Hmmm.. It's hard to tell from this.  Can you send us traceback()? Also, please make sure you update both R and Zelig to their latest version. Best, Kosuke -- Department of Politics Princeton University http://imai.princeton.edu On Mon, 9 Mar 2009, Kenneth Benoit wrote: > Hi - I am having the following problem with the plot.ci method: > > [note that pspend_total is the total spending, >        incumb is 0 or 1 depending on incumbent or challenger] > >> z.out <- zelig(wonseat ~ pspend_total*incumb+m, model="probit", data=dail) > > How to cite this model in Zelig: > Kosuke Imai, Gary King, and Oliva Lau. 2007. "probit: Probit Regression for Dichotomous Dependent Variables" in Kosuke Imai, Gary King, and Olivia Lau, "Zelig: Everyone's Statistical Software," http://gking.harvard.edu/zelig >> >> x.incumb <- setx(z.out, pspend_total=1:30, incumb=1, m=4) >> x.chall <- setx(z.out, pspend_total=1:30, incumb=0, m=4) >> s.out <- sim(z.out, x=x.incumb, x1=x.chall) >> plot.ci(s.out, xlab="% Candidate Spending in Constituency", > > +         ylab="Probability of Winning a Seat") > Error in plot.ci(s.out, xlab = "% Candidate Spending in Constituency",  : > x and x1 in vary on different dimensions. > > > Any ideas before I have to dig in and compute this from the return values of sim()??  Help much appreciated! > > Ken > > > Kenneth Benoit > Professor of Quantitative Social Sciences > Head, Department of Political Science > Trinity College > Dublin 2, Ireland > http://kenbenoit.net > Tel: 353-1-896-2491