We do not know what you are looking for when you ask "what is the best model". The three models tell you the same thing.
Say "the lower the IDe value, the better the model" or "the higher the R ^ 2 value, the better the model", is not accurate.
- L & # 39; id in a wording represents a bias point on the graph for a given subject.
- Higher regression value in a wording represents the criterion. In the case of Fitts, it is MT.
A particular model can be more predictive for your situation. To do this, you need to do more studies and compare the results to your regression.
You can not just say a wording and say "it's better".
The Fitts model is the original and it has been shown that it applies in a number of conditions.
The Welford Model is a 2-factor variation that takes as its goal distance and target width as influencers, that can not be directly compared to the Fitts factor equation 1. You can not draw Fitts' and Welford on the same graph and simply compare the two lines. If you really calculated the width as a distinct influence, looking at your data, I'm not sure if it was done, the Welford model generally provides greater predictive power.
Shannon's model is another variant. He has no doubt become more popular when he works in human-machine interaction. At least in part that ISO 9241 recommends its use. So if you're looking for the "official" variant of the international standard, use Shannon's.