Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • graphing partial correlations

    Hello! I am wondering if there is a command to plot partial correlations in STATA? When I use the pcorr command to calculate partial correlations using two continues variables (controlling for two other variables), I am getting a p-value of 0.026. Because I don't know how to plot partial correlations in STATA, I decided to calculate residuals values and plot them using a twoway scatter graph command. Thereafter, when I use pwcorr to look at the correlation coefficient between the residual values to make sure that my scatter graph mirrors the values I found using pcorr, I see that the correlation coefficient is the same, but my p-value has changed slightly to 0.017. I can't figure out why my p-value is different and I'm trying to circumvent using residual values, because something is altering my p-value. Does anyone have any ideas how to plot partial correlations directly in STATA? On the other hand, does anyone have an idea about what might be altering my p-value?
    Thank you so much for your consideration!

  • #2
    An addendum: the number of observations utilized in the calculations is the same for both pcorr and pwcorr using residuals

    Comment

    Working...
    X