My brother and I signed up for a market research company who are concerned with testing foods on the population before release. It was a fun process but I just realized how many potential flaws there are in the company's screening process.
When my brother turned me on to the company he said it was an easy way to make a quick $50 bucks. The appeal was that I could spend an hour or so trying new recipes for good pay. All I had to do was wait for a phone call that would determine whether I was a good candidate. My brother instructed me to answer yes to all of the questions in order to be accepted into each test and thus pocket the cash.
Whenever I received a phone call, I would answer however I thought the researcher wanted me to answer in order to increase my chances of being selected. If they asked me if I liked t.v. dinner I would say I love them. If they asked how often I watch game shows I would say almost every morning. I was lying a lot. Now, studying communication research, I'm more aware of how much I may have been contributing to the sample error. And I wonder how many other people acted the same way, more attracted to the incentive than the livelihood of the research. As I was thinking about this, I wondered how this company could improve their filter. Then I thought that a filter may not be necessary if the company started pulling samples from a larger pool of people that are proven to be interested in the beneficence of whatever product is being tested on them. In other words, why don't market researchers look for samples where they know the populations is tied to the research target.
For instance, check Neilsen ratings to find the key market for game shows, then base your calling list off of that demographic. This is pretty common sense I think and so I was surprised by the company I was involved with and how loose they were by selecting samples from random pools.