First of all, it was difficult for me to find a proper title for the question that I am asking here.

I have the following data:

Class       Branch  LA_type Method_type     Method_call     Branch_type     Branch_condition    Tested_parameter
Goal        12      Smooth  public static   never called    IFNE            TRUE                 String
TreeApp     20      Rugged  constructor     none            IF_ICMPGE  FALSE                     int
Password    4       Smooth  private         never called    IFEQ    FALSE                        int
XMLParser   9       Rugged  constructor     none            IFNONNULL   TRUE                     String
MapClass    33      Smooth  public          never called    IFGT    FALSE                        double

I want to know which of the variables (Method_type, Method_call, Branch_type, Branch_condition, or Tested_parameter) have an impact on the LA_type of each branch. For example, does the Method_type make the branch smooth or rugged? I need to ask the same question with the other variables. It is obvious that the Method_type and Method_call have an impact on the LA_type of each branch, since the constructor method type leads to rugged branch and a never called method leads to a smooth branch.

How can I answer such a question statically? In other words, what type of test is better to use to find out whether each variable has a significant correlation with the LA_type? Does the Kruskal–Wallis test work in this case?

  • $\begingroup$ Are you saying you wish to model LA_type as a function of the six other variables based on only these five records? $\endgroup$
    – whuber
    Commented Oct 7, 2019 at 15:15
  • 1
    $\begingroup$ @whuber this is just a sample. In reality, I have hundreds of records. $\endgroup$
    – Adam Amin
    Commented Oct 7, 2019 at 15:16
  • $\begingroup$ Okay--but that's not what you have written. Regardless, search our site for logistic regression. $\endgroup$
    – whuber
    Commented Oct 7, 2019 at 15:55
  • $\begingroup$ @whuber I'm sorry how can the logistic regression help? $\endgroup$
    – Adam Amin
    Commented Oct 7, 2019 at 16:11
  • 1
    $\begingroup$ It is the procedure of choice for analyzing how a binary response depends on other variables. $\endgroup$
    – whuber
    Commented Oct 7, 2019 at 16:12


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.