NewsAbout UsContactWebsiteBizcommunity
Make smarter campaign decisions with an agile survey tool designed to help you optimise in-market advertising performance.
Read more
Your guide to boosting digital media effectiveness and developing rapid understanding of which creatives are performing best, in context.
Read more
Screen, test and validate your innovations with a suite of agile market research solutions that deliver rapid insights at the speed of your business.
Read more
How the right consumer feedback tool can accelerate your insights career - if you choose wisely.
Read more

Breaking news - People's perceptions of the week's events and their fallout

TNS Research Surveys, South Africa's leading marketing and social insights company, yesterday conducted a study amongst 250 people with access to landlines on the events of the last week and their fallout. This survey was conducted amongst adults from the seven major metropolitan areas of South Africa, interviewing them by telephone in their homes, with a margin of error of under 7.0%.

President Thabo Mbeki's legacy is highly significant
TNS Research Surveys asked people to agree or disagree with the following statement:

  • “Do you think Thabo Mbeki did a good job as President of the country?”

Seventy percent of those polled agreed with this statement, 26% disagreed and 3% said “don't know”.

Over the last few years, Mr Mbeki's approval rating rose from the low 30s in 2002 to a high of 66% in both 2004 and 2005, before beginning a slow decline in 2006. This decline accelerated towards the end of 2007 with the axing of Vusi Pikolo as head of the NPA and reached a low of 32% in June 2008. These figures come from samples of 2 000 metro adults, interviewed face-to-face in their own homes. A landline sub-sample in June 2008 gave an approval rating of just 15%. So, the figure of 70% represents a feeling that Mr Mbeki has left a significant legacy, despite his approval ratings having fallen in recent times.

The level of agreement with this statement does vary by race but in all cases, a majority of people feel he did a good job:

  • Blacks - 76%; whites - 57%; coloureds - 69%; Indians/Asians - 75%

There are no differences by age, gender or city.

Approval ratings of Mr Jacob Zuma and Mr Kgalema Motlanthe

People were asked to agree or disagree with the following statements:

  • “Do you think Jacob Zuma is doing a good job as President of the ANC?”

  • “Do you think Kgalema Motlanthe is doing a good job as Deputy President of the ANC?”

Thirty percent of the sample agreed with the first statements whilst 43% agreed with the second. There are highly significant differences by race:

  • Approve of Mr Zuma: Blacks - 46%; whites - 21%; coloureds - 11%; Indians/Asians - 19%

  • Approve of Mr Motlanthe : Blacks - 48%; whites - 39%; coloureds - 40%; Indians/Asians - 28%

Of importance is that 26% gave a “don't know” response on the Jacob Zuma statement and 40% on the Kgalema Motlanthe statement, many people clearly adopting a wait-and-see stance. Nonetheless, at present, Mr Motlanthe has a somewhat wider support base than Mr Zuma. The figure of 30% for Mr Zuma compares to an approval rating of 37% in June 2008 amongst a wider sample of 2 000 in-home interviews (21% on a landline sample), representing a just-statistically-significant increase.

Whilst there are no difference by age, there is a somewhat lower level of approval for Mr Zuma amongst black females as compared with black males. Approval levels for Mr Zuma in Durban are much higher than average at 56%, whilst people in Cape Town gave much lower levels of approval (13%). There are no differences by area for Mr Motlanthe's approval levels.

A simulation exercise suggests that, had an in-home study been undertaken, approval ratings of Mr Zuma would have been near 40% (much the same as in June) and 43% for Mr Kgalema.

