Using American Community Survey to Understand Your Community

A Guide for Extension Professionals

By: Teja Pristavec and Morgan Stockham Photo from: American Community Survey data viewed on data.census.gov

A T T R I B U T I ON

e-pub: 978-1-955687-01-0

Publish Date: 08/03/21

Citations for this publication may be made using the following:

Pristavec, T., & Stockham, M. (2021). Using American Community Survey to Understand Your Community (1 st ed). Extension Foundation: National Extension Library . ISBN: 978-1-955687-01-0.

Producer: Ashley S. Griffin

Peer Review Coordinator: Heather Martin

Technical Implementer: Ashley S. Griffin

Welcome to the Using American Community Survey to Understand Your Community, a resource created for the Cooperative Extension Service and published by the Extension Foundation. We welcome feedback and suggested resources for this publication, which could be included in any subsequent versions. This work is supported by New Technologies for Agriculture Extension grant no. 2020-41595-30123 from the USDA National Institute of Food and Agriculture. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture.

For more information please contact:

Extension Foundation c/o Bryan Cave LLP One Kansas City Place

1200 Main Street, Suite 3800 Kansas City, MO 64105-2122 https://impact.extension.org/

T A B L E O F CON T E N T S

Attribution .............................................................................................................................................. 2

Table of Contents..................................................................................................................................... 3

Meet the Authors .................................................................................................................................... 5 Teja Pristavec...................................................................................................................................................................... 5 Morgan Stockham............................................................................................................................................................... 5

Acknowledgments ................................................................................................................................... 5

Introduction ................................................................................................................................... 6

Overview................................................................................................................................................. 6 Understanding the American Community Survey .............................................................................................................. 7 Case Study 1: Where in My County Could Children be at Risk for Being Left out of Remote Education? ................................................................................................................................... 10

Step 1: Search for your keyword............................................................................................................. 11

Step 2: Select your search results reporting format. ................................................................................ 12

Step 3: Search for your area. .................................................................................................................. 13

Step 4: Find your data table. .................................................................................................................. 15

Step 5: Find your variable....................................................................................................................... 17

Step 6: Customize your map. .................................................................................................................. 19

Step 7: View data and explore more. ...................................................................................................... 22

Case Study 2: Where in my County Could Residents be at Risk for Eviction, and How Does that Compare to Surrounding Areas? ................................................................................................... 24

Step 1: Search for your area and data table ............................................................................................ 25

Step 2: Find and understand the relevant housing indicators .................................................................. 26

Step 3: Compare your area to another geography ................................................................................... 29

Case Study 3: Who in My County is at Risk of Food Insecurity and How has That Changed Over Time?........................................................................................................................................... 32

Step 1: Search for your keyword............................................................................................................. 33

Step 2: Select the data table................................................................................................................... 34

Step 3: Review table contents and select the geographic area. ................................................................ 37

Step 4: Understand the data. ................................................................................................................. 41

Step 5: Explore data over time. .............................................................................................................. 43

Additional Tools ........................................................................................................................... 44 1. Narrative Profiles .......................................................................................................................................................... 44 2. Census Business Builder................................................................................................................................................ 45 3. OnTheMap for Emergency Management ..................................................................................................................... 45 4. My Congressional District ............................................................................................................................................. 46 5. My Tribal Area............................................................................................................................................................... 46 6. Census Flows Mapper ................................................................................................................................................... 46 7. QuickFacts ..................................................................................................................................................................... 47 Glossary....................................................................................................................................... 48 ACS Comparison Profiles................................................................................................................................................... 48 ACS Detailed Tables .......................................................................................................................................................... 48 ACS Subject Tables ............................................................................................................................................................ 49 American Community Survey (ACS).................................................................................................................................. 49 Census Block Groups......................................................................................................................................................... 50 Census Tracts .................................................................................................................................................................... 50 Margin of Error ................................................................................................................................................................. 51 U.S. Census Bureau ........................................................................................................................................................... 51

