Public opinion polling: answering common criticisms - Political Adviser

As campaign teams gear up for next year's electoral battles, one component of any strategy should be public opinion polling. But some candidates and groups resist this extremely helpful tool.

Leadership

The strongest resistance to public opinion polling is the notion that elected leaders should lead the American people, not follow them. When campaign consultants propose polling, political candidates are most likely to say that they know what the people want, and they do not need a poll to tell them what policies they should support or oppose.

Leading, rather than following, public opinion is a very strong argument, one often cited by Arianna Huffington, political pundit and California gubernatorial recall candidate. And it's one that needs to be handled very carefully.

First, you should tell the candidate or group that they should never change their beliefs based upon a poll or set of polls. People do want their leaders to lead and are forgiving if they disagree with them on a particular issue, provided that the leader has a strongly held view.

Voters want to believe that their leaders also listen, as well as act on conviction. Ronald Reagan was staunchly pro-life when he was elected president in 1980, a time when public opinion tilted more toward the pro-choice position. He did not attack pro-choice voters, but rather framed his beliefs as respectful disagreement.

In policy battles where groups fight for one version of legislation or another, finding where people stand on the issue is extremely important. In some cases, you might even find that the public is not engaged in the issue, which might provide cover for a legislator or agency official who wants to make a change. Polling can unearth information about the political landscape in order to enable them to lead more effectively.

Cost

After the leadership argument, the second biggest obstacle to public opinion polling is usually cost. Good research is not cheap, and cash-strapped challengers and non profit organizations on tight budgets often have difficulty justifying the expense of public opinion polling. Why spend tens of thousands of dollars on polling when you can hire someone knock on doors or help work the issue on Capitol Hill or in the state capitol?

The best response to this concern is that winning campaigns, whether they are for public office, legislative efforts or regulatory changes, all cost significant amounts of money. Most people involved in these kinds of campaigns know this, and the cost issue really is masking an anxiety about having enough funds for everything that they believe is important to help them win.

The key is to show that the costs of doing research, which helps define, test and reline strategy, is more likely to help than hiring an extra person to work the issue. If you have the wrong strategy, if you don't react enough to changing conditions during the campaign, it will not matter how many people are on the team. You are going to lose.

Accuracy

Once a candidate or group has overcome the hurdles of leadership and cost, the most common approaches to blocking public opinion polling are attacks on the accuracy, inconsistency and value of the research itself.

Here we move from the theoretical to the technical aspects of public opinion polling. As most pols have been consumers of re search at some point and believe that sometimes the numbers lie, why make the same mistake in your own campaign? Accuracy of polling is a hot topic in the industry today, as many academics, political professionals, media and the public are taking a more skeptical view of the numbers.

Almost everyone in politics has seen polling that just does not sound right. Most people can cite a particular bit of polling that said that a candidate was going to win by a wide margin only to see that person lose on Election Day. On policy matters, an official who opposes your side should be expected to aggressively question the accuracy of the polling regardless of how well the research was conducted.

The best way to handle accuracy concerns is to propose a tracking program with adequate sample sizes for more precise results. But even then, sometimes a good questionnaire with fair and balanced questions yields inaccurate results. It does not happen often, yet we must concede that there is a margin of error in every poll and that our findings are good in about 95 out of every 100 cases.

In the event that our findings are wrong on any particular poll, the best insurance is to track public opinion over time. Any mad spike in the trend can be discounted unless it has to do with a specific real world event. Moreover, you can minimize the margin of error by surveying more people, if the timing is critical and decisions need to be made immediately, you should survey larger samples of voters to get more accurate results.

Inconsistency

If polls can be defended as accurate, critics will likely attack the industry as a whole, citing various circumstances in which different research organizations come up with varying results. Inconsistency is a tough hurdle to overcome since it is so true. Who can forget about the polling at the end of the 2000 presidential campaign, which varied from company to company on how many points George W. Bush was ahead or behind in the closing days?

There are several ways to handle the inconsistency issue. First, not all polling organizations have the same standards. Yes, some folks do bad work and still get paid while others who do very good work get pulled down. It happens in every industry.

Your best insurance against inconsistency issues is to make sure you hire the right people to do the research. Ask for references, question standards and policies, and make a gut call on comfort level. Polling will be inconsistent during any campaign, and the best you can do is to feel good about your team.

Value

The last refuge of critics of public opinion polls is that the value of the information is not helpful. This usually is a result of a bad experience with polling or a lack of experience with it. Most of my first-time clients have resisted commissioning a firm because the last time money was spent on public opinion polling, the reports and data books ended up on a shelf to gather dust. It is unfortunate that some candidates and groups--and good corporations pay for research only to find out at the end that they are overwhelmed with useless data.

The key to ensuring that the research will be valuable is to constantly ask one key question as you help write the questionnaire: What will we do if we get this answer? For example, in an election campaign you ask a question about how likeable the candidates are. If you happen to have a candidate who is extremely likeable, but find out that the public believes the opposite, maybe what you do is get that person out in public more often.

If you find out that a particular issue is hurting the candidate, you may choose to run a particular response ad. On the policy level, you may find that the opposition is succeeding in distorting your side among a specific demographic, leading to a targeted response. As a customer, it's your job to keep us focused on what is important.

Conclusion

Critics of polling attempt to shame decision makers into dropping research. When that fails, they attack the science. Leaders and groups can remain true to their beliefs and become more effective in persuading the public if they also listen to them.

Polls that seem inaccurate or inconsistent may be easily explained with critical analysis and a commitment to tracking. Moreover, the cost of really knowing how your strategy is performing definitely outweighs the true cost of guessing. In politics, nothing is more valuable than victory. Without the proper tools, including public opinion research, you are taking unnecessary chances in achieving that success.

Michael Cohen, Ph.D., is vice president with Fabrizio, McLaughlin & Associates, a Republican research and consulting firm based in Alexandria, Va.

COPYRIGHT 2003 Campaigns & Elections, Inc.
COPYRIGHT 2003 Gale Group