Opinion poll

An opinion poll is a human research survey used to study public opinion. It uses a specific sample of people to represent a much larger population. A person who conducts these polls is known as a pollster. By asking a series of questions, pollsters use math to find generalities. They often report these findings using ratios or within specific confidence intervals. This process allows researchers to understand the views of millions by talking to a smaller group.

The mechanism of polling relies on careful sampling and data extrapolation. Researchers select a sample that they believe reflects the whole group. They then collect responses through various methods like telephone calls or in-person contact. Modern technology has introduced new tools like electronic clipboards and internet-based polling. Some organizations, such as YouGov, use large panels of volunteers. They then weight the results to ensure they match the demographics of the target population. Other methods involve analyzing social media content, such as posts on Twitter, to model public sentiment.

There are several distinct types of polls used for different purposes. A benchmark poll is typically the first survey taken during a political campaign. It helps a candidate understand their standing before they begin active campaigning. Another method is the tracking poll, which gathers responses over consecutive periods. These results are often calculated using a moving average of recent data. For example, a five-day moving average uses the most recent five days of data to show trends. Deliberative opinion polls are unique because they combine surveys with focus groups. In these polls, participants receive information and discuss issues before answering questions.

Exit polls are conducted as voters leave polling places during an election. These provide an accurate picture because the participants have definitely voted. They allow for comparative analysis across different geographic regions. This can help journalists understand how specific states voted rather than just looking at national numbers. However, exit polls can sometimes be misleading. During the 2016 U.S. primaries, CNN used exit polls to suggest a race was too close to call. In reality, the vote count showed Hillary Clinton had a large lead.

The history of polling shows a move from simple tallies to scientific research. The first known example occurred before the 1824 U.S. presidential election. It was a tally of voter preferences showing Andrew Jackson leading John Quincy Adams. In the early 20th century, The Literary Digest conducted massive national surveys. They correctly predicted winners in 1920, 1924, 1928, and 1932. However, they failed significantly in 1936. Their survey of 2.3 million voters predicted Alf Landon would win. Instead, Franklin Roosevelt won in a landslide. This error was caused by participation bias and a non-representative audience.

George Gallup helped transform polling into a scientific field after the 1936 error. He and others, like Archibald Crossley and Elmo Roper, used smaller but more scientific samples. This era saw the birth of major institutes. Jean Stoetzel founded the IFOP in Paris in 1938, the first European survey institute. In 1945, Gallup’s UK subsidiary was one of the few to correctly predict a Labour Party victory. Following World War II, the Allied occupation powers helped create survey institutes in Germany. These were used in 1947 and 1948 to help guide the process of denazification.

Polling has also played a major role in business and culture. During the Great Depression, businesses had to cut advertising spending significantly. In the late 1930s, corporate advertisers began a counterattack against their critics. They used scientific public opinion polls to understand market research. This helped them promote the concept of consumer sovereignty. By the 1940s, the advertising industry helped mobilize Americans during World War II. They helped define the "American Way of Life" through a commitment to free enterprise. This helped create the consumer culture seen in post-war society.

Understanding polling requires recognizing its complexities and potential for volatility. Tracking polls can experience dramatic fluctuations in their results. In the 2000 U.S. presidential election, Gallup showed Al Gore leading by eleven points one day. Just two days later, George W. Bush was shown leading by seven points. This volatility was partly due to an uneven distribution of political affiliations in the samples. Some pollsters try to fix this by adjusting the proportion of Democrats and Republicans in their samples. This practice remains a subject of debate. Despite these challenges, polling remains a vital tool for understanding the world.