But people are highly ambivalent towards the ANC

People were asked to agree or disagree with the statement:

  • “Do you think the ANC is doing a good job as ruling party of the country?”
    • Agree - 44% (blacks - 62%; whites - 32%; coloureds - 11%, Indians/Asians - 50%)
    • Disagree - 48% (blacks - 32%; whites - 65%; coloureds - 69%, Indians/Asians - 44%)
    • Don't know - 7% (blacks - 6%; whites - 3%; coloureds - 20%, Indians/Asians - 6%)

Black females show somewhat higher approval levels than do black males: taken with the results for Mr Zuma's approval rating, this suggests some dissatisfaction with Mr Zuma as against the party. Again, approval levels in Durban are much higher at 72% with Cape Town again being the lowest at 22%.

The level of approval of 44% rises to 49% in an exercise simulating the result amongst an in-home sample.

Most people feel that President Thabo Mbeki should not have been recalled by the ANC

People were asked to respond to the following question:

  • “Some people feel the ANC was correct to recall President Thabo Mbeki. Others do not. How about you?”
    • Correct - 31% (no differences by race but more males agreed; only 13% of white females agreed)
    • Not correct - 62%
    • Don't know - 7% (16% amongst white females)

It is surprising that there is so similar a response across all race groups: historically, one might have expected some significant differences. However, there are somewhat more people feeling it was a correct decision in Durban (46%).

Should a new political party be formed?

People were asked the following:

  • “There has been talk that those who support Thabo Mbeki should break away from the ANC and form a new political party. Do you think that is a good idea or not?”
    • Good idea - 34% (blacks - 22%; whites - 54%; coloureds - 43%, Indians/Asians - 19%)
    • Not a good idea - 60% (blacks - 75%; whites - 35%; coloureds - 51%, Indians/Asians - 75%)
    • Don't know - 6% (blacks - 3%; whites - 11%; coloureds - 6%, Indians/Asians - 6%)

Durban levels of agreement here are the lowest at 18%, with Pretoria and Cape Tow the highest at 46%

The simulation drops this figure to 28%. Whilst more people feel this is not a good idea, there appears to be a notable body of support for such an idea. That 22% of the black sample feel it is a good idea is notable.

The fallout

People were asked to agree or disagree with a set of statements on possible future actions and consequences. The levels of agreement are summarised in the table below, with people in Cape Town being more likely to say that the recall was disrespectful and that Mr Zuma should still stand trial for corruption (Durban's figures are lower for both these two statements) and less likely to feel Mr Mbeki should also stand trial:

It is important that Jacob Zuma still stand trial for corruption7558869481
The recall of Thabo Mbeki as President of South Africa in this manner was very disrespectful7670768984
Thabo Mbeki has served the ANC loyally for many years8680928691
Thabo Mbeki was wrong to dismiss Jacob Zuma as deputy president of the country in 20054458294047
Thabo Mbeki should be dismissed from the ANC142011119
Thabo Mbeki should also stand trial for corruption#4151333438
It is bad for South Africa that Trevor Manuel has resigned*7969908975
It is bad for South Africa that another 10 cabinet ministers had resigned by this afternoon*7475638081
Elections should be called earlier than originally planned4834546956

*These questions were framed before it was known that not all resignations would go through but asked once this had become clear. # 17% said “don't know”. All other “don't know responses averaged 6%.

These responses suggest that the ANC's action in recalling President Mbeki is seen as wrong both in absolute terms and in terms of the manner of how it was done. He is clearly seen to have served the ANC loyally for many years although 14% feel he should be dismissed from the ANC.

Despite all the turmoil of the past two weeks, it is interesting to note that the majority of people (75%) still feel that Jacob Zuma should stand trial for corruption including 58% of blacks. Interestingly, 41% also feel that Mr Mbeki should also stand trial for corruption, including 51% of blacks.

Just under a half of the sample feel that the dismissal of Jacob Zuma in 2005 was a mistake, although, at the time, a similar telephone study revealed that 74% felt Mr Mbeki was correct to release Mr Zuma.

Similarly, just under a half of the sample feel that early elections should be called, though this drops to a third amongst blacks.

