Argus Polls stink? SDSU professor says issues exist with Mason Dixon Polling
In an article in the Capital Journal this morning by Kate Turnbow, SDSU professor Gary Aguiar (remember him?) notes that this SDSU professor has labeled Mason-Doxon polling as "not the most expensive or sophisticated polling method."
Why? Aguiar notes that because of the use of cell phones and unlisted numbers, accuracy is being sacrificed. He also notes that while expensive polls (such as Zogby) will run a probability analysis on whether a respondent will vote or not, cheaper polls like Mason-Dixon will ask only if a respondent is voting.
As soon as the article is available this morning (about 9 or 10AM) I'll post some snippets of it.
I would note that Aguiar's boss, Professor Bob Burns disagreed.
Why? Aguiar notes that because of the use of cell phones and unlisted numbers, accuracy is being sacrificed. He also notes that while expensive polls (such as Zogby) will run a probability analysis on whether a respondent will vote or not, cheaper polls like Mason-Dixon will ask only if a respondent is voting.
As soon as the article is available this morning (about 9 or 10AM) I'll post some snippets of it.
I would note that Aguiar's boss, Professor Bob Burns disagreed.
Comments
Without reading the article in full, but basing it only on PP's reference to cell numbers, I have to say Gary Aguiar is full of shit on this one.
For someone who has put theirself out their as such a sophisticated politcal operator your comments show why you are still are only suiting up jv for the democrats, and your spon isn't much better.
MD may match their numbers based on the primary phone number but, and this a large but, federal law prohibits the use of automated dialing machines from calling cell phones when conducting polling activities. This means that the company must manually dial the cell phone subscriber in order to comply with federal law.
Today polling companies use a random, yet weighted for party, etc... sampling program that automatically dials phone numbers until the sample size has been achieved through the use of auto-dialers etc...
Now M-D in particular is responsible for pumping out cheap fast polling that is definitely not considered the gold standard amongst political professionals.
Now here's the secret regarding cell phone calls and polling: most firms including M-D are not going to conduct polling and take time out to manually dial the cell numbers unless the client requests that they do. In such an event the costs of the poll go up dramatically.
I highly doubt that the MD polling done here in South Dakota was top shelf, in terms of true sampling in regards to cell phone only users, probability strand analysis and so forth.
As to Gary Aguiar, he is on to something with his comments. As to Chad, stick to workig on legislative races and stay away from trying to spin the dependability of MD polling. With all that said MD is widely used and can be trusted to do good polling but, you have to pay for it and that is probably not the case here.
You are absolutely correct in everything you say.
Auto-dialers cannot be used to dial cell phones.
The MD poll for the Argus probably wan't the most sophisticated.
And like I said, I haven't see the full article.
But unless I see some evidence that excluding cell phone numbers some how skews poll results (keeping in mind that the MD sample looks good from what I've seen), I think this argument really isn't based in reality.
What this really is, is yet another attempt by the newest right-wing media hack in South Dakota (Kate Turnbow) to somehow energize the Unruh crowd because all the polling has shown them behind by a considerable margin.
[Cell only Americans] are younger, less affluent, less likely to be married or to own their home, and more liberal on many political questions.
Yet despite these differences, the absence of this group from traditional telephone surveys has only a minimal impact on the results. Specifically, the study shows that including cell-only respondents with those interviewed from a standard landline sample, and weighting the resulting combined sample to the full U.S. public demographically, changes the overall results of the poll by no more than one percentage point on any of nine key political questions included in the study.
Estimates of the respondents' likely congressional vote this fall, approval of President Bush, opinion about the decision to go to war in Iraq, and other important social and political measures are unaffected when cell-only respondents are blended into the sample. The relatively small size of the cell-only group, along with the demographic weighting performed when it is combined with the landline sample, accounts for the minimal change in the overall findings.
I'm no polling expert, but I can tell you this: It appears to me that about everyone under 30 does not have a landline phone. Common sense would seem to dictate that if you are not calling into that portion of the population, you're going to get less than accurate results.
On the other hand, based on polling data I've seen, if that group is under reported, that probably underestimates the support against the abortion ban and Amendment C. As to impact on video lottery, I haven't a clue as my observation is that most under 30s aren't playing video lotter. That's something old people like me do.
But I could be wrong.
As to Kate Turnbow, I would not dismiss her as a Rightwinger. She is a newcomer to the state and just needs a little educating on the scene here. I've talked with her several times and found her receptive and engaging.
And as the old adage goes, you can catch more flies with honey than with vinegar.
And Anon, Chad is damn good at what he does. He's not playing for the J.V., he's definitely on the AA varsity squad. To dismiss him and his skills is unfair and untrue.
Real intelligence is to be found in such words of wisdom.
The spin here is clearer than a flushing toilet bowl. You haven't quite earned the Dan Pfiefer toielt bowl spin award yet, but keep trying.
what data do you base your claim on that about everyone under 30 does not have a landline?
Can we assume that you are speaking about South Dakota as well because that is the context of this discussion?
It seems to me that the real pros rely on verified data and substantive authority
If you want more people to take you seriously, Chad, choose your words more carefully. (Saying Gary Aguiar is full if it would have worked.)
He doesn't say he's privy to the actual details of the MD poll, though I suspect he is mostly correct in saying that "the most sophisticated" methods weren't used.
He raises points that he has no evidence to back up, and Kate Turnbow decided to print them.
For instance, on the cost of the poll. I've seen many, many polls that cost in the 25,000 range that proved to be accurate at the end of the day. It doesn't take a six-figure poll to get good data, though that is what Aguiar suggests.
The cell phone issue is pretty much dead. Pew (see my above comment) looked into this because they were concerned with the accuracy of their own polls. They concluded earlier this year that it has had no affect on polling results.
So, at the end of the day, while I've agreed with Aguiar on a lot of things in the past, he's wrong about this one.
Who are you to say what you say?
What are your qualifications to make these judgements about a man and his job?
Those who throw anonymous personal and professional slurs are cowards.
Know now what you are. And as a fool shouting through a hole from behind a fence, you are not worth credence.
And I say that as a Republican