Allison,
A quadratic term can been seen as an interactive variable (of its self),
otherwise there is usually more appropriate to graph interactive
relationships using the predicted marginal effect of X on the full range
of the modifying variable z (unless the modifying variable is dichotomous,
then the marginal effect of X when the modifying variable is 0 and 1 is
enough).
For instance, Brambor et al (2006) provides a good introduction and Stata
code on how to do this. Although, several other options on how to
graphically display interaction effects is also available (see 'findit
interaction').
Best,
Johan Hellstrom
Brambor, T., Clark, W. R., & Golder, M. (2006). Understanding interaction
models: Improving empirical analyses. Political Analysis, 14(1), 63-82.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
