If you have "no place to go," come here!

How to read a poll

Valley Girl's picture

Thanks to Lambert, I discovered "fallacy files" taxonomy The taxonomy of logical fallacies. But, there's way lot more at the site if you poke around.

So, I invite you to take the poll from the site, here:
How to Read a Poll

I did, and had to laugh after, at myself. If anyone else takes the poll, and reports back, I'll reveal why I laughed at myself.


Submitted by hipparchia on

or an agreeable person, not sure which! ;) i chose disagree for all the answers.

so now you have to tell us why you're laughing.

Submitted by gob on

so please esplain!

Valley Girl's picture
Submitted by Valley Girl on

Disagreed on the others. Agreed it was a useful poll. And then laughed when I realized that it was a "self-selected" poll. Duh!

Submitted by hipparchia on

i almost agreed on the last one, but caught myself [yes, the self-selection point did sink in just in time].

Valley Girl's picture
Submitted by Valley Girl on

So at least you saved yourself, whereas I didn't.

I thought it was an interesting revelation of my own mind that I ended up getting the last question wrong. I mean, geez, I've spent a lot of time looking at various health care polls, and the various wordings (often prompted by my interaction with Kip Sullivan) and KNOW how much wording can bias a poll. And, the questions re: standard deviation were of course easy to answer, as we girl science geeks know *g*.

I trust you understand why I was laughing at my lack of "nous".

oh, I mean in the Brit sense- I lived in UK for 10 years, where I first learned the word.
2. Chiefly British Good sense; shrewdness

Submitted by gob on

I was thinking about why, later in the day. I think it has to do with an ambiguity in the meaning of "people's understanding".

The question was agree or disagree:

* This poll is a worthwhile measure of people's understanding of polls.

If it had read "This poll is a worthwhile measure of the understanding of polls by people in general," my thought process would have been quite different. As it was, I thought, yeah, the questions would make a pretty good quiz for an elementary stats class, to find out if my students had got the point about confidence intervals. In other words, I was thinking of it as a measure of an individual person's understanding.

It was only as a second thought that the other meaning (measure of understanding in the total population) took priority.

Kinda takes the LOL out of it though, doesn't it?

Submitted by hipparchia on

just between you, me, and the gatepost, i feel so parochial. i've never heard of any of those definitions of nous. i've only ever heard of entre nous.

yep, spotting the standard deviations from a mile away is easy for us girl scientists, which made it easy for the poll designer to distract us from the fact that the sample selection was also a problem.

in case anybody else is wondering what we're talking about, kip sullivan on polls, and really all his writing is a treasure trove.