Green Room

My friend Pete goes to the diner

posted at 11:24 am on October 7, 2012 by

Pete Ingemi – Da Tech Guy – and I were talking about micro-pollsters the other day, and I’m not alone in finding the sample size, em, lacking, to put it midly,

I took a look at the poll and it didn’t take long to notice some ….interestingissues to wit:

This is the sixth release of an 11-week tracking poll of Latino registered voters. Each week impreMedia and Latino Decisions will release a new rolling cross-section of 300 completed interviews with Latino registered voters across all 50 states. Battleground interviewers are combined across all six weeks and are 267 completed interviews, with Florida accounting for the largest share of battleground states.

So the battle ground figures in this poll are actually mini samples taken over six weeks while opinions might be changing. That’s bad enough but consider one more thing:

Battleground states: FL, NV, CO, AZ, OH, NH, NC, VA, IA, MO (all 7 points or smaller margin in current presidential polls)

Forget the insanity of trying to extrapolate the latino vote over fifty states based on a 300 person sample. Let’s look at the 10 battleground states 267 answers over 10 states.

Pete illustrates his point,

As I mentioned the other day, the Pew Hispanic Center reports that 23.7 million Latinos are eligible to vote in the USA (myself included).

Small sample? Huge margin of error?

Yes. Yes.

And we are to assume that is a significant enough ramdom sample?

Cross-posted at Fausta’s blog.

Recently in the Green Room: