I'm playing with some data, exploring really, testing it to see if there is a paper worth pursuing. As you can see above, I'm looking at how people "map" television programming. Ignore the "z" in front of all the variables, that's a naming convention I use in SPSS. For a more serious presentation I'd do it differently.
So what we have are 48 television programs and how a national sample "maps" them using a procedure called ALSCAL. Look at Big Bang Theory. All by itself high on the Y-axis along with, on the other quadrant, NCIS. How you do this is try to figure out what the X and Y axes are measuring. So looking only up and down, along the Y-axis, it looks like entertainment versus news, but below that horizontal line we all see Dancing with the Stars and American Idol. So maybe it's scripted television versus not? If we look left-to-right, along the X-axis, we see far to the right some news shows and 60 Minutes, while far to the left we see some Fox News progtams and Jon Stewart (this is 2012 data). We might guess that the X-axis is ideology in some fashion, as in ideological (far left) or not (far right). Or maybe it's cable vs broadcast? No idea.
Another way to do this is look at quadrants and see what holds those shows together as compared to the shows in a different quadrant. So the top left quadrant as Big Bang Theory, a bunch of Fox stuff, an MSNBC show, and the Mentalist, among others. What holds that all togehter as compared to the bottom left quadrant? Not much that I can tell.
And of course as I do this, I have in mind looking at diffent kinds of people and how their "cognitive maps" of television differ.