Innovations in Scientific Opinion Polling

This interactive timeline highlights innovations that make it possible to generate better predictions in domains where expertise matters – like atmospheric CO2.

1763-01-01 00:00:00

Bayes' Theorem

Modern forms of Bayes' Theorem have been used to bring clarity when information is scarce and outcomes uncertain.

1774-03-01 00:00:00

Movements of Celestial Bodies

Pierre-Simon Laplace applies Bayesian probability to complex scientific questions of his era: Would Jupiter smash into the sun and Saturn spin off into space?

1824-10-15 00:00:00

First Straw Poll

The first known example of an opinion poll is a local straw poll conducted by The Harrisburg Pennsylvanian.

1916-10-01 00:00:00

First National Survey

The Literary Digest embarks on a national survey (partly as a circulation-raising exercise) and correctly predicts Wilson's election.

1936-10-27 00:00:00

Sampling Bias Revealed

In 1936 the Literary Digest comes unstuck. The week before election day, it reports that Landon is far more popular than Roosevelt.

1936-10-28 00:00:00

Scientifically-Based Sampling

At the same time, George Gallup conducts a far smaller, but more scientifically-based survey, in which he polls a demographically representative sample.

1951-03-01 00:00:00

Cigarettes and Lung Cancer

Jerome Cornfield (NIH) uses Bayes' rule to advance the hypothesis that cigarette smoking is a probable cause of lung cancer.

1976-07-01 00:00:00

Tracking Soviet Nuclear Subs

John Nicholson, U.S. submarine fleet commander, uses Baysian computer analysis to figure out the most probable path of a Soviet submarine armed with cruise missiles.

1988-01-01 00:00:00

Iowa Electronic Markets

With an exemption from the U.S. Securities and Exchange Commission, the University of Iowa runs small-scale, real-money futures markets where contract payoffs depend on economic and political events such as elections. Operates as an educational and research project.

1989-01-01 00:00:00

"Sample of One" Reasoning

Past research (Ross et al., 1977) had shown that people tend to give higher than average estimates of the prevalence of their own opinion. Dawes (1989) argues that this pattern is consistent with rational statistical inference and is not necessarily a "false consensus effect."

Innovations in Scientific Opinion Polling

Copy this timeline Login to copy this timeline 3d Game mode