Understanding Predictive Factors for Merge Conflicts

Collinearity diagnostic results


* We use the Spearman correlation coefficient since it is based on rank data and does not assume a linear relationship
Ruby Sample
Factor Metric Strong correlation with cor * p-value
Changes to a common slice - - -
Number of commits Number of developers 0.73830206 0
Number of changed files 0.75464120 0
Number of changed lines 0.74091839 0
Duration 0.75540624 0
Number of developers Number of commits 0.73830206 0
Number of changed files Number of commits 0.75464120 0
Number of changed lines 0.88264661 0
Number of changed lines Number of commits 0.74091839 0
Number of changed files 0.88264661 0
Duration Number of commits 0.75540624 0
Conclusion delay - - -

Python Sample
Factor Metric Strong correlation with cor * p-value
Changes to a common slice - - -
Number of commits Number of developers 0.67053592 0
Number of changed files 0.74207924 0
Number of changed lines 0.69405794 0
Duration 0.72412574 0
Number of developers Number of commits 0.67053592 0
Number of changed files Number of commits 0.74207924 0
Number of changed lines 0.85374442 0
Number of changed lines Number of commits 0.69405794 0
Number of changed files 0.85374442 0
Duration Number of commits 0.72412574 0
Conclusion delay - - -