Beckee,
I think you've raised a lot of good questions throughout this discussion as pertains to perception and goals for students who may or may not eventually attend college. There is a big difference between talking about college in a classroom full of kids who already see college as a possibility and talking about college in a classroom full of kids who either have not considered it or who assume that they can�t have access. It occurs to me that more caution is needed in the classroom full of kids who are being raised with an expectation that they will go to college. Those are the kids who are likely to be defining success very narrowly and closing off other career options. Talking about college in a classroom full of kids who otherwise wouldn�t consider it is important. Because going to college is outside the norm for their community, they are less likely to close off other options in response, but may never consider college otherwise. That got me wondering about all of those statistics related to outcomes for college graduates and how they might apply differently in different subgroups of the population. I recall a friend of mine, who is African-American, talking about how she would never go to the grocery store in jeans and a tee-shirt or a pair of sweats because as an African-American woman she was treated with suspicion whenever she entered a store in something other than professional dress. In the context of this discussion, that got me wondering whether the lack of a college degree impacted African-Americans differently. In a culture that still has plenty of racism to go around, is the college degree necessary to acceptance in the job market in a different way than it is for white Americans? A quick search led me to:

this report

I have not read the whole chapter, but I looked to see if I could get an answer to that question and what I did find (in table 389) is that for those 25 and older, in the most recent year documented (2009), unemployment dipped below 10% ONLY in the segment of the African-American population with bachelor�s degrees. By contrast, within the same age demographic, unemployment for whites only climbed above 10% for whites without a high school diploma.
My point is not that everyone should go to college, but that we need to take into account that the impact of doing something or not doing something will play differently when combined with other significant societal factors. Just as it turns out that medical research done exclusively on males is often misleading when applied to females, discussions of college outcomes for a demographic that is disproportionately white can be misleading when applied to racial or ethnic subgroups.
Similar data
can be found here but it is less recent (2003)