True, but the results would likely still be useful, particularly when choosing colors for web sites which will be displayed on the same distribution of screens.
I think this is likely, but easily validated -- you'd probably see a couple of clusters or some latent variables, but the raw data would still likely be useful for getting average response characteristics.
True, but the results would likely still be useful, particularly when choosing colors for web sites which will be displayed on the same distribution of screens.