Using opinion polls wisely can win your charity valuable media coverage, and it can be done at a fraction of the cost of an advertising campaign. Dominic Wood offers some guidelines.
Using opinion polls to promote a cause has one key advantage for cash-strapped charities: it costs a fraction of the price of a major advertising campaign. Carried out correctly, polls can have a huge impact - even small charities can attain good media coverage with effective poll data. There are other advantages. Volunteering charity CSV uses polls partly to overcome its minimal advertising budget, but also because the charity's marketing team believes the way people engage with news editorial has three times more impact than advertising.
Opinion polls may be a cheap way of getting press coverage, but there are pitfalls. "Historically, the success of a campaign is measured by means of the thud factor of its press cuttings landing on your desk," says David Barker, head of communications at the British Heart Foundation.
"But that doesn't necessarily make a campaign successful, because you might not have hit the right type of media for you."
An equally knotty problem is being too eager to have poll results go your way. Encouraging too many supporters to take part in a poll can skew the results and the practice, which is known as 'piling in', is frowned upon by experts and academics.
This step-by-step guide shows you how to avoid the pitfalls and run a successful campaign.
1 Choose a suitable topic The topic of a poll is key to getting successful media coverage. "The secret is to provide a poll that shows the unexpected, reveals that public behaviour or attitudes are changing or gives a novel slant on an issue," says Clive Nancarrow, professor of marketing at the University of the West of England's Bristol Business School.
CSV took this route when it commissioned an ICM poll that found 17 per cent of 18-24 year-olds surveyed said, without any prompting, that volunteering had improved their sex lives.
"The poll was a bit of fun, but the important thing is for people to read about volunteering in relation to issues they often talk about," says Jason Tanner, head of press at CSV. "It's not enough any more to talk about volunteering in an 'oh, isn't it nice' way, so we created more impact with this poll."
Polls that go over familiar territory can have less impact. One source called the recent National Council for Voluntary Organisations' poll about charitable giving to the tsunami appeal a "meaningless set of questions", because the public already knew lots of people had donated when it was published.
2 Decide whether to pay a specialist pollster Once you've chosen the right topic, you'll have to decide whether to commission a professional polling company. The main advantage of a specialist pollster is that it can offer access to a random selection of people, which lends impartiality.
Using a well-known polling company can also give weight to your poll.
The source of your poll should be "professional, credible and independent", says Nancarrow. "Pollsters who work to professional bodies' codes of conduct can add to the perceived trustworthy nature of the research."
But how much is outsourcing likely to cost? Polling companies typically charge by the question. ICM charges about £500 per telephone poll question to 1,000 adults, with clients able to add as many questions as necessary.
But there are ways to reduce the costs. Shaun O'Leary, director of operations at the Prostate Cancer Charity, says: "We find the cheapest method is to get our question tacked onto a telephone omnibus, which asks people to answer a range of consumer questions in their homes."
Nancarrow recommends a thorough trawl around the websites of leading research agencies to examine the cost of omnibus questions.
Barker warns that agencies will usually charge more for open-ended questions simply because it will take longer to write up the answers.
Small charities that can't afford to pay for their polls can do it effectively themselves, says Joe Saxton, head of the consultancy NfpSynergy, who believes market research isn't as complicated as many charities think. "If they seek advice from a few polling companies and crib from other charities, DIY isn't a bad option," he says.
3 Choose the number of questions and sample size Polling can be a balancing act: it takes a larger number of questions to explore a complex issue, but you risk turning people off if you ask more than 10 during face-to-face polls. Many experts say the number of questions should be kept to a minimum, not least because extras will cost. Saxton says paper polls probably work best for larger surveys because people are less able to hold multiple questions in their minds during a telephone survey.
Ruth Scott, campaigns and parliamentary manager at Scope, says charities should poll at least 1,000 people if they are trying to make statements on behalf of the nation. Yet polling a smaller number of people is acceptable when the issue doesn't affect large numbers. Scope surveyed 300 disabled people about their access to communication equipment, which Scott says is a reasonable size in relation to the number of people who use the equipment.
4 Use polls to track change "Repeatedly asking the same question at different times can help to build a neutral picture of change," says Saxton.
The Prostate Cancer Charity subscribes to this view. For three of the past four years, its annual poll has asked: "What is the main function of the prostate gland?" O'Leary says: "The level of correct responses has improved by about 10 per cent in that time, but it allows us to avoid the delusion that we will make a profound difference overnight with one campaign on our limited resources."
But O'Leary also guards against being tied to the same agenda and says no question is too simple. "When working on a single issue, it's easy to forget how little people know about biology, so this year we asked: 'Who do you think has a prostate gland?'"
