As you can see on the Scatter Plot and Correlation page, "correlation" is generally about the similarity of variables in quantitative data. A measure called the "correlation coefficient" is calculated from quantitative data.
Data formats can be broadly divided into two types: quantitative and qualitative, but qualitative data also has a correlation.
In the case of weather data, points A and B can be said to be "highly correlated" because the weather is similar. On the other hand, it can be said that the correlation between points A and C is low because the weather is not similar.
This correlation is multivariate analysis because it looks at the relationships between variables . Independence Test and a href="ede1-2-3-3-2.html">Log-linear analysis have been studied for a long time.
The measure of correlation of qualitative variables is called the "Coefficient of association". The number of associations is calculated after converting a table that is arranged vertically for each variable as shown on this page into a contingency table .
Correlation of qualitative data is measured by the number of linkages, but there are also analysis methods that use the correlation coefficient to analyze qualitative data. Dummy Variable is performed to make quantitative data, and then the correlation coefficient is used.
Using this method is not an analysis of the correlation of qualitative variables, but an analysis of the correlation of individual categories.
For more information, there is a page for Correlation analysis of individual categories .
NEXT 
Log-linear analysis
