Sure, I was sucked in by the main story about how some polls show the presidential race as much closer than others. But what really captured my attention was the technology issue that was behind some of that variance.
Historically, pollsters have dialed random house phones to get their selection of voters. The problem is that more and more people, particularly young adults, don't have a landline.
The difference can be significant, as pointed out by a Pew research study last month. The organization conducted three separate polls--in June, August, and September. Each time, the difference between cell phone users and landline voters represented at least a 10-point swing (in Barack Obama's favor among cell phone respondents). The blended result was obviously more muted, but still showed a meaningful bump for Obama as compared with landline-only polling.
To be sure, the cell phone issue is just one of many complicating factors for pollsters this year, including how to weight party affiliation and other decidedly non-tech issues. Anyway, I thought the cell phone issue was one worth some attention.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment