We call our approach to Public Opinion Research “Decision Driven Polling.”  Our goal is to help clients collect and use public opinion research to make critical decisions on how to allocate scarce resources to generate the greatest possible return – the most favorable change in public perceptions and the most beneficial behavioral outcome. Following is a brief description of the proprietary methods we have developed to help with public opinion and polling research problems.

FPi Message Planner™: Message Planning and Testing – Setting Policy Priorities

Effective communication with the public requires an motivating, clear message. The message must be consistent with the desires of the public and must be in a form that they can quickly understand.

FPi Public Opinion Message Planning and Testing is a procedure designed to identify the optimal public opinion message, the one message among all options that will most likely cut though the clutter and move the most public opinion.

We test all possible message options against each other – using a sophisticated “trade-off” methodology which most simulates the kinds of choices people make everyday. The trade-off method forces the public to make difficult decisions, and choose messages they find most motivating against all other competitor messages.

In this way, every message is tested against all other messages, for its ability to drive public opinion to consider and act in ways beneficial to our client organization or individual. This leads to the discovery of the optimal public opinion message, and a rank ordering of every potential message. Results are delivered for the total population and for key target subgroups.

We use the same “trade-off” approach for identifying public policy priorities. There are a multitude of public policy options. Finding the options that are most acceptable and desired by the public is our job.  Our methodology tests each policy option against all others,  leading to a rank ordering of policy options, starting with the optimal policy, the one most acceptable or desired by the public.

Examples of the Message Planning Trade Off Method

Who would you vote for if the election were held today?

Person A who supports a tax cut for middle income families, or

Person B who supports an increase in the minimum wage, or

Person C who supports cuts to balance the budget.

Which would most motivate you to support Organization A?

They endorse the clean water act, or

They endorse a traffic bond issue, or

They endorse a gun control initiative?

FPi Public Opinion Persuasion Monitor™

 Persuasion is a process. Our Public Opinion Persuasion Monitor is designed to measure each discrete step in the process. The process involves first Awareness or Name Identification. The public needs to be aware of the person or organization before they can think about or take action.

Once awareness is establish, the public begins to form impressions. The public learns more about the organization or person. Though learning a public image is developed. The public image includes facts and feelings, and overall positive or negative attitudes and impressions toward the organization or person.

After a public image is established (and hopefully a positive one at that), public opinion is then typically asked to consider a behavior that would benefit the organization or person. This is called “Consideration” or “Behavioral Intention,” which are very strong predictors of actual behavior. The end goal of a persuasion campaign is to achieve high levels of consideration and behavioral intention.

Our job is to measure each of these steps in the persuasion process and identify problems or “bottlenecks” that might exist. We then offer recommendations for fixing bottlenecks at each stage in the process.

For instance, if awareness is the problem, then the organization must get louder. If there is awareness, but not enough positive impressions, then they recommendation is to improve the message. If consideration is the problem, then the organization or person needs to do a better job in highlighting a motivating difference.

Segmentation and Targeting: Identifying the Optimal Solution

 Who are the best target groups for your message and for building a coalition?

There are many ways to segment and target public opinion subgroups. Voting propensity, income, gender, party affiliation, age, education, race, attitudes, lifestyles, psychographics, the possibilities often seem endless.

Indeed, just 5 possible groups and 5 ways to divide each produces 3,125 segments that can be targeted! What is needed is a systematic and consistently applied statistical methodology that can sift though the thousands of possible combinations and identify the optimal way to segment, and the best possible targets.

At FPi we have developed statistical algorithms that sift though every possible segmentation scheme and targeting option, and compare each to an advanced set of criteria. The result is a segmentation scheme and targeting recommendations that represent the best possible options for achieving the organizations or individual objective.

Once the best way to segment and the targets have been identified, all the information available is used to present a complete profile of their demographics, attitudes and behaviors, including how best to reach them.

Research to Help Plan the Media Buy

Identifying targets is just the first step; organizations and individuals need to communicate with them to move public opinion. As part of the segmentation analysis, we profile targets based on media habits and other communication vehicles.

We can also build a statistical model from this information, which will yield a Return on Investment analysis. This analysis compares the costs of communicating with the target, the number of impressions and/or exposures, and the potential gains that may be achieve. A complete understanding of the returns of every possible media mix option against each of the targets is then provided. This will help organizations and campaigns buy media and build a cost-effective media mix.

Database Modeling

 “Birds of a feather flock together.” This is also true of neighborhoods and citizen public opinion attitudes and behavior.  Citizens tend to hold the same attitudes, and exhibit the same public behaviors as other citizens living in the same neighborhoods. Thus identifying the voting behavior patterns of neighborhoods will tell us a lot about the citizens that live that neighborhood and how they might behave toward an organization or individual’s message.

For any state, county, district, zip code, census track or even a precinct, data is available which can we will statistically model to understand the attitudes and behaviors of citizens in the area. For instance, one of the best sources of database information for public opinion research is the voter registration records. Precinct voting records are “mined” for patterns that will help maximize media buys, improve direct mail response, and help plan GOTV campaigns.

Database modeling is one of the newest forms of public opinion research. It involves complete statistical modeling of large amounts of data (requiring lots of computing power). It is also a form of “inductive” analysis, in which meaning must be given to patterns of data. This requires a strong background in public opinion research and the experience to understand the potential meanings of data sets.

© 2005 Scott W. Flexo, Ph.D. and Flexo & Partners, Inc. All Rights Reserved

TM Pending