Fizban said:
500,000ags said:
You are the only person I've ever heard say that user experience and user opinion are biased. It's amazing.
If you want a valid perspective you have to be like him and have no first hand experience at all...
For the both of you...
It's a sampling issue. When you say, "Everyone wants them because all my friends who have driven it love it. My wife wants her next car to be one!" that's
selection bias because the sampling frame, or the list/material your sample was drawn from, is
your friends and not some list of the general population. You can't extrapolate the opinions of your friends out to the general population and say demand is booming because any inferences generated from a sample are generally limited to the scope of the sampling frame it came from, so the idea that, "everyone wants one," is limited to your group of friends. What you're saying is like asking only democrats who they're going to vote for and then making inferences on the opinions of all voters based on that biased sample.
Their opinions aren't biased, your selection of them is. That's how you get, "Dewey Defeats Truman," Landon beating FDR, and Hillary being a lock. If your target population is the general population because you're making inferences about general consumer sentiment, then you need to randomly sample the general population or a sample frame that covers it.
If you're trying to determine reliability and not general sentiment, then you need a random sampling of vehicle owners and outcomes for those vehicles because the population you're measuring, a specific model of vehicle, is different. The population is different because the question is different. If I want to know how Texas voters will vote, then I need to limit my sample frame to Texas voters, not all voters. Sampling only Texas voters at that point is not just acceptable, it's what you're supposed to do. Sampling a list of all voters results in over coverage, and those who are in the sample frame but not in the population I want to study can bias the results of the sample, which is bad. The same goes for reliability sampling. Your sample frame should now be limited to owners, not the overall population.
And this isn't my, "perspective," it's
statistics.Quote:
Why Sampling Frames are so Important
Let's say you're doing a study on the opinions of US adults on current politicians.
Of course, you don't have phone numbers for *all* adults in the US. But you are able to get a master frame of all available cell phone numbers, which you can sample using random digit dialing.
The target population (US Adults) will in large part be "covered" by the sampling frame (those in the cell phone banks).
However, some cell phone numbers in the bank are owned by children, who are not part of the target population.
Likewise, adults with only a land line telephone or no telephone at all will be not be covered by this sampling frame.
If these adults are different in some way from those who own a cell phone on our survey outcomes, then selection bias may result. In this particular case, it's called coverage bias.
*bold emphasis added
Now, replace politicians with vehicle purchasing decisions, and take that list of cell phone numbers and trim it to your list of friends. That list obviously doesn't come close to full coverage of the general population, and since your small list of friends is not randomly chosen, it is safe to say that those excluded will likely differ on your survey outcomes. Oh look, "
selection bias may result."
If you want to argue the statistics of this instead of making stupid little quips, I dare you because I know you can't and won't. I've clearly explained this twice, and you two keep talking **** with no rebuttal whatsoever. If you think I'm wrong, then prove it. Put up or shut up.
