Focus groups: what are they for?

A cute puppy dog sleeping
Which political party leader is most like this puppy? (CC0 Public Domain image)

Questions about animals

If this blog post was an animal, what sort of animal would it be? That is the sort of apparently absurd question beloved of coverage of focus groups. But there is value in asking what sort of animal a politician would be, because the answers tap into a wide range of characteristics and emotional responses.

Asking people to pick an animal extracts expressive answers from those less confident with words. It teases out answers that tap into our instincts and subconscious rather than merely prompting people to give a rational, considered answer.

The small size of focus groups

What a focus group isn’t, however, is an opinion poll with a representative sample. What is lost from the smallness of the sample (the risk of atypical answers) is compensated for by the depth of what is heard.

A focus group involves inviting a small number of people – often in single figures, very rarely more than twenty – for a discussion lasting over an hour. The participants may all be from a target niche, such as swing voters, older men or students.

Often, they are given different stimuli to which to respond, such as being shown a short video clip of a politician, or the design of a poster. And sometimes there will be the seemingly off-the-wall question such as what type of animal a politician would be.

Focus groups shouldn’t involve percentages

The results of focus groups, when done properly at least, involve very little in the way of numbers. Being told that 25 per cent of a focus group of eight people agree with a particular view gives the finding an undue appearance of precision, given the smallness of the sample.

Giving percentages when speaking about a focus group treats it like a poll with a tiny sample – and so a humongous margin of error. (If 50 per cent of a focus group of eight gave the same response to a particular question, this finding comes with a standard margin of error of plus or minus 35 per cent – putting the range of likely truth between
15 per cent and 85 per cent.)

Focus groups are not polls

A focus group is not a small poll. It is very different. What makes for a great focus group is the incisive quote from a participant that captures the public mood, not the production of detailed statistics. Those quotes are often memorable, and can be highly impactful. ‘Get Brexit done’ – the election-winning slogan for the Conservatives in the 2019 British general election – came from comments in focus groups.

But for understanding the public’s views, always check the colour of what focus groups are saying against the numerical context of what public polls are saying.

The power of focus groups

That is why the best use of focus groups to understand politics is in tandem with opinion polls. Focus groups can help in exploring issues to identify the best questions to put into polls and how to word those questions – and then when you have the poll results, focus groups can help explain the reasons and motivations that lie behind the results.

To enjoy these benefits, you need sufficient focus groups to have confidence in their findings, as any single focus group may be off.

Unlike with polling and sample sizes, there is no similar rule of thumb for focus groups. But be suspicious of any findings based on just one. That may be driven by understandable time or budget restrictions, such as a media outlet doing one in the aftermath of an election debate, but it is still very brittle information.

What starts to become trustworthy is a series of focus groups. Credit to Times Radio, currently the main commissioner of public focus groups about British politics: they do a focus group every month, covering similar topics. So while each individual one should be treated carefully, you can look for – and rely on – patterns over time.

Even better is when you look at focus groups and opinion polls in tandem. This is because focus groups are no more an alternative to opinion polls than vowels are an alternative to consonants. Rather they are both best used in combination with each other. Polls can tell you the what, focus groups the why.

Find out more about focus groups and political polling in my book Polling Unpacked: the history, uses and abuses of political opinion polls.

One response to “Focus groups: what are they for?”

  1. great summary of the benefits and risks of focus groups. What is also interesting is paired interviews. An issue with focus group especially when talking about minor parties is a reticence to talk in front of others. In project traget voter from 1998 onwards we did not do any focus groups on swing voters relying on one on one or one on two depth interviews. We found out more about the obstacles to voting lib dems in this type of qualitative research than all the prior focus groups.

Leave a Reply

Your email address will not be published. Required fields are marked *

All comments and data you submit with them will be handled in line with the privacy and moderation policies.