Question

Election 2008 During the spring of 2006 the news media were already conducting opinion polls that tracked the fortunes of the major candidates hoping to become the president of the United States. One such poll conducted by Financial Dynamics showed the following results: "Thinking ahead to the next presidential election, if the 2008 election were held today and the candidates were Democrat [see below] and Republican [see below], for whom would you vote?" The results were based on a sample taken May \(16-18\), \(2006,\) of 900 registered voters nationwide. a. If the pollsters were planning to use these results to predict the outcome of the 2008 presidential election, describe the population of interest to them. b. Describe the actual population from which the sample was drawn. c. Some pollsters prefer to select a sample of "likely" voters. What is the difference between "registered voters" and "likely voters"? Why is this important? d. Is the sample selected by the pollsters representative of the population described in part a? Explain.

Short answer

Answer: The main difference between registered voters and likely voters is that registered voters have completed the necessary procedures to be allowed to vote in elections, while likely voters are a sub-set of registered voters who have a higher chance of actually participating in the election based on factors like past voting behavior, interest in the election, and their expressed intention to vote.

Step-by-step solution

a. Identifying the population of interest

The population of interest for the pollsters, who are trying to predict the outcome of the 2008 presidential election, would be all eligible voters who would participate in the 2008 election throughout the United States. This population includes citizens aged 18 and over who are eligible to vote and who would actually cast their vote in the election.

b. Describing the actual sampled population

The actual population from which the sample was drawn is described in the exercise as 900 registered voters nationwide. So, the sampled population is made up of people who are registered to vote, regardless of whether they will surely vote or not in the future election.

c. The difference between registered voters and likely voters

Registered voters are those who have completed the necessary procedures to be allowed to vote in the elections. They may or may not choose to vote on Election Day. On the other hand, likely voters are a sub-set of registered voters who have a higher chance of actually participating in the election. Pollsters usually identify likely voters based on past voting behavior, interest in the election, and their expressed intention to vote. The distinction between registered voters and likely voters is important because focusing on likely voters can lead to more accurate predictions of election outcomes since they represent individuals who are more likely to vote and influence the election results.

d. Representativeness of the sample

The sample selected by the pollsters includes 900 registered voters nationwide. It is not clearly stated whether the sample was taken using random sampling or any other sampling method. To conclude if the sample is representative of the population described in part a (eligible voters who would participate in the 2008 election), we would need more information about how the sample was selected. Ideally, a random and well-designed sampling method should be used to ensure that the sample accurately represents the population of interest. Without knowing the specifics of the sampling method, we cannot definitively say if the sample is representative of the population of interest.

Key concepts

Sampling Methods

When conducting opinion polls, selecting a proper sampling method is crucial to ensure accurate and reliable results. There are several ways to choose a sample from a population, such as simple random sampling, stratified sampling, or cluster sampling. Each method has its pros and cons depending on the nature of the poll.

• **Simple Random Sampling**: Each individual in the population has an equal chance of being selected. This method is straightforward but may not capture specific subgroups' variations.
• **Stratified Sampling**: The population is divided into subgroups (strata) based on certain characteristics, and samples are drawn from each stratum. This method ensures representation of different subgroups, but it requires detailed population information.
• **Cluster Sampling**: The population is divided into clusters, and entire clusters are randomly selected. This method is cost-effective for large populations but may lead to higher sampling error.

Understanding the sampling method is essential because it impacts the poll’s reliability. For the poll conducted in 2006, without information on the sampling method, the representativeness of the results is uncertain.

Registered Voters vs Likely Voters

In the context of political polling, distinguishing between registered voters and likely voters is significant for making accurate predictions. Registered voters are individuals who have officially signed up to vote, but this doesn't guarantee they will vote.

• **Registered Voters**: They have completed the voter registration process but may not be engaged or motivated to vote in the upcoming election.
• **Likely Voters**: These individuals are not only registered but also show a higher probability of voting based on factors like past voting behavior and expressed interest in the election.

Focusing on likely voters can lead to more precise forecasts of election outcomes because they are more indicative of who will actually vote. This distinction is critical because, in major elections, not all registered voters choose to cast their vote, making the likely voter sample more pertinent for predictions.

Population of Interest

The population of interest for opinion polls in an election context refers to all individuals whose opinions are relevant to the prediction or understanding of the election outcome. For the pollsters in the 2008 election scenario, this includes all eligible voters who would actually participate in the election.

When defining the population of interest, pollsters must consider:

• **Eligibility**: All adults over the age of 18 who have the legal right to vote.
• **Participation Inclination**: Those who are not only eligible but are also inclined to cast their votes.

This definition is broader than just registered voters, as it also includes those who might decide to vote during the election period even if they had not participated in past elections. Properly defining and reaching the population of interest is vital for ensuring the accuracy and relevance of poll results.