M E E T TH E AU THO R S

Teja Pr istavec

Teja Pristavec is a sociologist and a Research Assistant Professor of statistical sciences in the Social and Decision Analytics Division (SDAD), Biocomplexity Institute, University of Virginia. Her research interests include spatial analysis and GIS, quantitative methods, administrative data, health, and inequality. Her work has been published in the Journals of Gerontology and Geriatrics, Journal of Pain and Symptom Management, and Food, Culture, & Society. At SDAD she collaborates with an interdisciplinary research group using quantitative methods to develop evidence-based research and inform effective decision-making in government and industry. Teja has recently accepted a position as a Data and Research Specialist at the Coleridge Initiative.

Morgan Stockham

Morgan is a PhD student in Applied Microeconomics at Claremont Graduate University and visiting instructor at Pitzer College. Her academic research interests include crime and law economics, as well as community- based data science.

A C K NOWL E D GM E N T S

The authors would like to thank Mike Lambur of Virginia Tech, Virginia Cooperative Extension Service and the New Technologies in Agriculture Extension (NTAE) team for their support and thoughtful steering of this project.

Tira Adelman , Administrative Support & Reporting Karl Bradley , Leadership Beverly Coberly , Administrative Support Ashley Griffin , Publications Rose Hayden-Smith , Digital Engagement Chuck Hibberd , Catalyst

Megan Hirschman , Partnership Molly Immendorf , Professional Development

Akashi Kaul , Evaluation Rick Klemme , Catalyst

Heather Martin , Peer Review Coordinator Aaron Weibe , Marketing/Communications

Introduction

Have you ever wondered how many residents in the community you serve have access to a computer and internet and whether or not you could reach them electronically? Perhaps you are interested in better tailoring the distribution of a pamphlet about farmers’ markets that accept food stamps and want to know which areas of your county have a high proportion of residents who are eligible for such benefits. Maybe you are preparing a report, assessment, or a budget justification and want to include some numbers to illustrate a county’s need and support your case. In all these scenarios, data can help you better understand the community you serve and enhance your work as an Extension professional. This publication will show you how to use data from the U.S. Census B ureau’s American Community Survey (ACS) to find and interpret valuable information about your community that will help you make data-driven decisions. We will use step-by-step case studies that do not require any prior knowledge of the survey or technical skills to walk you through applied examples that you can adapt to your needs. The overview below explains our steps in more detail. We are excited to have you along for the ride!

O V E R V I EW

The first section of this publication introduces you to the ACS, the Census Bureau’s most extensive survey. The survey provides information on population, social and demographic characteristics, education and income, housing, social program participation, and other topics. Our case studies use the ACS to illustrate what data you have at your disposal to better understand your community and how to retrieve it using the Census Bureau data portal. The section provides a quick overview of key concepts about geographic areas and timeframes related to ACS data that may be helpful to know before walking through our tutorials. The main section uses three case studies to highlight the information available to you in ACS and demonstrate how to use multiple Census Bureau data portal features to retrieve this information. All three case studies use applied examples with step-by-step instructions that you can follow and adapt to find information relevant to your community. The first case study — “Where in My County Could Children be at Risk f or Being Left out of Remote Education?”— guides you through finding information about computer and internet subscription availability in households with children, showing you how to find detailed tables about any ACS topic and how to visualize results on a map of your area. The second case study — “ Where in my County Could Residents be at Risk for Eviction, and How Does that Compare to Surrounding Areas?”— highlights the breadth of housing information available in ACS. It introduces you to Data Profiles tables (a helpful tool that synthesizes information across multiple data points), shows you how to conduct a comparison between geographic areas, and demonstrates how to read tables and maps.

The third case study — “ Who in My County is at Risk of Food Insecurity and How has That Changed Over Time?”—uses information on the food stamp program to showcase ACS’s Subject Tables and guide you

through customizing tables. It uses ACS data to investigate change over time and shows you how to download the data you retrieve.

