Polls schmolls
I heard a story on NPR the other day that said in the last 4 years the percentage of households that have a fixed landline phone has decreased to less than 85%.
I happen to be part of the wireless phone only demographic, having dumped my landline around 7 years ago. Since I carried a wireless phone all the time anyway, it didn’t make sense to pay for the additional phone that hardly anyone called because they always tried the wireless phone first. Pretty much the only calls that came through on the landline were telemarketers or survey-takers, who I didn’t want to talk to anyway.
In election seasons like the one currently being inflicted on us, you’ll hear the results of an endless number of polls: Candidate A is ahead of Candidate B by X points, Y percent of people favor the rights of cats and dogs to cohabitate, Z percent of the population want the government to take away all of our rights for an illusory sense of security. The list goes on and on.
Here’s the part that makes me wonder: Polls largely rely on responses collected during telephone surveys. Landline telephone surveys.
So now the results of any poll you hear about is the result of collecting “data” from people:
That have a landline telephone
AND are home when the pollster calls
AND are willing to respond to pollsters
who may or may not actually have a clue what the hell they’re talking about
Scary huh?
5 Comments
Other Links to this Post
RSS feed for comments on this post. TrackBack URI
By Not "The LG", October 22, 2008 @ 4:29 pm
Interesting you should blog about this. I was thinking about something similar.
My take on this has to do with the Presidential election.
You see headlines (more as we get closer to the General Election) about one candidate having a certain margin or a wider appeal.
But isn’t this just political slant by pollsters and news organizations? While these types of polls are supposed to be scientific to a degree, I can’t read how they went about determining the polling population. Therefore, I can’t say that their methodology isn’t flawed in an additional way to what you state above.
These polls usually cover America as a whole and therein lies the flaw. The Presidential election is not a popular vote election. The electoral college determines who we place in office. Therefore, it doesn’t matter if 50% plus 1 or 100% of the voters prefer one candidate over the other in Hawaii. To the electoral college, it means the same thing.
As such, you can have wide ranging popularity increases for a candidate in certain states and have it make no difference to the election. The diffence is only in the states that could swing one way or the other.
If the polls don’t factor this in, the polls can misrepresent the predicted outcome of the election. Meanwhile, public perception begins to change, news organizations start talking about how a candidate is leading / trailing, what the candidate needs to do to stay ahead / make up ground, analyze strategy from the the ahead / behind perspective, etc. Suddenly, momentum begins to take over. One has now become described as desparate and the other confident.
Does perception then becomes reality? Are undecided voters in swing states swayed by these polls? Why aren’t the polls exclusive to individual swing states?
On a personal note, I’m not sure how much credibility I give to polls or surveys in general. They are usually not unbiased. Why? Because someone had to fund the poll. I was once told if you find out who funded the poll, the results often fall in line with the position the funder holds. I’ve checked this out a number of times and have found it to be true. But take it for as little as it’s worth. After all, it’s coming from me.
By Not "The LG", October 23, 2008 @ 9:03 am
Newsweek article today that seems interesting regarding polls.
http://blog.newsweek.com/blogs/stumper/archive/2008/10/23/is-this-really-a-one-point-race.aspx
By egeek, October 23, 2008 @ 9:13 am
79% of those surveyed think polls are a load of crap…
By the lone gunman, October 23, 2008 @ 2:43 pm
Have you ever seen Sneakers, the movie… Robert Redford, Dan Ackroyd, that dead guy, River, the guy who played Ghandi? “It’s all about the information”. When people hear that a bank is struggling… What do they do? They withdrawl their money… Effect, the bank fails. Same with the economy… People in the spotlight people say we are in a recession, the economy is broken, even thought the economy is still growing… People listen and start to worry… Put off buying the new car or painting the house… Next thing you know, the economy IS breaking/slowing.
Perception has a lot to do with reality. That’s why it’s going to come down to whomever had the better marketing strategy and NOT for what they stand for…
… People are dumb blind sheep.
That’s why polls don’t really mean too much. When they say +/- 3 percent they really mean +/- 30%.
Who needs “leaders”? Where are the guns, egeek?
By egeek, October 23, 2008 @ 3:12 pm
Speaking of guns, check out this post by Scott Adams on the Dilbert Blog: http://dilbert.com/blog/entry/the_upcoming_riots
The scary part there is that it sounds entirely plausible to me.