Journal
Week 10
This week we used R to perform statistical analysis on the intensity measures. First we ran a t-test on the mean intensities of all the sessions and of only the clean IPUs. I was happy to see a significant difference between the intensity measures, which shows that our preprocessing with AMT really makes a difference!
Here are some interesting results of our preliminary analysis:
- Smaller intensity differences between lawyers and justices than between justices and lawyers
This indicates that lawyers entrain more than justices, which supports the theory that the less dominant speaker is more likely to entrain to the more dominant speaker.
- No significant difference in entrainment between male and female lawyers
- We were excited to find that differences between justices and petitioners are significantly smaller when the justice sides with the petitioner.
This supports our hypothesis that justices entrain more to the lawyer that they eventually side with! However, we also found that differences between justices and respondents are also significantly smaller when the petitioner wins the case.
For some reason, it seems that justices entrain more to both lawyers whenever the petitioner wins the case!
These are very preliminary results, but there are endless possibilities! There are still a lot more sessions to be preprocessed, more features to be extracted, and more analysis to be performed. But I'm thrilled that I got to see some early results of this project :)
At the end of the week, I gave a final presentation at the lab. I also worked on writing my final paper. My presentation slides and final report can be viewed here.
This has been a wonderful experience, and I am truly grateful to my mentor, DREU, and everyone who made this possible. THANK YOU!
[PREVIOUS]
|