+ Reply to Thread
Results 1 to 1 of 1

Thread: Spearman's rho or Pearson's r

  1. #1
    Points: 4, Level: 1
    Level completed: 7%, Points required for next Level: 46

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Post Spearman's rho or Pearson's r




    Dear all,

    For my master thesis I held a survey. For the sake of simplicity let's say part A asks people to rate a job offer on ten job features using a 5 pt likert scale, whereas part B asks people to indicate their job pursuit intensity by measuring five statements using a 5 pt likert scale as well. For now, all statements are answered via a likert scale and (as far as I know) are ordinal, hence correlation should be tested using Spearman's rho.

    However, for the ten ratings of job features I multiplied the likert scale answers (1, 2, 3, 4, or 5) by an importance index. Some features are deemed more important, some aren't, so the constant factors I used to multiply differ per job feature.

    For the job pursuit intensity, I summed up all five answers to get the 'total' pursuit intensity per cadidate.

    In short, I used a small formula to generate ten new variables for part A and one (the sum) new variable for part B. Since these are not the direct answers of the likert scale, but since they are derived from it, should I use these new variables as ordinal as well and apply Spearman's rho, or can I consider them ratio/interval variables and test for correlation using Pearson's r?

    Please know that my knowledge does not extend any further than the things I've just discussed so please try to not make it any more complicated than it already is to me.

    Thank you very much in advance.
    Last edited by JayPee; 01-19-2014 at 10:07 AM.

+ Reply to Thread

           




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats