Obviously people with the ability to drive anywhere they want aren’t cut off, in any town large enough to have any real grocery stores. Duh. The entire and only point of the concept is with respect to people who can’t jump in the car any time.
Then it still appears to be a solution in search of a problem, for the most part. In 2010, in the depths of the recent recession, somewhere north of two thirds of poor people had access to one or more cars or trucks (cite). People in Manhattan who have to ride the subway to the grocery store also does not strike me as incredible hardship, requiring societal intervention.
Regards,
Shodan
However, a food desert map is also probably largely correlated with “walkable” communities in general. If I had restaurants and grocery stores a block or so from me instead of 1/2 to a mile, I could get some of my exercise in in a pleasant walk after work or a weekend, and save gas and car mileage at the same time.
That’s true. I was too reductionist there; it is relevant to non-poor and/or non-carless people. I’d have appreciated better grocery options within walking distance (which is long, to me) of some places I lived in the city, even when I did have a car there.
This is another caveat to the map. How many of those low-level food deserts are actual more “barren” than they appear because there are few/no crosswalks and sidewalks?
My grandmother lives four miles from the closet grocery store. She’s voluntarily given up her car keys, but despite being 89 years old, she’s still ambulatory enough to walk a little ways. Just not four miles. Especially four surburban miles. She ends up having to rely on my aunt, since there isn’t any public transit in her area. That’s hardship on my aunt and on my grandmother, who has always valued her independence. It also makes me worried that she’s so dependent on my crazy-ass aunt for something as basic as food. So I think she lives in a food desert. But I don’t think she would be if she had access to public transit. It seems to me that the transportation infrastructure is a vital component to any discussion about “food deserts”, so it is unfortunate that it is not included in the USDA atlas.
To be fair, the ridiculous paucity of pedestrian facilities in most of America is its own problem, and arguably more serious.
I am curious. When you right-clicked on the area, what information did the pop-up window provide?
I am half a mile from a grocery store yet live in a de facto food desert.
Cheese? Crap-ass store-brand. No other choices. Ditto frozen vegetables.
Chicken breasts? Most of the time no. The rare occurrance the breasts are there they are so frozen you can’t use them that night.
Deli? Meat mortuary.
Markup? At least 10% more than any other store. Usually more like 20%.
Next closest grocery is over 8 miles away.
it just says “is a low income tract”, is a low access tract".
I don’t think you have a good grasp of the economics of grocery stores.
Most grocers want customers who will buy lots of the high-margin items: more expensive cuts of meat, prepared foods/hot case, fresh-baked bread, floral, seafood, and so forth. Most poor people, however, aren’t buying these items: notwithstanding urban legend, they buy a lot more of the bulk packs of cheap hamburger and very little lobster and ribeye. The margin on generic products, canned goods and pantry staples is pathetic, though, and only a few grocers with very specific marketing plans (e.g., Aldi) have found it worthwhile to specialize in such goods.
The big chain retailers can’t really get away with having very different pricing regimens in poor neighborhoods (it’s a PR nightmare), and they can’t make money just by selling canned vegetables, generic cereal, and 5-lb logs of cheap hamburger (what are loss-leaders in richer places), so they don’t locate their stores in such locales. It’s not that the inhabitants don’t want the stores; it’s that the stores don’t want only customers who don’t buy the items on which the stores make most of their profits.
So poor neighborhoods, therefore, the stores that stay in business are the ones that don’t have all of the fancy departments. Without the fancy departments, though, the stores have to make all of their profits by jacking up the prices on the goods they do sell, so you end up with high prices and limited selections.
The maps are using census tracts, which are preexisting defined areas for which income statistics (and other information such as household composition and demographics) are readily available. If the USDA wants to define its own areas with more regular boundaries, then they lose easy access to all of the population data, which sort of defeats the purpose.
Sure, but in the areas that I was looking at- specifically the ones near my home (NE Dallas, east of the I-635/US-75 interchange), it’s hideously inaccurate. In one case, the tract is literally bounded by a huge, modern grocery store, and only the farthest corner of the tract (the part tucked up near the freeway interchange) isn’t within a mile of it. The other ones, more curiously, are in reality, overlapped by no less than 2, and in some areas, 3 grocery stores within a mile.
The issue isn’t so much with the use of the census tracts as such, it’s the fact that the way that they’re using them means that the results are very dependent on the geometry of the individual census tracts.
Search ‘9299 Markville Rd. Dallas, TX 75243’ for a point midway between the two sets of tracts I’m talking about.
On the left-side tract, there’s a huge Kroger grocery store on the SE corner of Greenville Ave. and Forest Ln. On the right side tracts, there’s an Aldi on the corner of Audelia @ Forest, an Asian supermarket at Audelia and Walnut, a Fiesta Mart and a Wal-Mart at Buckingham @ Plano Rd, another couple ethnic supermarkets at Plano @ Walnut, and a monstrous Wal-Mart Supercenter at Abrams @ Forest. The tracts are literally bounded by grocery stores in all directions, and yet somehow, that’s a food desert? I suppose if you drew 1 mile radius circles from each of those, you could possibly find some tiny portion of each of those tracts that’s more than a mile from each store, but that’s misleading as hell. The area’s awash in grocery stores- if there’s an area more than a mile from one, it’s because it’s a “commercial desert” as chappachula points out. In the left-side tract, the parts that aren’t within a mile of the nearest grocery store are generally tucked right up against freeway, parkland, or a huge construction staging area for the I-635 construction project that’s going on. There isn’t squat in terms of ANY commercial development in that particular corner of the freeway interchange.
