I definitely believe that there are likely a number of new factors that affect credibility that probably weren’t addressed in the Fogg study, due to the overall newness of the WWW. Elements specifically pertaining to design might be difficult to parse out – it seems as if the original study may have tried to get at that a little bit (“site has been updated”, “site looks professionally designed” sort of peripherally address this?), but I do agree that given the time that has passed, there are definitely sites that appear dated and less “credible” overall.

I am thinking about doing a follow-up study where I might address questions about “new” site elements that people take into account that weren’t prevalent in 1999 (“The site uses AJAX”?). I would also be extremely interested in getting at social questions, given how much our navigation of the web is now socially-mediated. It would be interesting to look at how things like “This site was linked to by a friend’s FB status” or “One of my Twitter followers @’d me with a link to this site” or “This is bookmarked by 1000 people on Delicious” might affect perceptions of site credibility.

I suspect that the MTurk population worked well specifically for a study like this because they tend to be somewhat internet-savvy compared to the current general population – much like the population in Fogg’s study, given that he also recruited online.

Incidentally, this also suggests that sampling bias in MTurk is either less of a big deal than we thought, or that everyone who completed Fogg’s survey is now on MTurk.