|
Correlation analysis is an important tool for studying relationships among different variables. The correlation coefficient r(A,B) expresses the degree of linear dependence among variables A and B. The QC.Expert program computes three types of correlation coefficients: pairwise, partial and multiple correlation coefficients.
Example output
Data:
| Weigth |
Width |
Strength |
Hardness |
| 122.4 |
12.10 |
2.45 |
1715 |
| 134.6 |
12.97 |
2.69 |
1728 |
| 127.7 |
12.41 |
2.38 |
1865 |
| 141.1 |
13.50 |
2.60 |
1830 |
| 140.1 |
13.43 |
2.46 |
1831 |
| ... |
... |
... |
... |
Protocol Output:
|
Correlation analysis |
|
|
Task name : |
Pills |
| |
|
|
No of rows : |
28 |
|
No of columns : |
4 |
| |
|
|
Pairwise correlations (Pearson) |
|
|
Column |
|
|
weight1 - thickness1 |
0.757458252 |
|
weight1 - strength1 |
0.169614309 |
|
weight1 - hardness1 |
0.147036074 |
|
thickness1 - strength1 |
0.098601682 |
|
thickness1 - hardness1 |
0.121716524 |
|
strength1 - hardness1 |
-0.717341174 |
| |
|
|
Multiple correlations |
|
|
Column |
|
|
weight1 |
0.785863793 |
|
thickness1 |
0.758314793 |
|
strength1 |
0.770106008 |
|
hardness1 |
0.767650632 |
| |
|
|
Partial correlations |
|
|
Column |
|
|
weight1 - thickness1 |
0.731468725 |
|
weight1 - strength1 |
0.310442616 |
|
weight1 - hardness1 |
0.288575726 |
|
thickness1 - strength1 |
-0.052821162 |
|
thickness1 - hardness1 |
-0.029867326 |
|
strength1 - hardness1 |
-0.761645673 |
| |
|
|
Spearman correlations |
|
|
Column |
|
|
weight1 - thickness1 |
0.743295019 |
|
weight1 - strength1 |
0.162561576 |
|
weight1 - hardness1 |
0.152162014 |
|
thickness1 - strength1 |
0.077175698 |
|
thickness1 - hardness1 |
0.145046524 |
|
strength1 - hardness1 |
-0.666119321 |
Graphical Output (Statistically significant pair correlations are marked with red line):
|
QC-Expert 
Show Cart