poverty statistics

I doubt very much such a statistic is available. Crimes occur no matter which country, culture, or educational level.

Predisposition is an oxymoron, because nobody is immune to crime.

I agree that you'll not find race a great indicator.

First of all, do you consider what people are charged for, or what they have actually done? It's understood by many that those put in jail aren't necessarily all guilty of their crimes. Inmates are proven innocent all the time.

Second of all, different regions will have different crime rates by race.

Anyway, try looking at the US census bureau data pages.

This might be a good place to start - it won't really answer your question, but it may help you develop some better search engine key words.

http://www.ojp.usdoj.gov/bjs/glance.htm#Corrections

You might find something in this groundbreaking book.

http://www.quarterlyconversation.com/TQC_7/Poor_People.html

http://www.harpercollins.com/books/9780060878825/Poor_People/index.aspx

THANKS, lol THE CENSUS BUREAU, ITS SO OBVIOUS I SHOULD SLAP MYSELF IN THE FACE.

Be mindful that the US Census Bureau's statistics starts with the wrong assumption based on income for all. Their stats doesn't take into account the "cost of living" index for those stats which makes them somewhat useless. There are other issues, but the fundamental premise is wrong; based on income.

Looked at another way, all those in the US making $50,000/year are not "out of poverty." Many are probably up to their eyeballs in debt.

It is nonsence to hold "race" responsible for crime rates. First of all, causation cannot be firmly established even for socioeconomic status and that seems to make some "causal" sense, e.g., if you are poor and powerless (excluded from the "legitimate opportunity structure") you will be more desperate and likely to enter (an "illegitmate opportunity structure) into the crime world.
But "race" is a false concept. The American Anthropological Association no longer considers it be an actuality, and has petitioned the federal government to exclude it as a consideration in official census work.
Genes exist, of course, but they do not naturally organize into biological units or races. If anything "social race" exists. This consists of the ideas people have about categories of the human species--what were once considered subspecies units. These pseudo biological ideas operate in everyday life as quasi-political tools. As such, "social race" is of interest to sociologists but not to physical anthropologists.