0

I have here with me, two methods of intervention; a standard care method, and a specialised intervention. Different individuals receive different method, its all randomly allocated. I am trying to find the association of the specialised intervention and a 'well being score' which is a numerical value. What is the best way to find any associations? E.g. logistic regression model?

kjetil b halvorsen
  • 63,378
  • 26
  • 142
  • 467
squashguy919
  • 43
  • 3
  • 1
    Would you consider the well being score to be continuous or ordinal ? – Sal Mangiafico Jan 04 '20 at 22:40
  • 1
    it is indeed, continuous. – squashguy919 Jan 04 '20 at 22:50
  • Probably the most common method as a measure of association between a dichotomous variable and a continuous variable is point biserial correlation. It is my understanding that this is the same as Pearson correlation where you use the dichotomous variable as one of the variables. A significance test can be performed for this correlation as well. – Sal Mangiafico Jan 05 '20 at 12:44
  • For association measure, see https://stats.stackexchange.com/questions/102778/correlations-between-continuous-and-categorical-nominal-variables but logistic regression could also be useful. – kjetil b halvorsen Jan 05 '20 at 14:31

0 Answers0