# Linear Correlation

If Linear Correlation test is selected, the correlation test is calculated. For this calculation there must be at least 3 or more observations. This calculation calculates if there is a significant correlation between the two compared columns. When there is a relationship (correlation) between the two variables this relationship can be calculated. See Regression.
The correlation does not work in DOE mode because there are no pairs in this mode.

## Interpretation

• The smaller the p value is the more likely there is a significant correlation between the variables.
• Always check the correlation graph.
• Develve uses the commonly accepted value of p < 0.05 for significance.
• For a good power (0.8 in Develve) the amount of observations must be bigger than the minimum calculated.
• Alway check the correlation graph.
• To see what the relation is between the data sets check the Regression result.

### Colors of the cells

• Green
No significant difference
• Yellow
Significant difference  Linear Correlation No correlation

## Formula for calculating the correlation r value. ## Formula to convert the r to the student t value With the t value and the degrees of freedom minus 2 can the program calculate the p value.

### Sample size The result is the minimum sample size of pairs.

## Legend

n = Amount of pairs = correlation = significance 0.05 in the sample size calculation (2 sided) = Power 0.80 in the sample size calculation

## Example

Between data-sets A and D is a correlation (Row Correl p <0.05), and the sample size is big enough.
Between data-sets A and C is no significant correlation and the sample size is to small. Data file

# Multiple Linear Correlation

Basically the same as the Linear correlation but then for more than 2 data sets.

## Interpretation

• The smaller the p value is the more likely there is correlation between the variables.
• Develve uses the commonly accepted value of p < 0.05 for significance.
• For a good power (0.8 in Develve) the amount of pairs must be bigger than the minimum calculated.
• Alway check the correlation graph.
• To see what the relation between the data sets check the Regression result.

### Colors of the cells

• Green
No significant difference
• Yellow
Significant difference

## Example

To use the Multiple Correlation test "Menu: Tools=> Multiple Correlation". Then select the data-sets to test. Only between data-sets A and D is a correlation (Row Correl p <0.05). When clicking on Graph the a matrix of correlation graphs will be displayed. In graph it is visible that there is a correlation between A and D. Data file