Week 3: Representative Surveys
Overview
We start the topic of public opinion with representative surveys. The main reason why social scientists conduct surveys is to infer real-world attitudes and behaviors. However, this presents a very difficulty measurement problem. How should questions be asked to produce an accurate measure of the attitude or behavior of interest?
As the name suggests, a representative survey is administered to a sample of respondents that resemble the demographic characteristics of the underlying population.
The punchline of this week is that conducting a true representative survey via random sampling is near impossible. Instead, most surveys use some form of quota or stratified sampling to ensure that the sample obtained resembles the population of interest.
The lab assignment for this week compares different sampling techniques.
Reading
Schwarz, Norbert. 1999. “Self-reports: How the questions shape the answers.” American Psychologist 54 (2): 93-105
Stephenson, Laura B., Allison Harell, Daniel Rubenson, and Peter John Loewen. 2021. “Measuring Preferences and Behaviors in the 2019 Canadian Election Study.” Canadian Journal of Political Science/Revue Canadienne de science politique
Slides
Click on the frame below and use the arrows to navigate. You can type ? (or shift + /) to see useful commands.
Lab
Sampling and descriptive inference
Monday, January 29, 5 PM