That’s what I’m talking about- nobody in their right mind would think that NE Dallas/SW Garland/S. Richardson is a food desert, not even for ethnic or specialty foods. Yet, due to their sketchy data visualization, the USDA seems to think so.
Trying to frighten people who live in undustrial parks? Or in large drainage channels? Most people looking at that kind of data are going to see that they live exactly where they thought they lived, which isn’t very frightening at all.
Real-estate development is the most important thing my local and regional govenments do. It is, I think, the largest and most influential source of campaign funding, influencing all the major infrastructure decisions like roads, trains, tunnels and towers. My local city council built and sold off a shopping centre in a new area, to boost development and increase real-estate values.
My wife studied urban geography. My cousin-in-law makes a living doing this, employed by a major urban developer. My local real-estate advertiser publishes that kind of information to get people to pick up their advertising. It’s both interesting to a lot of people, and hugely financially important to a small number of rich people. Enjoy.
Huh? Most people who look at that data are not interested in seeing where they live.
Most people who look at that data have an agenda.
A simple google for “food deserts” brings up lots of websites with names like “food empowerment”, or warning you about “this sobering map” showing all the terrible food deserts in the country.
The map is absurd, and totally misleading (see my post above).
Yet it is an official document from the US government, so naturally many people will, in good faith, rely on it.
And that is not only absurd…it is dangerous. Public policy makers are relying on bad data.
The green areas are “food deserts”? Then yes, this is a flawed map.
I checked my neighborhood - Mesa, AZ. There is a huge “desert” that covers the airport and Boeing, and a golf course. But even with that, across the street (literally, corner of Greenfield and McKellips) from the “green zone” is a Wal*mart, and there is a Fry’s super market two miles east.
And yet, there is a true “desert” by their definition that isn’t identified. There aren’t any businesses immediately to the west of the zone for about two miles (purely residential, and orange groves) and yet it isn’t a “desert”.
I actually bothered to look at the documentation, and it explains the various “mysteries”.
They are defining a “food desert” as a tract where at least 100 households do not have a car in low income tracts which are a specified distance from a grocery store, or 500 households that are more than 20 miles from a grocery store. You are able choose the distance from a grocery store to display (.05, 1 or 10 miles— so rural areas can look at distances appropriate for driving). For determining distance, they break the housing down into quarter mile blocks.
Information is displayed in terms of census tracts because that is how the demographic information (income, car access, etc.) is collected. In other words, they could get into more detail with the distances, but they can’t do that and also include information such as income, etc.
We don’t collect (or at least, we don’t display) income and car access on a household level.
Supermarkets are determined from the 2012 list of stores accepting SNAP which have all common grocery departments (produce, dairy, meat, packaged foods, frozen foods, etc.) I assume that is the most recent complete data available.
Anyone who clicks the link, glances at the map and makes a major policy decision would definitely be making a mistake. But once you actually know what the map is displaying, I don’t think it’s horribly wrong, and it’s the best that you are likely to get with existing data. You just have to know what the actual information being displayed is.
Not picking on you - just using your quotes to illustrate the map’s problems.
In my example, the green desert of Falcon field and the Boeing plant (and the golf course) there are ZERO people living there. Three square miles with absolutely no one living in them. Zero people with zero income with or without zero cars per person.
And no matter the size of the grid used for mapping, having a situation where there is supermarket kitty corner or directly across a street from a “desert” renders the definition useless.
The green areas don’t represent “food deserts” per se. They represent areas that likely contain food deserts in them, as defined as a low income, low car ownership area with more than 100 households further than the selected distance from food. Somewhere within that green block are 100 households who likely don’t have good access to food. I would hope this definition was ground truthed in some way.
It would be more helpful to get down to a more granular level to pinpoint exactly where the affected houses are on a block by block basis, but unfortunately the data for that just doesn’t exist. We can’t get more accurate than census blocks.
It’s like when you look at, say, a map of the world where, say, countries with malaria are shaded. Within those countries, some areas have malaria and some don’t. But we are visualizing it on a higher scale and just identifying countries with the problem.
It’s not a useful map for laypeople, but it can help a city or state get an idea of where to look more closely.
Public health and planning folks deal with maps like this all the time. It’s not cost effective to do the kind of surveys that would make for the data that you want, so sometimes you have to cobble together a map using existing public data that was likely collected for another purpose, and may not be perfectly tailored to your needs. As long as policy makers understand the limitations of that data, that’s fine.
A better indicator about how “troubled” an area is than just whether or not there is a grocery store close is to couple that with a count of how many bars there are in the same area. There are several towns in this area that have no grocery store but that do have multiple bars. Most of the bars can legitimately be described as nuisance bars. None of these towns are likely to be anybody’s first choice of a place to live. OTOH, there are also areas that are upscalish (for here anyway) have no nearby grocery stores but no nuisance bars either. They aren’t bad places at all.
Yeah, after looking at the map I really can’t take worries about the dangers of ‘food deserts’ seriously. Near me there’s a big marked ‘food desert’ right across the street from a grocery store. I mean if you start in the neighborhood marked as a ‘food desert’, go down neighborhood streets to the edge of the food desert, then cross a 2 lane road with lights that borders the ‘food desert’, you’re in a food lion parking lot. It seems like a good term for political scaremongering, but doesn’t seem to have anything to do with food nearby.