CHAPTER 16Statistical Testing of Differences and Relationships

Photo illustration of a woman working on her computer. Another computer is seen parallel to her.

Statistical testing is another one of those areas that has been long settled and has not changed for many years. All of the tests discussed in this chapter apply to samples and assume that the sample we are applying the test to is a random sample. They all address the issue of whether the difference or the relationship we are observing is greater than would be expected due to chance. The chance we are talking about relates to sampling error. All of these tests are inappropriate and not meaningful when applied to any sample that is not a random sample. These tests also assume that we have no measurement error. This issue is discussed in the chapter.

We find that clients sometimes rely too much on statistical significance and sometimes too little. By far the larger group is the one that relies on statistical significance ...

Get Marketing Research, 11th Edition now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.