What is Statistical Association?
What do we mean when we say two or variables ”associate” or ”relate”? Understanding
this question is critical to understanding a lot of the statistical universe.
Essentially, association means that the pattern of data in one variable seems to occur in a certain manner
that is related to the pattern of data in one or several other variables.
If this seems a little technical or hard-to-digest, consider some really simple examples:
-
You may run the market research for a retail chain, and notice that daily sales of water and ice-cream tend to rise or fall together. When water sales increase, so do sales of ice-cream.
-
People with blond hair are more likely to have blue eyes than people with dark hair. There is a pattern relating the two variables (hair and eyes).
-
In the textbook case example (see Chapter 1), you may find that customer size (an ordinal variable measured as ”small,” “medium” and ”big”’) seems to be related to sales. Later, we will see that medium-sized companies actually have the highest sales, then big customers, then small customers. There is an association between size and sales.
We see here that the data in one variable seems to relate to what is happening in
another variable. This is variable association.
Last updated: April 18, 2017
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.