People happy with their choice of party in 2009 more than twice as likely to vote as those no longer happy with their 2009 choice.

A survey of 2 000 metro adults conducted mid-February showed that, amongst those happy with their choice of political party in 2009 and who would vote for the same party on 18 May, 89% claimed that they would vote. This drops to 43% amongst those no longer happy with their previous vote and who would change the party they vote for. Whilst both these figures are likely to be overstated, they do suggest that people still happy with their political affiliation in 2009 are more than twice as likely to vote as those who are no longer happy. This was revealed today by TNS Research Surveys (Pty) Ltd, South Africa's leading marketing and social insights company.

This is a clear indication that dissatisfaction with one's previous party will be likely to be one of the causes of not voting on 18 May. Service delivery is at least a part of this dissatisfaction: 56% of those planning to vote are not happy with the service delivery they receive from their local municipality but this rises to 67% amongst those not planning to vote. It must also be borne in mind that some parties have also merged with others and that this may make some people feel that they have no political home. Indeed, it does appear, overall, that there are a large number of people lacking a political home at present.

Turning this around, 14% of people polled and who said that they planned to vote said that they would not be likely to vote for the same party in the 18 May local government elections as they voted for in the 2009 national elections. The company added that 73% said they would vote for the same party whilst 13% gave a don't know response.

This also suggests that, even amongst those who do plan to go to the polls, over an eighth in February were still to decide whilst a seventh represent the potential swing vote.

Amongst those who said that they did not plan to vote, 56% said that, if they did, they would not vote for the same party as before. Of these, only 22% would support the same party as before whilst 22% gave don't know response. It seems from this that, if those not planning to vote could be persuaded to vote, the swings might be much larger.

How do turn-out levels vary?

Whilst claimed turn-out levels are likely to be over-stated, it is possible to examine relativities. The study conducted by TNS Research Surveys suggests that turn-out in metro areas will be highest in Durban, Port Elizabeth, Bloemfontein and the East rand and lowest in Soweto, the Vaal Triangle/South Rand and East London.

Differences by age are lower than might be expected but claimed turn-out levels do drop amongst younger people 

Technical note

The study was conducted amongst 2 000 adults (1260 blacks, 385 whites, 240 coloureds and 115 Indians/Asians) in the seven major metropolitan areas: it has a margin of error of under 2.5% for the results found for the total sample. The studies use probability sampling techniques and are fully representative of the major metropolitan areas. The studies were conducted by TNS Research Surveys (Pty) Ltd as part of their ongoing research into current social and political issues and were funded by TNS Research Surveys.

Statement issued by Neil Higgs, TNS Research Surveys, May 10 2011

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