction of 300,000 voters or 2.8 per cent.

Is it not amazing that Uganda has not registered new voters in the past five years? What is the integrity of the voters roll?

The Electoral Commission says it cleaned up the register to remove ghost voters and to deal with the problem of double registration in the 2001 voters roll. But how does one explain the lack of growth? According to statistics from the Uganda Bureau of Statistics, the population has been growing at a rate of 3.4 per cent for the past 10 years.

This means that every year, the voters roll should grow by 3.4 per cent and on a non-compounded scale gross 17 per cent over a period of five years. The register should thus have 12.5 million registered voters. Instead, it has 10.4 million.

What could have happened? One, the base number used in 2001 was inaccurate and indeed the Electoral Commission has admitted to this. Two, the official population growth figures are inaccurate.

How could a register that is not accurate affect the results of the just concluded elections? First, it can affect the logistical planning of the Electoral Commission. If you don't have accurate figures, you will be wrong in your estimate of number of officials hired, number of ballot papers printed, litres of indelible ink ordered and the budgeting process in general.

An inaccurate voters register opens loopholes for vote rigging. Since nobody knows the true number, no one would know the right number of votes per polling station. Neither is it possible to accurately calculate the turnout. It makes it possible for someone to come up with an arbitrary figure of votes cast. Clearly, a shrinking voter register defies all statistical logic.

Consider another statistical riddle in the Ugandan elections. According to the voters roll, 10.4 million people registered to vote out of a population of 28 million. According to official statistics, 55 per cent of the population falls below the voting age – 15.4 million are ineligible to vote. 

Kenya has a population of 30 million with 50 per cent falling below the voting age in the 2002 general elections. Those who qualified to vote were 15 million yet only 10.4 million or 69 per cent registered to vote.

Even more recently in the 2005 Kenyan constitutional referendum, there were 11.5 million registered voters of a possible 16 million. (71 per cent vote registration).

What inference do you make from this? First, Uganda has a smaller population yet had the same number of registered voters in 2001.

Second, Uganda has a higher ratio and absolute number of those who fall below the voting age than Kenya, yet it still has the same number of registered voters.

Third, Kenya’s voter registrar grew by 10.6 per cent between 2002 and 2005 while Uganda’s figure shrank by 2.8 per cent between 2001 and 2005.

Fourth, Kenya has a lower voter registration rate (71 per cent) than Uganda (82.5 per cent). 

An even more interesting statistic in the Ugandan elections is the nature of voter distribution. 

As a statician, if you go by the measures of what is known in jargon as the central tendency – mean and mode – you are not likely to get a good picture on the nature of vote distribution.

One is better off looking at measures of dispersion – percentiles, standard deviation and variances. Why? Because what you have here is a country of 69 districts where 24 of the districts account for 58 per cent of all the votes.

Some four districts account for almost one in every five voters in the country. These four districts – Kampala, Masaka, Wakiso, Mukono, all from the central region – account for 1.9 million votes compared with 1.5 million for the whole of the northern region. 

The mean vote per polling station in these four districts is 633, which is much higher than the national mean of 553, again showing the uneven nature of the distribution in the whole country. 

Indeed, any of the two regions, combined – east and central, central and west, or west and east – were enough to provide the n eeded 50 per cent for a presidential win. 

The north, with 15 per cent of the votes, is statistically marginalised. 

In other words, the electoral data frame in Uganda is not designed for fairness, not even for equality. It is strongly weighted. Perhaps there is a case for proportionate representation.

The overall voter turnout in Uganda was 68 per cent, with almost 7 million people turning out to vote out of the 10.4 million registered voters.

The turn-out from the four key districts was on average 61.2 per cent. Kampala had 56.5 per cent, Mukono 61.9 per cent, Wakiso 58.8 per cent and Masaka 67.5 per cent. 

It is quite evident that the mean is made meaningless by the wide differences. 

The turnout in these four districts was however much lower than the national average, except for Masaka. 

The overall registration ratio (82 per cent) and the overall turnout (68 per cent) are much higher than those in Kenya. 

In the 2002 K enyan elections, 71 per cent of those who were eligible registered and 56 per cent of those who registered actually turned out to vote. 

Either Ugandans have a higher civic commitment than Kenyans or the statistics are inaccurate. 

So how did they vote? The overall official tally showed Museveni had 4,078,677 votes (59.3 per cent) and Besigye 2,570,572. 

The rest got a total of 3 per cent, the implication being that even a combined opposition would not have received the required 50 per cent to force a run-off.  The voting pattern by district put in the context of voter turnout presents an interesting picture. 

Museveni beat Besigye in 49 out of the 69 districts. In the 20 districts in which Besigye won, he beat Museveni with a mean percentage score of 35 per cent, which is lower than his overall score. 

These twenty districts had 3,161,504 registered voters or 30 per cent of the register. Only seven of these districts are in the list of 24 that account for 58 per cent of the registered voters. 

---

Yahoo! Mail
Use Photomail to share photos without annoying attachments.

_______________________________________________
Ugandanet mailing list
Ugandanet@kym.net
http://kym.net/mailman/listinfo/ugandanet
% UGANDANET is generously hosted by INFOCOM http://www.infocom.co.ug/


The above comments and data are owned by whoever posted them (including 
attachments if any). The List's Host is not responsible for them in any way.
---------------------------------------