stat 22000
You have completed a survey of 1,000 voters from the 2020 election. For each voter, you record the following variables: Quantity measured Variable name Income in 2020 income_2020 Income in 2016 income_2016 Age age Education level (HS, Some College, College, educ Graduate) Party registration (Republican, Democrat, party Independent) 2020 Presidential vote (Biden, Trump, Didn't Vote, vote Other) For each of the following research questions, give a formula for the statistic you'd calculate and the test you'd use (2, t, or chi-square). If your test includes standard errors or degrees of freedom, write down the formulas you'll use to compute those. You may use notation like mean(age) to refer to sample means, and SD(age) to refer to sample 803. If you need to refer to differences, you can use notation like mean(income_2020 - income_2016) or SD(income_2020 - income_2016). Where appropriate whether you'll conduct a one-sided or two-sided test. If you will do a regression, say what you will use as the predictor variable x and what you will use as the response variable y. If you need to refer to a subset of the sample, for example the average age of HS graduates, you can write it like this: mean(age | educ=HS). If you need to refer to a proportion, you can say prop(educ=HS). If you use another notation, that's fine too, as long as it's clear what you're trying to say. For example, if you are asked to test the hypothesis that the average age is equal to 30, you'd calculate the statistic t = (mean(age)-30)/SE , SE=SD(age)/sqrt(n) , df = n-1, two-sided. a) Did incomes increase between 2016 and 2020? b) Did 2020 votes depend on education level? 0) Predict income given age. d) Was the average income in 2020 less than $50,000? e) Trump voters have lower average income than Biden voters. f) Trump voters are less likely to have college degrees