Because this publication serves only as an introduction and not an exhaustive guide to using ACS, its final section provides you with tools that you can use to explore these data further on your own. The Census Bureau and other agencies and organizations make many datasets publicly available and build tools that allow you to interact with the data without any prior training or technical skills. At the end of this publication, we provide a list and short descriptions of Census Bureau data tools that may be relevant to your work, as well as a glossary of terms we use in our case studies for your reference.

Understanding the Amer ican Communi ty Survey

We hope that this publication serves as a practical introduction to using data — ACS in particular — to better understand the community you serve and gives you the skills you need to incorporate it into your Extension work. Let’s get started!

American Community Survey

The U.S. Census Bur eau’s American Community Survey (ACS) is an ongoing, nationally representative household survey. Administered in all fifty states, the District of Columbia, and Puerto Rico, it is the largest survey the agency conducts; approximately 3.5 million households complete it every year via mail, phone, internet, or in-person interviews. ACS contains information on demographic characteristics like gender, age, race, ethnicity, family and household composition, educational attainment, employment and income, migration, ancestry, citizenship, disability, insurance, social program participation, housing, and more. It is a rich data resource that helps local and state governments, federal agencies, non-profit institutions, and other organizations make informed decisions about funding and planning. It also allows them to learn about their constituents and improve their communities.

Learning to use data to better understand the community you serve is also the goal of this practical guide. Given the ACS’s wealth of information on individuals’, families’, and

household s’ social and economic characteristics, we will use the survey in our examples and show you how to navigate the Census Bureau’s portal to find data that can inform your decisions. The two sections below provide background on the ACS geographic areas and time frames that we will cover in our examples.

ACS Geographic Areas

ACS data are available for multiple geographic areas. While we can use it to find information about the entire nation, a state, or a county as a whole, its granularity also allows us to retrieve data for geographies smaller than a county. For Extension professionals and other stakeholders working with communities, this is an important feature that helps draw insights about social and economic conditions in local areas. For example, the ACS not only would tell you the unemployment rate for the state of Virginia, but it will also tell you the unemployment rate in a given county, like Patrick County, and even for smaller, neighborhood-level areas within Patrick County.

Our guide focuses on three geographies that will be most useful for your work and are the smallest geographies available through the Census Bureau data portal: counties, census tracts, and census block groups. Before we delve into our step-by-step data case studies and tutorials, let’s review these areas and describe how they are related to each other. To learn more about the geographies that the ACS covers, see the complete listing of Census Bureau geographic hierarchies on the agency’s Geographic Program website. As you know, counties are legal units that make up a state. In some states, the equivalent units are called parishes, boroughs, or municipalities, and some states have independent cities in addition to counties. The ACS refers to this level of geography as “county or county - equivalent.” The Census Bureau data portal provides access to information on two geographic areas that are smaller than a county. Census tracts are statistical subdivisions that are smaller than counties but larger than census block groups. Their populations range in size from 1,200 to 8,000, with an average of about 4,000 residents. Breaking census tract areas down further, census block groups are statistical subdivisions of tracts that describe areas of 600 to 3,000 individuals. They are smaller than tracts but larger than individual census blocks. We will be using county, census tract, and block group data in our case studies, as block data is unavailable through the Census Bureau portal.

Figure 1: County, census tract, and census block group maps of Patrick County, Virginia.

ACS Time Frames

The ACS started in 2005 and is an ongoing survey, releasing products every year as single-year estimates and five-year estimates. ACS single-year estimates are timely, but they are available only for areas with 65,000 residents or more. This is because the ACS is a survey, which means that only a sample and not all Americans respond to the ACS questionnaire. The Census Bureau then uses statistical methods to provide us with representative data about the entire population. In a single year, the Census Bureau collects enough responses that represent large areas well. Still, it cannot collect enough information in each small geography for them to be representative of everyone who lives in that small geography. However, when the Census Bureau combines data over five years, the sample size for these small areas becomes larger. The information is reliable enough that it describes well the population of counties, census tracts, and census block groups. Thus, to get our data on the small geographies we discussed, we will be using the five-year estimates in this tutorial.

