# E5: Conduct your own statistical analysis

### Objective

- Conduct a statistical analysis to answer research questions
- Get more familiar with Fitts’ law experiments

### Back on Fitts’ law experiment

At the end of the second class, you participated in this Fitts’ law task experiment designed to investigate the effect of mouse/touchpad vs touchscreen interaction in pointing tasks. Source code is available here.

A first group of participants ran the experiment on their personal mobile device and a second group of participant ran it on their personal laptop. The instructions were to adjust their speed/accuracy trade-off by trying to follow a 4% error rate.

The research hypotheses were the following:

- H1: Pointing using indirect interaction (e.g. using a mouse/touchpad) results in lower error rates, especially for small targets
- H2: Pointing times increase with longer distances and smaller target widths

In addition we make the following hypothesis:

- H3: Pointing times follow Fitts’ law for both direct and indirect devices

The data collected during the experiment is available here. The CSV file provides columns with:

`participant`

number`ismobile`

(True, false), indicating whether the participant ran the experiment on a mobile device or a laptop`d`

(0.35, 0.7), corresponding to the distance to the target, expressed as a percentage of min(screen width, screen height)`w`

(0.035, 0.07), corresponding to the target width, expressed as a percentage of min(screen width, screen height)`orientation`

(30, 60, 90, 120, 150, 180, 210, 240, 270, 300, 330), corresponding to the orientation of the target in degrees. 0 degree has been removed because it corresponds to the first trial after a break`time`

corresponds to the total time to successfully select a target in ms`err`

corresponds to the number of errors made before selecting the target

### Goal

Your objective is to run appropriate statistical analysis to validate / invalidate the three hypotheses above.

Some recommendations:

- present your results in a clear an elegant way to make it as easy as possible to follow your analysis: make good use of markdown syntax, present tables using kable, use graphs when appropriate to illustrate some results…
- justify each step of your analysis
- interpret the results of the statistical tests and write conclusions as it would be done in a paper
- for H3, you will need to run linear regressions, not covered before, using
`lm`

command

**Note that this final exercise will be graded on 6 points, corresponding to twice the weight of other exercises.** You have two weeks instead of one to finish it.

### Submit

Follow the submission instructions on the course information page. Provide the Rmd file and its html output (using the Knit button on the interface) showing the results. Make good use of markdown syntax to present your results in a way that is easy to read and understand. In your solution notes (integrated in the Rmd file), describe any problems you ran into, and the main resource(s) you used (blog posts, online tutorials, stackoverflow posts, papers, textbooks, etc.). These resources should have brief descriptions of what the resource is and how it helped you.