CBS and Dan Rather are really really sorry for putting forth forged documents and doing shoddy journalism, so everyone in the blogosphere be nice to them now. Let’s also not speculate exactly who the documents originated from, because that’s just petty.
In other news, it is true the DNC is now giving me money, but that will not affect my blogging.
Archive of entries posted on 20th September 2004
I’m Not a Highly Paid Political Consultant But…
Just to Be Mean
Question for Readers
I Demand More Free Ice Cream! More!
I’ve been spoiling you people lately with all the posting, but works heating up so it won’t be as frequent.
Still, I can post at any moment and you’ll never know when!
Why, I could even be posting… NOW!
BTW, according to Drudge, Kenneth finally found the frequency. Took him long enough.
What Would Glenn Reynolds Call This?
Poll My Finger
What’s with all these wacky polls? Some have Bush up by a huge amount, and some have Bush up by only a lot. Why the variance? Let’s take a look at how polling is done to understand.
First, scientist create a scientifical list of people to call up using science. A good scientifical list is the most important thing to good polling. By law, cell phone numbers can’t be solicited, so people who use cell phones as his or her only phone are excluded from polling. Such people are weirdo freaks, though, and best not heard from.
Next, you need to call everyone on the scientifical list. This takes people making one nickel over minimum wage working in a phone bank. They are given a very specific procedure on calling people. Usually it goes like this:
POLL TAKER: “Hello. Would you like participate in this poll?”
POLLEE: “No!” (hangs up)
(two seconds pass; phone rings again)
POLL TAKER: “Hello. Would you like participate in this poll?”
POLLEE: “I already told you NO!!” (hangs up)
(two seconds pass; phone rings again)
POLL TAKER: “Hello. Would you like participate in this poll?”
POLLEE: “Fine! Just give me your @$%& poll!!!”
What if the person initially agrees to the poll, you ask. Then this is the procedure:
POLL TAKER: “Hello. Would you like participate in this poll?”
POLLEE: “Sure.”
(poll taker hangs up; two seconds pass; phone rings again)
POLL TAKER: “Hello. Would you like participate in this poll?”
POLLEE: “I already said I would.”
(poll taker hangs up; two seconds pass; phone rings again)
POLL TAKER: “Hello. Would you like participate in this poll?”
POLLEE: “Will you just give me your @#%& poll!!!!”
The most important thing when polling someone is that the pollee be very angry. Angry people tend to be more opinionated, thus making better poll data.
So what to ask? Well the most important thing when polling someone is how the questions are phrased. Consider for instance the question “Will you be voting for Bob or Ed?” versus “Will you be voting for Bob like all decent, God-fearing people are, or will you be voting for Ed along with the pedophiles?” Despite those questions being about the same thing, that phrasing will get different data. Crazy but true!
So, if you aren’t getting the poll data you want, change the questions. Then you’ll have polls validating what you think. It feels good, just like endorphins.
So now we get back to the question I asked at the beginning: Why are the current polls on the presidential election so different from each other? I have no idea, but that query made a good lede, don’t you think?