In the latest installment of “Poli-Sci Perspective” over at Wonkblog, I look at a conference paper in which computer scientists Tae Yano and Noah A. Smith teamed up with political scientist John D. Wilkerson to look at the features of House bills which predict that they will make it out of committee. Once we know the words in a given bill, knowing additional information including the bill’s sponsor tells us surprisingly little about the bill’s fate. And in the words that do predict bill survival, we can see more than a trace of our current fiscal logjam. Words associated with spending (including “authorize”) make bills more likely to win committee approval. Not so much for words like “revenue,” and for others associated with taxing.