A Data Analysis Dive Into Reddit's Most Upvoted Comments Using R + Other Topics

Topic 1

Have you ever seen a top comment on a reddit post and noticed the time stamp matches the post (i.e. both happened "7hrs ago")?

Essentially, the question is this: Just how time-lopsided are top comments? This question has various interpretations/corollaries:

1. If you make a comment super early, is it more likely to be upvoted?
2. If you make a comment later, how unlikely is it to be upvoted?
3. Are votes based on content, or just being toward the top (and seen)?
4. Because the reddit default sort is "best," users have to scroll to read newer comments; will they, or is there a high "scroll burden"?
5. Do these metrics/phenomenon vary by sub?

Join us to learn what an analysis of 3 million reddit comments from 7000 posts in 57 subs reveals.

Topic 2

How does one interpret data and examine the validity/strength of arguments presented in a theories? John debunks a commonly cited Michigan voter fraud theory that has been convincingly presented using data and statistics. Bring an open mind!

Topic 3 - If time allows

The main reasons behind most state road construction projects are to replace or fix aging infrastructure and to alleviate traffic congestion.
Before these construction projects start, the state & cities will make presentations about the benefits. However, after project completion, an analysis to show the level of benefits realization (such as commute time on I-35W between 7-9am has been reduced from 35mins to 16 mins) is never made. John has analyzed vehicle traffic data before and after major Minnesota road constructions. The results are interesting!

--------

Presenter: John Henderson

John is a Product Development Specialist at a Minnesota Fortune 100. He is a mechanical engineer who has been using R for data analysis and visualization for over 10 years.