[Home Page] [Previous Page]
[Old Roundtable Archives] [New Roundtable]
Posted by Mike Curtis on June 29, 1998 at 15:31:44:
In Reply to: Re: Re: Re: Banner Ad / Online Survey Combinations posted by Randy Hill on June 29, 1998 at 10:40:42:
At a basic level, it is a purely self-selected sample with no known probability of participation. There is no statistical representativeness tying the data to any known population. If 50,000 people (or 50 people) respond to the survey question, I have no knowledge about anything other than those specific people. In fact, I could conduct a survey among a random sample of 100 people in my defined population and have a much better idea of who those people are than I would with my self-selected sample of 50,000.
It isn't even clear to me what kind of people would complete the survey. Would it be anyone interested in the research topic? Only those interested in the sponsor's products? Those that wanted a free coupon, but had no interest in the research topic? Some combination of these characteristics?
You're right that "meaningless" is a strong term, and on occasion I have been known to make extreme statements. :^)
To restate it slightly, for me as a survey researcher, the information would be meaningless because it completely lacks external validity. From a sample survey perspective, that is a critical point. I am willing to concede, however, that others may find meaning in the results.
Subject: Re: Re: Re: Re: Re: Banner Ad / Online Survey Combinations