If there is correlation, further analyze the correlation strength and direction.

The following table is a summary of commonly used correlation analysis, so that everyone can find an analysis method suitable for their own data.

Please click to enter a picture description.

Classification variable:

Disorder: gender (male, female) and blood type (A, B, O, AB);

Orderly: obesity grade (severe obesity, moderate obesity, mild obesity, non-obesity)

1 correlation analysis

Analyze the correlation between quantitative variables, such as people's height and weight; Correlation between relative humidity in the air and rainfall

Type:

Pearson correlation coefficient (suitable for quantitative data, and the data satisfies normal distribution)

Spearman correlation coefficient (used when data does not conform to normal distribution)

Kendall τ-b correlation coefficient (ordered classification variable)

Case: Study the relationship between human height and weight.