>> |
Anonymous
>>374927 Done that? Good. Now you can start collecting data. You're going to need to have people rate a variety of these pictures. Probably everyone will not be able to rate every picture, if you have a large set of pictures. But for each person who participates in the study, you need to take down their gender, age, race, place of origin, and any other data you think relevant, as well as keeping their ratings associated with them. This is going to be important when you go to analyze the data, since you'll need to ensure that you're looking at a representative sample of the population. E.g. you don't want to accidentally use only black men's opinion's, or only those of white women, because you'll skew your data hard core.
Now that that's done, and you've ensured the statistical validity of your data, or at least identified biases, you can try to make sense of it. A self-organizing map looks like a shoe-in for visualizing this data.
tl;dr—You're just blowing smoke: you need to rethink your method. At the very least you need to get more info from your respondents here, though asking us to correctly identify our preferences is still going to render your data pretty much useless.
|