Interpreting results
Lecture 17
Aidan Combs
Duke University
SOCIOL 333 - Summer Term 1 2023
2023-06-15
Logistics
- Four classes left. No class Monday. We’re almost there! :)
Wrapping up the project
- Project component 3: submit for grading by next Tuesday (June 20)
Final paper
- use (revised!) elements of your proposal, descriptive stats, and results to turn your question and analysis into a (concise) paper.
- More detailed instructions and example will be posted soon
- Due Wednesday June 28 11:59pm
Final presentation
- Short! ~5-10 minutes
- Briefly introduce your question, say a bit about why it’s interesting, show your results, and talk about what you conclude from them.
- Most on Monday June 26 (last day of class)
Final presentation logistics
The (approximate) data analysis process
- Determine topic ✓
- Find data; learn what observations and variables are available ✓
- Write research question ✓
- Describe distributions of relevant variables ✓
- Prepare data frame for analysis ✓
- Describe relationships between variables ✓
- Perform statistical tests ✓
- Communicate results
So you have a p value…
- Rejecting/failing to reject hypotheses is useful
- But it’s also statistics-speak, not effective communication
What do you want to be able to say?
- Something about what the world is like!
- Translate back from hypothesis test language—zoom back out. If you were describing your results to a friend, what would you say you learned about the world?
What can’t we say?
- Knowing the limitations of your conclusions is super important!
- Limitations can arise from several places.
Sampling and generalizability
- Who can your results speak to?
- Is your sample big enough?
- Was it selected randomly?
Causality
- Association vs causation: What’s the difference?
- Association: two things are related
- Causation: one thing drives another thing
- It’s nice to be able to say something about causation, but it’s hard!
- Hypothesis tests speak to association. They sometimes speak to causation as well—it all depends on the set up of your study!
Exercise Q1
- Does this provide evidence that using the internet helps people live longer? Why or why not?
- Write down a plausible explanation for this relationship.
Reading for Tuesday
- There’s an article linked on the course calendar. Mess around with the interactive element!



Interpreting results Lecture 17 Aidan Combs Duke University SOCIOL 333 - Summer Term 1 2023 2023-06-15