The questions about the resignation of cabinet ministers were framed before it became known that not all the resignations would go though but it is clear that these resignations caused widespread concerns, especially the resignation of Mr Trevor Manuel.


On this telephone study, 60% of people said, at the end of the study, that they are positive about South Africa and its future. This compares to 49% in a June study of 2 000 conducted in home and represents a rise to the levels seen in the first quarter of this year. No differences by area are evident.

Our take-out

Mr Mbeki leaves an impressive legacy in the minds of most people despite his drop in approval levels in recent months. His recall is widely felt to be both wrong and poorly handled and he is felt to have been a loyal member of the ANC for many years. There is ambivalence towards the ANC and even a third of blacks polled feel they are not doing a good job. Whilst support for a break-away party from the ANC is in the minority, nonetheless, it is significant.

None of the events of the past few weeks has obviated the need in the minds of very many for Jacob Zuma still to stand trial for corruption, and a notable number even feel that Mr Mbeki should also stand trial.

However, optimism levels are still relatively good and somewhat improved on June 2008, no doubt partly fuelled by the drop in the fuel price and fact that interest rates have stopped rising for the moment.

Technical note

The in-home studies were conducted amongst a sample of 2 000 adults (1260 blacks, 385 whites, 240 coloureds and 115 Indians/Asians) in the seven major metropolitan areas: they have a margin of error of under 2.5% for the results found for the total sample. The telephone study was conducted between 17h30 on 23 September 2008 and 16h00 on 24 September 2008 amongst 106 blacks, 72 whites, 35 coloureds and 32 Indian/Asians. The simulations simulated the profile of the samples of 2 000. 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. For more details, please contact Neil Higgs on 011-778-7500 or 082-376-6312.

About TNS

TNS is a global market information and insight group.

Its strategic goal is to be recognised as the global leader in delivering value-added information and insights that help its clients make more effective business decisions.

TNS delivers innovative thinking and excellent service across a network of 80 countries. Working in partnership with clients, TNS provides high-quality information, analysis and insight that improves understanding of consumer behaviour.

TNS is the world's leading provider of customised services, combining sector knowledge with expertise in the areas of Product Development & Innovation, Brand & Communications, Stakeholder Management, Retail & Shopper and Customer Intelligence. TNS is a major supplier of consumer panel, media intelligence and audience measurement services.

TNS is the sixth sense of business™.

26 Sep 2008 14:35


can this survey even be taken seriously with such a low number of ppl polled? c'mon!
Posted on 26 Sep 2008 15:43
It is poor methodologies like this that gives research a bad name. TNS should have known better!!!
Posted on 29 Sep 2008 21:08
2000 not 250!!! Biz community owes TNS a BIG apology for not proofing headlines-
And the two comments slating the methodology obviously were by people who didn't bother to actually read the article. Read on and you will see that "The in-home studies were conducted amongst a sample of 2 000 adults (1260 blacks, 385 whites, 240 coloureds and 115 Indians/Asians) in the seven major metropolitan areas: they have a margin of error of under 2.5% for the results found for the total sample"
Posted on 30 Sep 2008 20:48
Lovely - read the article carefully, and see where the error is...-
no more comment required
Posted on 30 Sep 2008 20:52
Even sadder if you are too lazy to read, and see Biz com made the error-
Soo sad
Posted on 30 Sep 2008 20:53
Neil Higgs
Sample size is 250-
Thanks to all who commented. The sample size is indeed 250 but I do make some comparisons with a sample of 2 000 run earlier in the year. My apologies if this is unclear - it is not bizcommunity's fault. See also the technical note at the end of the article where I specify the sample breakdowns for both samples.
Posted on 2 Oct 2008 15:32
Yes, even sadder if you have your facts wrong...-
See Neil Higgs's comments. The sample was indeed 250. Maybe you should get your facts right before you criticise other people. Sad, yes very sad that people like you try and understand (and then even comment on) articles published here.
Posted on 6 Oct 2008 23:02