Food Insecurity:
A persistent problem in Southeast Michigan

Food insecurity, defined by Healthy People 2030 as “…a household-level economic and social condition of limited or uncertain access to adequate food,” has been an ongoing reality in Southeast Michigan.  Since its founding in 1990, the Forgotten Harvest Food Bank has played a significant role in reducing food insecurity and hunger in the Southeast Michigan counties of Wayne, Oakland, and Macomb through more than 100 agencies (food pantries).  Recognizing the constant and growing need for food in the area, as well as its commitment to helping to meet this need, in 2017 Forgotten Harvest undertook a major assessment of its operations.  It was in essence wanting to align supply (its pantries) with the demand (need).

A Challenge:
Measuring the extent and location of food insecurity

A major challenge Forgotten Harvest faced in this effort was how to measure demand; more specifically how to quantify demand so that the organization could evaluate whether its existing and future pantries were optimally located to meet residents’ needs.  That’s where Data Driven Detroit (D3) got involved.  D3’s approach was to identify areas (census tracts) that were likely to be food insecure.  To do that D3 drew on a methodology used by the U.S. Bureau of the Census to identify areas that were hard-to-count in the Census.  A central feature of the methodology is the development of an index composed of a limited set of variables relevant for a specific factor:  In this case, the estimated number of residents either experiencing or at risk of experiencing food insufficiency in an area.  The Food Insecurity Index was calculated for each of the 1221 census tracts and later, ZIP codes, in the tri-county area and mapped as shown below.*

Map showing food insecurity scores by census tract for the Detroit tri-county area of Oakland, Macomb, and Wayne Counties

Note: The Food Insecurity Index measures populations that have a likelihood of experiencing food insecurity, and is comprised of six indicators: Households Experiencing a Housing Burden, Lower-Income Individuals who Moved in the Past 12 Months, Households with No Vehicle Available, Residents 25+ without a High School Diploma, % of Land Area over 1/4 mile from a Bus Stop, and % of Families with Children that only have 1 Parent. Index values can range from 6 (fewest potentially food-insecure people) to 66 (greatest number of potentially food-insecure people).

How Forgotten Harvest used the map

After the analysis was completed, the map gave Forgotten Harvest an immediate overview of the areas that might have higher need in the tri-county area. This geographical representation highlighted, for example, clusters of potential food insufficiency as well as isolated areas of food insufficiency. With the addition of physical markers depicting the locations of pantries, Forgotten Harvest had a starting point for wide-ranging discussions of the complexities and factors, both quantitative and nonquantitative, affecting supply and demand related to food insecurity.   

Expanding quantitative and qualitative actors affecting demand

Recognizing that demand was in part a function of the population size of a census tract Forgotten Harvest added a population weight to the calculation of the score.  For example, when comparing census tracts with comparable Index scores, the organization might give more weight to the ones with larger populations.        

Incorporating qualitative factors influencing demand was also important for giving a fuller, more nuanced picture of demand and the forces affecting it.  The changes wrought by the pandemic provide a stark example of this. When people lost their jobs many could not continue to pay for housing and had to move in with relatives or friends.  This movement of households could affect the quantity and types of food needed in an area.  While such nonquantitative information would not be reflected in the Index score, electronic notes could be attached to the map to bring it to the attention of the reader.

Markers on the map of supply

Supply in this context refers to the numbers and locations of food pantries.  Forgotten Harvest used the base map to test various scenarios regarding placement of pantries.  For example, if a criterion for pantry placement is that the service areas of pantries be at least one mile apart, one could draw circles of one-mile radius around all pantries and make visible the areas of overlap.  Forgotten Harvest could then identify areas that weren’t covered by these radii and also scored highly on the food insecurity index to prioritize for placing new pantries.  Of course, there may be good reasons for having overlapping service areas if, for example, pantries near each other serve different subpopulations with distinctive food needs or their days and hours of operation complement each other.


In pursuit of its goal of aligning its operations with the need for food in the tri-county area of Southeast Michigan, Forgotten Harvest took the quantitative geographic base of food insufficiency in the tri-county area provided by Data Driven Detroit and added immeasurably to its usefulness by bringing in an abundance of relevant information about food need and supply. In essence, D3 provided the framework that was then fleshed out by Forgotten Harvest’s deep knowledge and expertise and used to inform their actions and programming.

The 2017 census tract level Food Insufficiency Index drew five of its six component measures from the 2014 five-year American Community Survey (ACS).  A recent analysis** updated the original work, calculating the Food Insufficiency Index for ZIP Codes as opposed to census tracts (to allow for comparative analysis with ZIP-code level health outcomes data that they had received from Henry Ford Health System), and drawing the measures from the 2021 five-year ACS. It also limited the analysis to Detroit and rescored the 2014 data for ZIP codes. The authors of the updated study concluded that “[t]he differences gleaned between the 2014 and 2021 food insecurity indexes suggest that food insecurity vulnerability has not changed on average between both time points, but that vulnerability is increasing and declining at smaller scales in opposite ways across the city.”

*The original blog describing construction of the index and the source of this map is found at


Read an update to the Food Security Index by a couple PhD candidates at Wayne State University, Brenna Friday and David Criss

Do you need data to to enhance your project? AskD3 for free to get started!

Copyright © 2022 Data Driven Detroit. All Rights Reserved.