POLI 001 Lecture Notes - Lecture 9: Twin Study, Alf Landon, Hate Speech
Russia and the 2016 election
- Indictments against trump campaign aids
- Special counsel Robert Mueller accuses Trump campaign aids of tax evasion,
money laundering, and not disclosing activities(lobbying) undertaken on the
behalf of a foreign government. Unrelated activity to the 2016 election.
- George Papadopoulos, a campaign foreign policy advisor to Trump, pleaded
guilty to lying to investigators about meetings with Russian agents
- There is no formal investigation at this time into the Clinton campaign's ties to
Russians.
Public Opinion
-What is public opinion?
-The collective attitudes of citizens on a given question.
-How do we measure public opinion?
-We measure public opinion through surveys by sampling a population of
interest.
-A sample does NOT exactly match the population; it reflects the
population with some predictable degree of accuracy.
-What is Critical to the Accuracy of a Sample?
-Is the sample randomly selected?
-Each individual should have an equal probability of being selected.
-How big is the sample size?
-The larger the sample, the more accurately it represents the
population(depending on random selection!)
-Random Selection is the key.
-Selection is random if every unit has an equal chance of being chosen
-Random selection eliminates all known and unknown biases
-How many respondents do you need?
-If you have a randomly drawn sample, you can calculate exactly how
many people you need for a representative sample.
-1500 typically works well
-Why Randomness Matters
find more resources at oneclass.com
find more resources at oneclass.com
-For the 1932 election, the Literary Digest Poll was mailed to addresses
culled from phone books, auto registrations, and other sources.
-Millions of respondents
-Predicted FDR would win and he did
-Used the same process in 1936.
-This time they predicted that Alf Landon would beat FDR in a
landslide.
-They were wrong!
-Had LD suspected social class bias they could have controlled for
it but had they used random selection it would have been
automatically controlled for even if they did not suspect it.
-Public confidence in polling was further eroded with the 1948 election
-Pollsters predicted that Thomas Dewey would defeat Truman in 1948.
-They were wrong.
George Gallup said that polling was still “an infant science”
Sampling
-There are also potential problems with the surveys themselves
-Problems measuring public opinions:
-Question Wording Effects( Welfare vs. the poor)
-Question Order Effects( Freedom, freedom, freedom, abortion)
-Question Loading Effects( giving information that would make one favor
a certain choice)
-
Double-barreled
Questions( Decreased Education and Welfare)
-Confusing Questions(Did Nazis happen)
-Social Desirability Bias
-People don’t want to appear clueless
-People don’t want to give socially unacceptable answers
Polling: A Challenging but Worthwhile Endeavor
-What does Public Opinion Look Like?
-To understand public opinion, we must understand the shape and the
stability of the distribution of public opinion.
-Shape of the Distribution: the shape of the opinion distribution depicts the
pattern of all the responses when counted and paired.
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Special counsel robert mueller accuses trump campaign aids of tax evasion, money laundering, and not disclosing activities(lobbying) undertaken on the behalf of a foreign government. George papadopoulos, a campaign foreign policy advisor to trump, pleaded guilty to lying to investigators about meetings with russian agents. There is no formal investigation at this time into the clinton campaign"s ties to. The collective attitudes of citizens on a given question. We measure public opinion through surveys by sampling a population of interest. A sample does not exactly match the population; it reflects the population with some predictable degree of accuracy. Each individual should have an equal probability of being selected. The larger the sample, the more accurately it represents the population(depending on random selection!) Selection is random if every unit has an equal chance of being chosen. Random selection eliminates all known and unknown biases.
