Monday, August 24, 2009

Polls

Just because polls give certain information about the questions Americans were asked, doesn't mean the results reflect what is best. Any of us who took statistics and research courses know that results depend on how the survey is formulated, which method is used, assurance of a representative sample, inclusion of variables, and all that good stuff! I participate in Harris polls and don't like some of the answers I have to choose. Some of them allow me to express my opinions pretty well, but some don't.

What I'm saying is that I don't put a lot of credence in polls in general. The validity of the poll is usually unknown, and we all know about how many Americans vote. Just because the majority of people choose certain answers and vote a certain way doesn't make it the best choice. Look what they do on American Idol, Dancing with the Stars, and other competition shows where the audience is allowed to vote, and especially on the important elections where it affects the whole country, states, counties, and cities.

8 comments:

Pseudonymph said...

Interesting view on polls, Joy. Reflected by the question 'Have you stopped beating your wife?' Yes/No
Pose that to the nation, then graph the results...

miss alaineus said...
This comment has been removed by the author.
miss alaineus said...
This comment has been removed by the author.
miss alaineus said...

aside from my usual disregard for all things related to extraneous key strokes i can't spell today!

my grammar has apparently seen better days as well. sheesh i hate getting up early and the trying to think!

lol i was really into stats (and still am)

statistics 101: numbers lie. the answers depend on what is asked, who does the asking and whom they ask it of.

when i teach this to my students i set out with a question like 'how many 5th graders like school pizza' and then i proceed to only ask the people in the class that i know live and die by the little caesar's we get for our school lunch.

my fave abusive trick of statistics involves the use of equivalent fractions, when the pollsters ask 5 people and 4 people agree and then they go on to make the leap that the results are based on 80% of the *population* and not 80% of the *population polled*

again, numbers lie. the answers depend on what is asked, who does the asking and whom they ask it of.

of course to preface this i am the one who trusts the validity of the enquirer and TMZ over faux news.

xxalainaxx

Beth said...

So true. I always take a poll with a grain of salt. Every once in a while, I'll put up a poll about something on my blog (my favorite one was to determine whether Rahm Emanuel is a hottie--he is, according to 86% of those who voted in the poll), and it's amazing how easy it is to word the questions in such a way to skew your poll. If it's easy for me to do, you knew it's even easier for professional pollsters! Hugs, Beth

Joy said...

Funny, Miss A! I know what you mean. I used to do that kind of thing when I was teaching, too.

Good point, Beth!

Joy said...

Right you are, Pseudonymph! (no surprise there)

Berry Blog said...

Popular is all that counts...we all know that silly girl