Page 7:
“When interpreting food security statistics in this report (except for appendix
D), it is important to keep in mind that households were classified as having
low or very low food security if they experienced the condition at any time
during the previous 12 months. The prevalence of these conditions on any
given day is far below the corresponding annual prevalence. For example,
the prevalence of very low food security on an average day during the 30-day
period prior to the December 2008 survey is estimated to have been between
0.9 and 1.2 percent of households.”

First, the WP article says “16.6 percent of Americans experienced “food insecurity” — the feeling of not knowing where one’s next meal is coming from — during 2009.” Note the “during 2009,” that means that these 16.6 percent were food insecure at any time during the year. To figure out how many people were food insecure at any ONE time, you need to know how long people were food insecure. In the survey itself, it says “Per the USDA, households experiencing food insecurity experience this condition in, on average, in seven months of the year.” ( http://feedingamerica.org/our-network/the-studies/map-the-gap/map-methodology.aspx ) So, the closest you can come to a point-in-time estimate with this data is to multiply 16.6 by 7/12ths yielding, 9.68 . Now, I’m sure there are seasonal factors that would make this vary within the year, but the study doesn’t go in to them.

So we now have that 9.68% of America is experiencing food insecurity. You should know that this study wasn’t a survey, it didn’t go out and interview people, it just used updated poverty, unemployment, and racial statistics to generate estimates of the USDA’s existing food-insecurity data by state and county. So to find out what this means, you have to look at the USDA’s definition: “Food Insecurity: USDA’s measure of lack of access, at times, to enough food for an active, healthy life for all household members; limited or uncertain availability of nutritionally adequate foods.” ( http://feedingamerica.org/our-network/the-studies/map-the-gap/~/media/Files/research/ExecutiveSummary_Final.ashx?.pdf ). I have no idea where the Post got its description of “food insecure,” that isn’t even close to what the measure means.

Food insecurity is not a politically-correct synonym for hunger, it’s trying to be a broader measure of those who aren’t getting ADEQUATE food over some length of time. You are counted as food insecure now if, for example, you don’t have enough to buy groceries before your next paycheck; that doesn’t mean that your cupboard is empty this instant. To make say how many Americans “don’t know where their next meal will come from” you’d need data on how often food insecurity coincides with hunger. It could be 25% of the time (which would suggest 2.4% “don’t know where their next meal will come from”), it could be 10% of the time (which would suggest 1%), but by the very definition, we know it’s not 100%. The study doesn’t touch this. Read the article, one of the examples is an unemployed IT administrator who went to the food bank. Is he food-insecure? You bet. Does he know where his next meal is coming from? Yes, he has steaks in the freezer that his friend gave him.

Look, it should go without saying that food insecurity is bad, and that 9.68% of Americans is a lot. It’s an important story, one that someone should report. Blowing the numbers hugely out of proportion doesn’t help the case, it makes people (including myself) think “that can’t possibly be true,” and dismiss it. They’d be right to do so; those inflated figures aren’t true. The best way to win the argument is to use figures that are correct.