bensoltoff
commented
7 years ago - Initial thought - the test is biased against the static image because there is no indication that there is even a time dimension to the data
- Think about presenting results in terms of the error rate - which graph leads to erroneous conclusions? Could then test using a basic t-test or something similar. This is more similar to the approaches from the readings
- I don't think the research design properly matches your research question. Why would we expect animation to work better in this instance? Also animation is used to visualize the time variable. For a fair test, the static visualization needs to convey time in some sort of way. Either by labeling the points, or even mentioning each point represents a different year of data.
Homework was rendered as an html file to ensure the animation component ran in the browser window