For Questions 1 and 2: Refer to this “The Daily” podcast (Link listed below) on the Nov. 2019 telephone survey poll conducted by the New York Times and Siena College, with the goal of identifying which Democratic candidate was most electable against Donald Trump.
For Question 3: Refer to this NPR podcast (Link listed below), which compares and contrasts challenges of polling in 2016 and in 2020.
For each answer:
Find a specific piece of information from the podcast to summarize in your response—no direct quotes! I want to see what you think, not what the podcasters do.
Define any bolded words in the questions below your own words, as part of your answer.
**Each answer should be a short paragraph; make sure you keep to the overall 300-400 word limit for the Memo.**
1. How did the representativeness of the sampling methods used in 2016 lead to errors in predicting who would win the 2016 election?
2. Explain what the researchers did in their 2019 survey to correct for the mistakes in their 2016 research design. To answer this question, you can either focus on their decision:
(a) to limit the study to six key states, or
(b) to “weight” their sample to reflect the influence of certain kinds of voters. If you choose this point, explain what you understand “weighting” to be, in your own words.
3. In 2020, pollsters faced a different set of challenges because the pandemic prompted so many people to vote by mail ahead of election day, rather than in person. What challenges did that create around how voting questions were worded in their surveys, and how did that trigger social desirability in ways that were difficult to resolve?
**Important once again to note that this memo must be between 300 and 400 words, points will be deducted if this isn’t included.**
Links to both:
https://podcasts.apple.com/us/podcast/the-daily/id1200361736?i=1000455979482
https://www.npr.org/2020/11/15/935261940/polling-your-leg-the-pitfalls-of-predicting-an-election
