'Happy Meal' Legislation Cites Faulty Data
One thing is for certain, however. Alarming statistics cited in the proposed law's legislative digest -- "According to 2003-2004 data, San Franciscans consume over one-third of their food, as measured by weight, at fast food and pizza restaurants" -- are demonstrably false.
That citation comes from a 2008 study called "Think Globally -- Eat Locally" undertaken by the American Farmland Trust. As its title implies, this study is largely about whether San Francisco can sustain itself on food produced within 100 miles of the city. It has exceedingly little to do with pizza, fast food, or Happy Meal toys.
Yet it includes a citation -- on page 24 for those who click on the link above -- noting that 56 percent of San Franciscans' food budgets are spent on home-prepared meals. In the very next paragraph, the study notes, "When food intake is measured by weight rather than dollars, fast food and pizza restaurants claim the largest share of total food consumed - a whopping 35 percent."
This is problematic -- as the source for this claim is the 2003-04 United States Department of Agriculture "What We Eat in America" survey. This is a national study of roughly 5,000 Americans' diets reported over two days -- not a survey of San Franciscans' eating habits, as the juxtaposition implies.
It is simply not accurate to use this national data to claim one-third of the food "San Franciscans" consume by weight is pizza and fast food. That the American Farmland Trust would so casually jump between hyperlocal data and incredibly broad national data is odd -- and has led to Supervisor Eric Mar's staff making an understandable mistake when drafting this proposal.
But is The American Farmland Trust's claim that the 2003 national survey shows 35 percent of the food Americans eat by weight is pizza and fast food correct? Even the research leader in charge of the USDA's Food Survey Research Group couldn't tell us that.
This is because What We Eat in America's wondrous and myriad findings are all but impenetrable. In order to determine, by weight, what the 5,000-odd respondents ate in 2003-04, first you must download the data from Day One and then the data from Day Two. Then you've got to scour the USDA site -- we couldn't find it -- to locate the weights assigned to individual food items ranging from various types of pizza, burgers, fillets o' fish, etc. This is beyond onerous.
So, here's what you need to know: Is the 35 percent figure for pizza and fast food realistic? Alanna Moshfegh, the aforementioned USDA research leader, said it could be. Does it mean anything? Maybe not.
"I don't know how come they focused on that number," she said of food weight. "Liquids are very heavy. Cotton candy is very light." You could eat cotton candy from dawn until dusk and still only consume a pound or so of food -- but no one would argue that this is a reasonable thing to do.
The USDA, in fact, doesn't list "food weight" among the pertinent data for its What We Eat in America findings. It does publish "nutrient intakes from food," "Percent of Energy From Protein, Carbohydrate, Fat, and Alcohol," and how nutritious meals and snacks are, and where they are eaten.
In the end, what light does this far-fetched data shine on the wisdom of regulating Happy Meals in San Francisco? Not much. And what chance does Mar's proposal have of surviving a promised mayoral veto when Supervisor Bevan Dufty is running for mayor and almost certainly won't touch this with a 10-foot Happy Meal toy? Even less.
Follow us on Twitter at @TheSnitchSF and @SFWeekly