Tuesday, June 14, 2005
Polling
Just one comment while I'm thinking about it. I've been on Zogby's online polling list since 2000, and I've seen and answered a lot of questions. It's usually easy to tell which side is asking the questions, even though they are carefully and neutrally phrased, but it's not easy for the pollster to tell which side is answering the questions. They always stop short of acquiring real information. Why? Because they assume 'teams' that don't necessarily exist. Polls always ask which issues you're most interested in, and sometimes go into great detail to get the strength of your passion. But I've NEVER, not even once, been asked which direction my interest takes.
Example: Let's say I rank my interest in environmental issues as #1. By the 'team' assumption, the poll's consumer will count me among those who think the EPA should control everything and humans should be destroyed. In fact, I think the EPA should be abolished and humans should rule the world. Similar malassumptions happen on many other subjects. "Are you concerned about how the war in Iraq is going?" Answer, yes. The poll's consumer will count me among the weak-kneed jellyfish, when in fact I'm concerned that we didn't kill Saddam when we captured him, and I'm concerned that we failed to transform Mecca into a glowing crater on 9/12.
Why is this especially bad? Because the administration gets no indication of how many people want it to fight HARDER. Say 53% of the country "thinks we're going in the wrong direction". The assumption is that these 53% are all dhimmi, so the admin tends to moderate its efforts accordingly. How many of those 53% are actually displeased with chickenshit moderation? The poll doesn't give any signal. ¶ 11:35 AM
Example: Let's say I rank my interest in environmental issues as #1. By the 'team' assumption, the poll's consumer will count me among those who think the EPA should control everything and humans should be destroyed. In fact, I think the EPA should be abolished and humans should rule the world. Similar malassumptions happen on many other subjects. "Are you concerned about how the war in Iraq is going?" Answer, yes. The poll's consumer will count me among the weak-kneed jellyfish, when in fact I'm concerned that we didn't kill Saddam when we captured him, and I'm concerned that we failed to transform Mecca into a glowing crater on 9/12.
Why is this especially bad? Because the administration gets no indication of how many people want it to fight HARDER. Say 53% of the country "thinks we're going in the wrong direction". The assumption is that these 53% are all dhimmi, so the admin tends to moderate its efforts accordingly. How many of those 53% are actually displeased with chickenshit moderation? The poll doesn't give any signal. ¶ 11:35 AM
