|
|
Post by Onyango Oloo on Oct 21, 2005 21:50:08 GMT 3
Poll Shows Kenyans Rejecting New Constitution
By Katie Nguyen, Reuters
Nairobi - The first opinion poll on Kenya's constitution referendum next month shows President Mwai Kibaki heading for defeat because most people think the government-backed document is flawed and would be difficult to amend.
The poll - released as the "Yes" and "No" camps formally launched their campaigns on Friday in a contest that has sparked violence and split the east African country - showed the "No" camp 10 percentage points ahead.
The Steadman/Gallup International survey of more than 2 000 people across the country said 42 percent of registered voters opposed the first complete overhaul of the constitution since independence from Britain in 1963.
The percentage in favour was 32 percent, with those undecided at 22 percent. Kenyans will vote on November 21 in the first referendum in their history.
Elected in December 2002, the government, which promised to deliver a new constitution within 100 days of coming to power, is leading the "Yes" campaign with the symbol of a banana.
The opposition, and a "rebel" party in the ruling coalition, is pushing the "No" vote with an orange.
"Both sides have made this a vote of confidence in the government - that's of course why the stakes are so high," said Tom Wolf, a political analyst working for Steadman.
Analysts say the undecided voters could swing the end result either way in the lead-up to the referendum.
"The gap is not insurmountable by any means," Wolf said.
Critics, including some of Kibaki's ministers, say the draft charter fails to curb the sweeping powers of the president which they fear promote nepotism, tribalism and corruption.
But Kibaki's supporters insist his powers have been clipped in the proposed text.
Many Kenyans say politicians have lost sight of the issues as the constitutional battle ignites tribal hostilities and turns into a major political showdown ahead of 2007 elections.
Informal campaigns have been marred by riots, stone throwing, the arrests of journalists and politicians and fist-fighting among lawmakers.
On a bright note for Kibaki's supporters, the poll showed 33 percent of voters would endorse him as president compared to 29 percent in a July poll. In contrast, opposition leader Uhuru Kenyatta's popularity rating fell to 17 percent from 28 percent.
Reuters
Published on the Web by IOL on 2005-10-21 10:15:56
© Independent Online 2005.
|
|
|
|
Post by job on Oct 22, 2005 2:38:29 GMT 3
Some statistical flaws in poll methodology.
----------------------------------------------------------
1) 52 out of 74 districts sampled. More of the 12 districts excluded, fall generally in NO (Orange) zones. Stedman should have injected a little more cash to do a complete national survey in all districts.
2) While the urban:rural sampling ratio of 35:65 is fair, proportionate sampling with respect to population densities ought to have been considered. It does not make sense to sample 50 respondents in Garissa & 50 respondents in Gucha for instance.
3) On the Presidential poll, it is outrightly inaccurate to sample respondents in Kibaki's home district of Nyeri, then fail to sample respondents in the home districts of other contenders. Notice failure to sample respondents in Mwingi (Kalonzo's home district) & Kangundo; & failure to sample respondents in Thika (Uhuru's home district) & neighboring Gatanga. This would significantly alter the poll outcome portrayed in statistically significant margins.
4) Given the close-ended nature of the forthcoming referendum question, i.e either YES or NO, the % of undecided voters in the Steadman poll, is apparently too high, ( 22%). In more scientific polls, this may indicate a methodological flaw from numerous factors that the surveyors could possibly need to address next time. If you give three options of responses in the question, i.e YES, NO, or UNDECIDED, and a different pollster gives only two responses YES, or NO (as would be offered on the referendum day), the later pollster may reflect more accurate depiction.
5) The referendum question in itself may induce or influence short term bias with regard to prefered presidential choices, hence concurrent sampling of the presidential poll inherits innate
errors which may be divorced right after the referendum day.
For instance, somebody may currently prefer any candidate other than Kibaki just because they dislike the Wako draft, or another may just prefer Kibaki since they percieve the other contenders as Kibaki sabotours with regard to the Constitution. Thus the presidential poll question was bound to be skewed,..... this is not the right time to get an accurate Presidential poll.
Lastly,
a) I contend that the overall NO lead is far greater than portrayed. (statistical assumption)
b) That the undecided vote is much smaller than portrayed.
c) That the undecided voters, can swing either way before November 21st.
d) That both NO & YES sides have solid base constituencies that are unlikely to change their minds before november 21st.
e) That a lot of money has been wasted in CKRC conducted civic education since very few citizens may even attend such classes (reportedly 10% have so far attended). The lessons may influence very little in the voting patterns.
f) That a lot of our national resources have been wasted in trying to woo support for the Wako draft, & that may all come to naught on November 21st.
peace.
unedited.
Job.
|
|
|
|
Post by Onyango Oloo on Oct 22, 2005 3:29:24 GMT 3
Even though I am an Orange damu, I find that these Steadman polls are quite dubious.
I will say more later.
oo
to
|
|
|
|
Post by njoroge on Oct 22, 2005 9:16:37 GMT 3
What are the districts that were sampled and what are the districts that were left out?
The Steadman poll notwithstanding, who were the folks who penned the Wako draft (Mongrel) and since we are not exactly in some sort of an EMERGENCY hurry to enact a new Constitution, why can't we sort out the so-called CONTENTIOUS issues slowly, hold grassroots civic education on the final draft lasting a few months to educate Kenyans and then have a proper vote that is largely de-politicised?
Njoroge
|
|
|
|
Post by aeichener on Oct 22, 2005 15:10:24 GMT 3
hold grassroots civic education on the final draft lasting a few months to educate Kenyans Being done already, and rather diligently, though with a very tightly packed schedule: www.kas.de/proj/home/events/31/1/index.htmlAlexander |
|
