PARTY

Markinor predicts two-thirds majority for ANC

Polling company says that if undecided voters are allocated ruling party should get 67%

It was clear from the results of the latest Ipsos Markinor opinion poll [1] that around one in every ten (10%) of likely voters [2] are still undecided. On 16 April 2009 we reported in a press release that the ANC would potentially win between 64% and 66% of the vote on the national ballot and could, depending on the voter turnout, achieve the coveted 66.6% (two thirds) majority.

Traditionally, the ANC has benefited most from late deciders and high voter turnout. Voter registration was brisk and all other indications are that there is a lot of interest in this election, therefore we can expect a higher voter turnout than the 15.8 million and 15.9 million of the previous two national elections. Ipsos Markinor uses a ballot paper resembling the actual election ballot paper and asks respondents to complete this ballot paper by drawing a cross next to the party they would support in the election.

At the time of fieldwork about 10% (one in every ten likely voters) was undecided and did not indicate their party choice on the ballot paper. However, in this special pre-election Socio-Political Trends survey a series of other questions were also asked and the results of these questions were used to determine the preference of these "undecided" voters. The 3 key aspects that were factored into this analysis were:

Likely voters' feeling about each of the eight biggest parties in the country was established by employing a holistic scale. Each voter had to indicate how he/she feels about each party on a 6-point scale ranging from "I reject this organisation completely and on principle" to "I would definitely vote for this organisation if there were an election".

Leadership and the image of leaders are important elements of the political scene in our country and therefore a question linking leaders and parties was also included: and read as follows:

"I am now going to read out some voting options that may be possible in the next election. Please tell me which one of them comes closest to how you feel right now:

1. I will vote for COPE under the leadership of Mosiuoa Lekota or Mbhazima Shilowa
2. I will vote for the ANC under the leadership of Jacob Zuma
3. I will vote for the DA under the leadership of Helen Zille
4. I will vote for another opposition party
5. I don't think I will vote"

The third measure determines the relative closeness to a political party. In the first place we determine whether they feel close to a party and then secondly which that party may be.

Using the results of these questions advanced modelling was done to arrive at the preferences of "undecided voters". - in other words how these voters may potentially vote and how this would influence the final outcome of the election. These results and those of the ballot paper were combined to arrive at a possible prediction of the election results. All sample surveys are subject to a certain margin of error, as opinion polling is not an exact science. Results have to be evaluated within this margin of error. For the study as a whole this is between 0.72% and 1.66%.

Party

Predicted average %

Possible range of results

ANC

67

65.34% to 68.66%

DA

13

11.34% to 14.66%

COPE

11

9.34% to 12.66%

IFP

3.5

Up to 5.16%

ID

1.5

 

UDM

1

 

ACDP

1

 

FF+

1

 

Other parties

1

 

Methodology

These results are from a special pre-election Socio-Political Trends survey conducted by Ipsos Markinor. A total of 3531 respondents, representative of adults of voting age (18+) in South Africa, were interviewed between 24 February and 10 March 2009. Respondents were randomly selected to represent South Africa's adult population and all population groups, provinces and types of settlements were included. Interviews were done face-to-face in the homes and home languages of respondents. All sample surveys are subject to statistical error and the results for this poll have to be evaluated against this background. Depending on the response rate, the margin of error for the poll as a whole is between 0,72% and 1.66%. Data for this report is analysed amongst likely voters.

Footnotes:

[1] Fieldwork was done by means of personal in-home interviews by trained interviewers from 24 February to 10 March 2009.

[2] "Likely voters' are registered, have a green bar-coded ID book, they indicate also that they want to vote and are likely to vote.

Statement issued by Ipsos-Markinor April 21 2009

Click here to sign up to receive our free daily headline email newsletter