How statistics (however good or bad) can be used to form an arguement

So I was wandering around on the internet, reading through articles about Big Dairy vs. Raw Milk after I a CF post about this topic had caught my eye. So then it was off to google. The first article I found had an interesting statistic, followed by an interesting statement about the statistic. I will summarize to the best of my ability:

Statistic: "Official studies showed that last year in the US raw milk caused 200 food borne illnesses, while that number pales in comparison to the total 76 million food borne illnesses in the US. That's .000002% of cases."

Statement: "Big industry and government continue to ignore these statistics and keep the blinders on".

Really? I am thinking there is some missing information that needs to be there in order to understand the relevance of the 200 food borne illnesses, 76 million total illnesses and the percentage the 200 comprise when compared to the subject of what they are implying: the healthiness of raw milk. That is because the case against raw milk is that it is unhealthy, or so I have been lead to believe.

Of course this is just some article pulled from a random google search (I just typed in Big Dairy vs. Raw milk) so it is not like it is a major newprint article.

Are there other examples of poorly or deceptively used statistics? Is it a product of bad journalists (notice I did not say journalism), incomplete access to information, motive, or something else? Another reason I bring this up is the Daily Show with Jon Stewart ran a short clip of a compilation of cable TV news anchors and pundits quoting various "polls" about the same subject (debt ceiling if I remember correctly). It seemed laughable how the "percentages of people who believe...x" changed and how the wording of what people believed changed from poll to poll, but all polls seemed to have the exact same statement.