English subtitles

← Overplotting and Domain Knowledge - Data Analysis with R

Get Embed Code
5 Languages

Showing Revision 3 created 05/24/2016 by Udacity Robot.

  1. In the last exercise, we used alpha and jitter to reduce
  2. over plotting, but it turns out that there's more that we can
  3. do. Let's hear from Mora about how she used her domain
  4. knowledge and a transformation to make an adjustment to her scatter plot.
  5. >> The next thing that I did, was
  6. to take again, the perceived audience size, and their
  7. actual audience size, but this time I transformed
  8. the axes. So this time, it's as a percentage
  9. of their friend count. Some people in this study had
  10. 50 friends, some had 100, some had 2,000, and so,
  11. it actually makes more sense to think about your audience
  12. size as a percentage of the possible audience. All of
  13. the people in the study had shared their post with
  14. friends only privacy, so you'd expect that it would be
  15. bounded by their friend count. So, what we found when
  16. we plotted it this way was that all of the points
  17. are below this line of perfect accuracy, this diagonal line, really
  18. well below. And one other thing I should note about this
  19. plot, we actually ran two different surveys. We ran one survey
  20. where we asked people in a single post, how many people
  21. do you think saw, saw your post? But we also asked
  22. a different set of people, in general, how many people do
  23. you think see the content that you share on Facebook? So
  24. that's what this plot is showing. This is the in general
  25. question, and their guesses are a little bit
  26. higher. But still, people typically think people that maybe
  27. 10% of their friends see their content when in
  28. reality it's more like 40% or 50%, even 60%
  29. of their friends will see their content in
  30. a given month. So that's what this plot is
  31. showing, is the percentage of friends who actually saw
  32. their content in the last month, again, they're underestimating.