Unlikely Voter

Poll Analysis and Election Projection

Grading the Pollsters: Pennsylvania Edition

Per Real Clear Politics there were six pollsters who took on the Pennsylvania primary, which Joe Sestak won handily by 8, 54-46 over Arlen Specter.

Let’s see who got it right, and also give credit to a particular pollster where it is due.

First, the losers.

  • Quinnipiac’s final poll had Sestak just barely up, 42-41 (51-49 among non-undecideds). Sorry.
  • Muhlenberg College/Morning Call had it tied (50-50 among non-undecideds). Nope.
  • Research 2000/Daily Kos had Sestak slimly ahead 45-43 (51-49 minus undecideds) Try again.
  • Franklin & Marshall had Sestak up 2, 38-36 (51-49 minus undecideds). You lose.

I went into the primary wondering if it would go to a recount, given that almost everyone had this thing close. But the counting barely started when I started to hear Specter was toast, which he was. But now, the winners who defied the norm and had Sestak clearly up:

  • Rasmussen had Sestak up 5, 47-42, which translates to 53-47 when we subtract the undecided. That’s only one point from Sestak’s final tally.
  • Suffolk also got close. They had Sestak up 9, 49-40, which would look like an outlier and easily dismissed when compared with the field. But when subtracting undecided that takes us to 55-45, also one point off of Sestak’s actual result.

Congratulations Suffolk University and Rasmussen Reports. You got one.

Also deserving a nod is Susquehanna Polling and Research. They, too, defied the trend that said the PA-12 special election was neck and neck by saying Mark Critz was ahead of Tim Burns 44-38 (54-46 subtracting undecideds). I said that was suspect because everyone else said otherwise.

Well, Critz beat Burns 53-46. Well done, Susquehanna.

Possibly Related Posts

Comments

No Responses to “Grading the Pollsters: Pennsylvania Edition”

Write a Comment

Comments are closed.