Pingu wrote: > I assembled a phonebook of 10 names. In each condition, every name was > dialled in a random order of presentation. Time for a successful dial > was recorded, being the time from the presentation of the name, to the > system ackonwledging selection of the correct name by the user. I was > the only participant in all conditions. I am now left with 10 times in > 8 conditions and a mean dialling time for each condition.
OK. You have a situation similar to that in some areas of psychophysics and neuropsychology where you are looking at data from just one person. The first place to start is probably just to report/plot the means and median times (the medians may be more useful here because the times may be skewed as Bruce, I think, pointed out). As this is a "pilot", I'm not convinced you need to do much more than look at the means and distributions in something like a boxplot. If you are just going to select the most promising configuration for later work then _maybe_ that's all you can reasonably do without much more time/resources. This isn't the best way to evaluate the configurations (though it is slightly better than doing so on the basis of no data at all). It depends too heavily on one person's characteristics (e.g., prejudices, hand/finger size, previous experience and so forth). Thom . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
