FAIL (the browser should render some flash content, not this).

High Tech and High Touch: The Use of Polls and Focus Groups in Political Campaigns

By  Brad Bannon


Survey research is more than numbers; it is about words and feelings. For this reason, survey research should be about focus groups and not just polls.


In political research, polling and focus groups should go together like a horse and carriage. But, often the only kind of research that campaigns conduct is a poll. Polls serve an important need in politics but they are rigid, structured and formal.


If a political campaign is an effort to build a candidate and win an election, the information from the poll would provide the skeleton and the focus groups would supply the skin. Conducting a poll without doing focus groups is a lot like having an ice cream sundae without the whipped cream topping.


But, what are focus groups and what do they do? Focus groups are in depth discussions with ten to twelve voters for a period of one and a half to two hours that deal with candidates, issues and verbiage. They are meetings with voters selected at random by phone within defined demographic parameters that offer in-depth information that mold the campaign into a being.  


Political insiders like to believe that they know everything about the issues and images that surround a campaign but the focus groups give voters an unfiltered chance to tell us what they think is important. In this period of political discontent, anytime you give voters the chance to sound off, the better you will be to understand a hostile political environment.


A professional moderator guides the discussion to acquire the information that the campaign requires.  The time you have to talk to voters in focus groups is an important part of the process. There is just so much information that you can get from voters in a 20 minute baseline survey.


The discussion in a focus group gives the researcher the luxury to probe in some detail the nuances of an issue that you can not begin to deal with in a 20 minute baseline telephone survey.


The focus group experience offers valuable vocabulary lessons for the campaign. Political insiders use specialized language or jargon that is either incomprehensible or misleading to voters. Focus groups give you the chance to learn the language that voters use to describe the issues that they worry about.


I once conducted focus groups in suburban Virginia for a coalition of environmental groups. The purpose of the groups was to discuss the problem of suburban sprawl. The problem was that the word that my clients liked to use to describe the problem, “sprawl”, had a positive meaning to voters. When I asked focus group participants to tell me what they thought of when they heard the word, “sprawl”, they told me that it meant having room to be comfortable. Overdevelopment was a much better word for the environmental groups to use in their communications because that word had a clear negative connotation.


If you do decide to conduct focus groups, and you should if you have the budget, it is important to keep them loosely structured. Many researchers make the mistake of conducting very formal and structured focus groups.  Focus groups are an opportunity to collect impressions not more numbers.


While polls are very structured and are used to complement the data you get from a baseline survey,  focus groups should be informal so that voters have the chance to raise their own issues and concerns. The best way to organize the discussion in a focus group is to get participants comfortable with the moderator and each other. Have everybody introduce themselves and tell a little bit about their kids or jobs. I like to start groups by talking about my kids so that the participants can identify with me.


Then start the discussion by asking participants whether or not they think things in the country, state or county are going in the right or wrong direction and ask them why they think that way. The questioning can become more direct as the group continues.


One of the decisions that the campaign has to make is whether to conduct the groups before or after the baseline survey. There are arguments on both sides but my opinion is that focus groups are most valuable before the campaign does the baseline survey.


The best reason to do focus groups first is that the information from the groups may provide valuable insight into the construction of the baseline questionnaire. The people in the campaign will have strong ideas about the questionnaire based on their knowledge of the area. The researcher will also have firm ideas about the content of the questionnaire on the basis of his or her polls in other areas.


But if you do the focus groups before you conduct the baseline survey, the voters in the groups will raise issues that neither the researcher nor client would have come up with on their own.


There are, of course, limitations to focus groups. A poll is a systematic and scientific measurement of public opinion based on the random selection of voters to interview If you are careful and you select a truly random sample of 600 voters in Virginia or any other state, you can be confident that you are accurately measuring public opinion  within a margin of plus or minus 4%. But there is little chance that talking to a collection of 12 voters in a focus group is representative of anything.


To deal with the reliability problem, you have to be very careful how you conduct the groups and interpret the information you get from a focus group. You should always conduct focus groups in pairs among specific types of voters. If you believe that you have problems or opportunities with women over the age of fifty, older women would make an attractive focus group opportunity.


The most important thing however is to use the qualitative information from the focus group in conjunction with the quantitative information you get from the baseline survey. You may learn from the poll what issue is most important to voters, but the focus groups will tell you how to talk about that issue.


Bannon Communications Research © 2011 All Rights Reserved| Home | Contact Info