How's this for a loaded question?

The Tribune likes to run it's little non-scientific opinion poll on the issues of the day. This is no place for a statistics lesson, even assuming I could give one. Generally speaking, though, in order for a poll to provide an accurate representation of the views of a body of people, there are mathematically determined numbers of people that must be questioned and, of course, they must be questioned at random, not by self-selection.

The Tribune 'poll' likely fails the first hurdle on most days because not enough people respond to provide a number truly representative of Great Falls. Also, the respondents are self-selected, i.e. people choose to respond or not depending on whether they are interested in the topic. Thus, it's really nothing more than a little Gannett party game.

The problem arises, I guess, from the fact that we don't really know how many people can tell the difference, or know the difference, between a scientific poll and a non-scientific poll. Thus, how many people read this thing, assume it represents the views of the community, and adjust their views accordingly like good little sheeple?

Today, the Tribune ran a front page piece about the State of Connecticut suing the federal government over the No Child Left Behind law. As an accompaniment, the Tribune ran it's "Opinion Poll" on the same subject.

The question? "Should the federal government fully fund the 'No Child Left Behind' program?" Now, is that a loaded question or what? They have a place to say "yes," a place to say "no," but amazingly no place to say "I think it is already funded adequately."

Can you say "push poll?" By our newspaper, no less! Amazing.

1 comment:

david said...

I "enjoy" the totally foolish questions, such as: "Is summer over?"