Polling+-+Methods+&+Types

__Polling Methods & Types__ Written By: Tom Landro, Nathalya Diosa, and Kate Perdion. Polling today occurs in several types and by several methods. Since there are no laws or regulations on how research institutions conduct polls, many methods are used; all of them have advantages and disadvantages. No polling method is perfect, and some are better than others in certain scenarios.

Straw Poll The **Straw Poll** method was a methos used by the magazine //Literary Digest// in the early days of public opinion polling. The polls were called "amazingly accurate" in predicting the outcome of the popular vote for the election of 1916. The disadvantage of the straw poll was it being based on unscientific survey methods, therefore making it obsolete after the election of 1936 where the poll predicted a Republican win. A major error of the poll that contributed to its failure was a poor **sample** **selection**. The sample, part of the population chosen for questioning for gauging opinion, was based off of telephone and automobile owners. That means middle and upper class people, who are more likely to vote for a Republican. George Gallup correctly predicted the 1936 election outcome with the help of using those who read newspapers. The Gallup Organization still predicts outcomes of elections to this day.

TYPES OF POLLS
 * Telephone polls** operate as their name suggests, by telephone. In the early days of telephones, responders were excited to speak their views to pollsters. However, in recent years it has become increasingly difficult to conduct telephone polls due to the decreasing number of landlines and the decreased desire by the public to be contacted by outsiders.
 * Exit polls** are conducted at designated voting places. A pollster asks voters a set of questions as they leave the voting place. The media conducts exit polls, asking every tenth voter who emerges, and uses the results to predict the outcome of the election.
 * Tracking polls** were introduced to help presidential candidates keep watch on short-term campaign developments and how their strategies effected their campaign. Since 1992, when they were first introduced, they have been used by the media as well. The polls are taken every day and usually to only registered voters. The polls can be grouped into averages of a few to five days.
 * Internet polls** are similar to telephone polls only using the internet. They both have the same problem of relying on the willingness of people answering them. While most internet polls are scientific, there are plenty that are not. Some can resemble the obsolete straw poll, asking only a certain audience who will probably answer one way, such as on a biased-news website. The polls can include the interests of only a limited number of people.
 * Push polls** are conducted in order to persuade the participant. They include biased questions that are designed to either support or bash a candidate in order to influence voters. They do not necessarily provide much information and are conducted amongst thousands of people, while other polling methods use smaller numbers to reduce the **margin of error**, or sampling error in a survey's results. If the margin of error is 3%, and a result is 50% of the questioned chose one answer, then it is believed that 47-53% of the public is the same.

CONDUCTING POLLS There is a serious amount of consideration that polling organizations must take before conducting a poll or survey. Every aspect of a poll is critical to the accuracy of the results, and a small error can drastically change the results. Question phrasing is an important aspect that makes an accurate or inaccurate poll. A survey question should be made for anybody to understand what it is asking. Poorly phrased questions can result in a ridiculous outcome, such as a high percent of the United States believing that the Holocaust never happened, or result in a high amount of "not sure" or "unknown" responses. Sample selection is equally as important as question phrasing. **Random sampling** gives anybody in the public the same chance to get selected. Random sampling in theory is the best method for getting a completely unbiased sample. A variation of random sampling, called **stratified sampling**, splits the population into subgroups, then weights them based on demographics of the entire nation. Stratified sampling does not allow volunteers, as volunteers usually have different beliefs than those who do not volunteer to be questioned.

__Vocabulary:__
 * Straw Poll:** An unofficial ballot conducted as a test of opinion.
 * sample selection:** A type of bias caused by choosing non-random data for statistical analysis.
 * Telephone polls:** polls taken by telephone, and even by text messaging using a cell phone.
 * Exit polls:** A poll of people leaving a polling place, asking how they voted.
 * Tracking polls:** An opinion poll in which the same sample, such as a small number of voters, is questioned periodically to measure shifts in opinion.
 * Internet polls: P**olls taken on the internet in which some allow many participants.
 * Push polls:** An opinion poll that seems true but is not necessarily true. The true objective is to sway voters using loaded or manipulative questions.
 * margin of error:** An allowance for slight error or miscalculation; an allowance for changing circumstances.
 * random sampling:** The selection of a random sample, each element of the population has an equal chance of being selected.
 * stratified sampling:** The population is divided into to sub populations (strata) and random samples are taken of each stratum.

__Section Review Questions:__

1. Which type of polling is considered to contain biased questions that are designed to help support or bash another candidate for voting influence? A. Straw Polls B. Internet Polls C. Push Polls D. Tracking Polls

2. True or false? There is only one method to polling.

3. What is an important aspect that makes an accurate or inaccurate poll?

4. Which types of polls are similar? i. exit polls ii. telephone polls iii. push polls iv. internet polls A. i, ii, iii, iv B. ii, iv C. i, iii D. ii, iii, iv

5. What is random sampling and why is it, in theory, the best method to use?

6. True or False? There are no regulations on how research institutions conduct polls.

7. Why were tracking polls introduced?

__Current News:__

Obama Upper Hand to Romney- http://www.cnn.com/2012/09/23/politics/battleground-polls/index.html

Distracted By Polling Noise- http://fivethirtyeight.blogs.nytimes.com/2012/10/15/oct-15-distracted-by-polling-noise/