Hi,

I'm having trouble creating and understanding plots that show a non-linear effect and how this non-linear effect is moderated by another variable. I'm basically trying to create figures like Figure 1 (plot of the non-linear effect, page 1381) and Figure 2 (plot of the moderation of the non-linear effect, page 1382) shown in the following paper. I hope that you have access to the paper, since I'm probably not allowed to insert pictures of the paper.

Kohtamäki, M., Partanen, J., Parida, V., & Wincent, J. (2013). Non-linear relationship between industrial service offering and sales growth: The moderating role of network capabilities. Industrial Marketing Management, 42(8), 1374-1385.

For Figure 1 from page 1381 I use the following syntax:

Code:
regress DV c.IV##c.IV

margins, at(IV=(1(0.5)7))
marginsplot, recastci(rarea) recast(line) plotopts(lcolor(black)) ciopts(color(gs13) fcolor(gs13) lwidth(none))

While this syntax shows my non-linear effect very well, I'm unable to make any statements regarding from what values of my IV the relationship is significant like they do in the above mentioned paper (Figure 1 - the corresponding text is on page 1380, right column). I guess I have to plot the marginal effects to make such statements? However, when I use the following syntax, I just get a linear upward-sloped line and don't see my non-linear effect anymore.

Code:
margins, dydx(IV) at(IV=(1(0.5)7))
marginsplot, recastci(rarea) recast(line) plotopts(lcolor(black)) ciopts(color(gs13) fcolor(gs13) lwidth(none))

I'm quite confused. Can anybody help?