Like all good researchers, the authors of the first study saw an opportunity for a follow up study which also included questions that might play to left or progressive leaning individuals. A nice summary of the new findings is here, and the link to the full article is here.
The bottom line: when the correct answers to the questions are consistent with more liberal view points, the leftward leaning survey takers do a lot better than their conservative counterparts. In fact, the authors conclude:
However, the new test consisting of all 17 questions yielded results that vitiated prior evidence of the left being worse. Now, all groups do poorly, with each group tending to do relatively poorly on the questions challenging its positionsHopefully the Wall Street Journal will publish a follow up to their original article...although I'm not betting on it.
Via: Tyler Cowan, Marginal Revolution.