5-2-13 What creates a valid survey?

First, let’s be clear: If a survey is valid, that means the results apply to the entire population from which the sample is drawn, within a stated margin of error.

The word “survey” is used a lot, but pay attention to whether the term “Statistically valid” is attached.

A lot of non-reliable information gets passed off as a “survey.”  With Survey Monkey and other simple ways of sending out questionnaires available on line, it’s easy to believe that if you just send out a survey on the internet, it will produce reliable information.

If someone chooses to respond to an on-line survey, which by definition is a “self-select” survey,  this is not a statistically valid process.

Telephone interview surveys are still considered to be the most cost-effective way of getting statistically valid information.  There is a margin of error, based upon the size of the sample (generally, a minimum sample of 200 is the industry standard for reliable data about any population segment).  Even with the advent of cell phones, there are companies that create accurate phone lists that marketing research companies purchase.  Overall, random sample telephone interview surveys provide reasonably accurate information about the population from which the sample is drawn.

Then, there’s the matter of how questions are asked…..next week’s Blog.