5 Phrase questions carefully It is important to spend time framing questions, because you can't change them afterwards. "A subtle difference in wording can make a big difference to what is being asked," says Scott.
A good maxim to follow is 'the more complex the issue, the simpler the question'. Scott believes people will express negative attitudes if their ignorance makes them perceive their answer to be acceptable. "Our simple question 'Could a blind person do your job?' is not leading, but people respond to what they think is a positive question," she says. "Our interpretation shows they are being negative if they say no and are probably unaware that assisted technology could help disabled people to do their jobs.
The answers may not be interesting but the impression gained is very illuminating."
Saxton warns that voluntary groups should be wary of asking loaded questions.
Greenpeace recently used volunteers to ask 5,000 people in London streets "Should gas guzzlers like 4x4s have to pay a higher congestion charge than other cars?" Saxton doesn't know why the question was worded like this, but comments: "I presume Ken Livingstone treated the result of that survey with the pinch of salt it deserves."
Greenpeace spokesman Ben Stewart admits that the mayor "wouldn't give it as much credence as he would to a poll by Mori", the firm it usually employs to conduct its surveys. "He can look at the result that showed 80 per cent were in favour of a higher charge, know how it was arrived at and make up his own mind," he says. "We weren't disingenuous about the wording to the public or the media. It characterises what Greenpeace thinks about gas guzzlers, and we presented the question and its result as a Greenpeace campaigning poll, not as a scientific poll."
Nancarrow warns that questions must be sound to avoid accusations that they are biased. In particular, he says, earlier questions must not direct respondents' thinking in later ones.
6 Take care when packaging the results for the media Writing press releases that don't accurately reflect the poll figures can make the whole exercise worthless. Chris Menzies, research executive at ICM, says this is a common problem and recommends that charities that are unsure about methodology should get their press releases checked for accuracy by the polling company before sending them to the media. "Typical examples include reading the wrong column on data tables and using different language in the press release to that in the questionnaire," he says.
Sometimes the best way to get a message out is to package the results of different questions in separate press releases. CSV's benefits of volunteering poll found that almost half of respondents had improved their health and fitness. To get this news into the media, the charity staggered its press releases over some weeks to separate the main findings about sex lives and fitness. There was a risk that journalists could become weary of reading claims that volunteering was a panacea, but the strategy paid off: the poll produced 130 pieces of media coverage, including the attention of nine national newspapers.
It is also important to consider all the avenues to media coverage. "One national newspaper journalist said that most charities don't think anything exists in the national press apart from news," says Saxton. "But it's often better to feed details into features coverage."
Nancarrow suggests gauging media interest before commissioning polls or approaching journalists likely to be sympathetic. He advises charities to promote survey results to the press during the so-called silly seasons, such as Christmas and high summer, when slow news days are common.
However, the secret, according to O'Leary, is to team poll results up with other news.
CASE STUDY - SCOPE
When Scope wanted to assess public attitudes to disability, it commissioned YouGov to encourage 2,150 adults to explore their attitudes through an online poll.
The findings showed that 36 per cent could not name a single famous disabled person, despite the success of well-known figures such as Professor Stephen Hawking, Olympian Tanni Grey-Thompson and David Blunkett. Two-thirds also said it was acceptable for a non-disabled person to play a disabled role on film despite the general offence it can cause.
Scope used the poll to get the less than media-friendly issue of disability into the public domain. Its successes included 17 regional newspapers, seven news websites and one TV show.
The findings were also used to spark studio debates with 37 regional and two national radio stations involving disabled actors, Scope spokespeople and the public. Ruth Scott, Scope's campaigns and parliamentary manager, says the poll was deliberately kept simple to grab media attention: "The statistics were self-explanatory and didn't require too much explanation or translation, so journalists tended to use our information without much manipulation."
OTHER USES FOR POLLS
As well as securing news coverage, opinion polls are a useful tool for measuring the success of public education and awareness-raising campaigns.
They can help you understand your audience and strengthen objectives at the outset of a campaign. British Heart Foundation head of communications David Barker recommends that charities spend about 5 per cent of their communications campaign budget on research polls and planning. "Charities rely on gut feeling to measure the impact of their campaigns, but that's not good enough when trying to present a solid business case," he says.
In last year's anti-smoking campaign, Give Up Before You Clog Up, the BHF measured people's understanding that heart disease can be caused by smoking cigarettes both before the campaign began and after it, to assess any shift in attitude. "The success of our campaign is measured by an 18 per cent rise in the number of people being aware of the link," says Barker.