I am trying to normalize my data in which I collected number of time people pick their hair and their stress level which ranges from 1 (no stress) to 5 (high stress) and try to create a graph in order to see if stress level has an effect on the number of times people pick their hair. Should I normalize my data to stress level 1 which can be considered as a possible negative control? If I do the percentages are mostly over 100% since higher stress seems to have higher number of hair picks. What should I do to normalize my data?

  • $\begingroup$ Can you just plot a bar chart with number of hair picks in each of the five bars? It isn't clear to me why any kind of normalization would be warranted. $\endgroup$ – Dave Nov 30 '20 at 15:42
  • $\begingroup$ Hi! since it was a survey the groups of stress levels have different sample sizes so I was thinking that I should normalize my data. $\endgroup$ – 93386 Nov 30 '20 at 15:52
  • 1
    $\begingroup$ Try dividing each count by the total count. $\endgroup$ – mdewey Nov 30 '20 at 16:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.