Rea­son to smile?

Today, Ras­mussen Reports pub­lished results of a poll per­formed yes­ter­day. In it, Newt Gin­grich is a mere three points behind Mitt Romney…exactly the sam­pling mar­gin of error for the poll. In other words, assum­ing there isn’t some other form of bias, the two can­di­dates are sta­tis­ti­cally tied.

To some degree, the surge is bol­stered by the CBS News/​New York Times poll, also pub­lished today, which cov­ers the past five days. That poll shows Rom­ney with a seven-​​point lead, which would make sense if the surge hap­pened after the last debate. It’s also worth not­ing that this poll cov­ers 340 reg­is­tered vot­ers, while Ras­mussen cov­ers likely voters.

On the other hand, Gallup’s poll (also pub­lished today), cov­er­ing the past four days, and includ­ing a thou­sand reg­is­tered vot­ers, indi­cates a 17-​​point lead by Romney.

One of these polls is not like the others…

What might it mean? One thing we don’t know about the CBS or Gallup polls is the dis­tri­b­u­tion of respon­dents over the var­i­ous days. If there is a late surge, a bias of respon­dents toward the ear­lier days could mask the surge some­what. The polls end­ing prior to the debate show Rom­ney with a com­fort­able lead in the neigh­bor­hood of 20 points. That the post-​​debate polls all show tight­en­ing, albeit to vary­ing degrees, sug­gests that the Gin­grich Surge might be real.