First of all, we need to distinguish between different cancers, because the risk factors for one are not the same as the risk factors for another. 'Cancer' isn't just one condition, but a whole group of conditions. Some risk factors have an effect across many of those conditions (alcohol would be an example as consumption increases the risk of lots of different cancers from liver to breast cancer), some are more specific (asbestos might be an example, since it tends to be linked more specifically to lung cancer).
Then we need to put the evidence in context, which is not about certainties but about risks. Some people will eat lots and lots of red meat and never get colorectal cancer, just the same as there are those who have smoked 40 a day for 40 years and never got lung cancer. This does not mean, however, that there is no link statistically between regular consumption of large portions of red meat and colorectal cancer, or large consumption of cigarettes and lung cancer. However, to say that something is a risk factor also does not mean that it is the sole possible cause of a disease. There are people with lung cancer who never smoked at all, and people with colorectal cancer who have not consumed red meat in years.
I think one important thing missing here is the difference between absolute and relative risk. I muddied the waters between the two in my example above, so I'm going to try to straighten that out a bit here. As far as I understand it, all studies try to do is to measure the effects of one variable on another in an 'exposed' group against a 'control' group. So you might measure the colorectal cancer incidence in a group who regularly consumption of large portions of red meat against colorectal cancer incidence in a control group who have a lesser/no degree of exposure. Once you've done all the followup and all the sums, you can come up with a rough measure of the 'relative risk', which is a ratio.
So, in our case, it would be a ratio that showed the probability of colorectal cancer in the high red meat consumption group compared to the probability in a group with lower/no consumption. A relative risk of 1 means there is no difference between the groups. Greater than 1 means it's more likely in the exposed group. (You have to do another bunch of calculations to work out whether the difference is 'significant' or not, because small differences can just be an effect of the sampling you're using).
If you look at the data above, there are relative risk (RR) values presented throughout, and some indication of significance/confidence intervals. Hopefully these are helpful in allowing people to make an informed decision!