This project was used in a study that used the Mechanical Turk platform to test a series of bar chart embellishments for the paper: Drew Skau, Lane Harrison, Robert Kosara, An Evaluation of the Impact of Visual Embellishments in Bar Charts, Computer Graphics Forum (Proceedings EuroVis), vol. 34, no. 3, pp. 221–230, 2015.
The study is built on the excellent Experimentr.js project.
Both data files showing the resulting data are available in the data directory. This is a summary of the data fields in the Bars Data file, the Demographics are pretty self-explanatory:
| Field Name | Description |
|---|---|
| time_start_time-trial | Unix time for beginning of the question |
| bars | (For relative questions) Which of the three bars were being asked about |
| time_end_time-trial | Unix time for ending of the question (if this is missing, it is the same as time_end_trials from the demographics data) |
| bar | (For absolute questions) Which of the three bars was being asked about |
| data-trial | The data used for each bar |
| number-in-series | Position within a given chart type |
| log-error | Log absolute error as reported in the paper |
| totalSequence | Position within the entire study for that participant |
| question-trial | Abbreviated plain english version of the question |
| judged-true | Judged answer minus true answer |
| ans-percentage-diff | (given-correct)/correct |
| max-mid-min | whether the bar in question was the tallest, middle, or shortest |
| number-in-section | Position within absolute or relative questions |
| adjacent | (For relative questions) Whether the bar in question was adjacent to the one it was compared to |
| question-type | Absolute or Relative |
| ans-trial | Given answer |
| postID | ID generated by experimentr |
| chart-embellishment | Embellishment name |
| time_diff_time-trial | Time in miliseconds that it took to answer the question |
| correctAns-trial | Correct answer |