Outcome is Ranking Variable--What is the best test?

#1
Hi all,

I am working with a dataset in which the outcome is a rank or an ordered variable. Essentially, participants are given some information that they can read, and they choose the order in which they read it. The context is that they are able to view eight different snippets of this information, and must "click to expand" and read the rest of any one piece of information. Once they have read that piece of information, they click to go back. When they go back, the information that they've just read disappears, and now they must choose among the remaining seven snippets of information to read. They continue this process until they have read all 8 pieces of information.

What is of interest to me is which information the participants choose to read first. The information presented varies on certain characteristics (such as valence). I want to see if I am able to predict which reviews they choose first based off of some linear and categorical predictors. I can use a Friedman test, certainly, but this doesn't allow for prediction of anything other than whether certain pieces of information are read first more than others. Is there another test that will provide more information or be more useful? Is rank regression an option? Are there any recommended readings to start learning about this?

Any help would be appreciated, thanks!
 

hlsmith

Less is more. Stay pure. Stay poor.
#2
How big of a sample do you have? What is "valence"? In general, these scenarios can be difficult to define. Is your sole question, what are predictors of first selection or is it, what are predictors of average when a section was view, or other question?