After the framework paper meeting, a few ideas were tossed around about how to view the data.
Overall ~Violin~ Plot
While it looks great, violin plots might not be appropriate anymore. They should be updated to swarm plots where high and low scores are side-by-side and the data points are colored according to the participant.
TST Scatter Plot
Now we can include all the data, but draw linear regression lines for each participant and include either an overall regression or an average regression.
The idea to add a third panel was tossed around, but this time with a random effect (?) introduced to each participant. CC mentioned using the LME4 package in R with pid as random effect.
REM and Restful
Remove the outlier
Beacon Heatmap
The beacon heatmap wasn't so much of a hit, but include all 3 beacons with all four variables to see at least what is going on.
UT3000 Visuals to Update/Add
After the framework paper meeting, a few ideas were tossed around about how to view the data.
Overall
~Violin~ PlotWhile it looks great, violin plots might not be appropriate anymore. They should be updated to swarm plots where high and low scores are side-by-side and the data points are colored according to the participant.
TST Scatter Plot
Now we can include all the data, but draw linear regression lines for each participant and include either an overall regression or an average regression.
The idea to add a third panel was tossed around, but this time with a random effect (?) introduced to each participant. CC mentioned using the LME4 package in R with pid as random effect.
REM and Restful
Remove the outlier
Beacon Heatmap
The beacon heatmap wasn't so much of a hit, but include all 3 beacons with all four variables to see at least what is going on.