Hi saralarsen, and welcome to WebmasterWorld.
As I remember, Compete data is derived from Toolbar data. WebTrends reports, on the other hand, are base on actual referrer data to your site.
For your site, therefore, the Compete data will be at best a very small sample of your actual referrers, perhaps with sampling gaps large enough to explain many differences. Demographic skewing of Compete data may also account for some of the difference. Toolbar users tend to be more tech-oriented, eg.
It's well-known that it doesn't take many toolbars to skew Alexa data significantly. This is likely to be true also for Compete.
Other services (Hitwise, eg) use ISP data, or a combination of ISP and Toolbars. Various demographic factors involve what ISPs are sampled, what browsers the Toolbars work in, etc etc.
And on sites that are business-to-business, particularly corporate-to-corporate, it's very likely that many site visitors aren't using either ISPs or Toolbars... so some kinds of b2b data wouldn't be represented well in Compete or in Hitwise.
All this suggests, though, that Compete data is a subset of WebTrends data, and that, within your WebTrends referrers, you should be able to see the referrers that Compete reports. The Compete data might be missing so many referrals, though, that the top phrases might not even be in the same order as on WebTrends.