Case Study 1: Where in My County Could Children be at Risk for Being Left out of Remote Education? The COVID- 19 pandemic significantly disrupted elementary, high school, and university students’ ability to receive their education in a traditional classroom setting. With social distancing requirements in place, to keep providing instruction, educational institutions had to quickly pivot to delivering their content online and later through hybrid in-person and virtual models. However, while remote education may provide continuity and minimize the impact of disruptions on schooling, not all students can participate in online learning equally. Without access to a computing device, like a laptop or smartphone, and a reliable and fast internet connection, many vulnerable students across the country were at risk of being left out of remote education. Local governments, organizations, and schools responded to this challenge and began providing devices to students and installing wi-fi hotspots. As an Extension professional, you may have participated in similar efforts. Perhaps you partnered with university researchers and staff to conduct outreach and distribute computing devices or were asked to report on areas of need. You may have relied on the wealth of knowledge you already have about the community you serve to provide your partners with information. How could you have supported your response with data? Computers and the internet are one of the topics that the ACS covers. This case study will guide you through how to use ACS data to learn where in your county children may be unable to participate in remote education because they lack the tools to do so. We will walk step by step through how to answer this question with a map.

S T E P 1 : S E A R CH F O R YOU R K E YWO R D .

We will begin our search for data at the data.census.gov portal. This portal gives you the ability to search across the Census Bureau’s datasets, including ACS data. Because we are interested in finding areas in your county where school-aged children and youth may not be able to participate in remote education, we will use the simple search bar displayed in the center of the page to search for a keyword for relevant clues. For this example, let’s use “computer.” We can type the keyword into the search bar and click “SEARCH,” as Figure 2 shows.

Figure 2: Using the search bar.

S T E P 2 : S EL E C T YOU R S E A R CH R E S U L T S R E P O R T I NG F O RMA T .

Our search results will come in multiple forms: tables, maps, and pages. Because we are interested in identifying geographic areas in our county where children and youth are at risk, we will focus on maps. As Figure 3 shows, the top-left bar of our search results gives us an option to filter result types. We will click “maps” and explore tables in a later section.

Figure 3: Selecting maps to filter.

S T E P 3 : S E A R CH F O R YOU R A R E A .

To view a detailed map of your county, we must next filter our results by geography. The first step in doing so is selecting the “filter” option that now appears under our “maps” selection, as shown in Figure 4.

Figure 4: Selecting search result filters.

After clicking “filter,” a new window with filt er options appears, as displayed in Figure 5. On the left-hand side, the “browse filters” list includes filtering by topics, geography, years, surveys, and codes. Because we are interested in seeing results for a particular area, we select “geography.” This action opens additional options that allow us to provide more information about what kind of geography we would like to see on our map. Recall from our introductory section that the Census Bureau uses statistical geographies to provide data. Each county is divided into census tracts, and each census tract is further divided into census block groups. Census block groups are areas with populations of 600 to 3,000 and will give us a good level of detail. We select “block group” from the geography list. Finally, we navigate to our state and county of choice. The next filter option allows us to pick a state. For this example, we want to know more about remote education barriers in a Virginia county, so we select “Virginia.” You can follow along or select your state. Within your chosen state, we now see a list of counties, and for this example, we select “Patrick County, Virginia;” again, you may navigate to your county instead. As a final step, we click the box that confirms we would like to see a map for “all block groups within Patrick County, Virginia.” Once we are done with our selection, we click the “done” button on the top -right side of the filter window to collapse the filtering panel and return to our results.

Figure 5: Filtering search results to a limited geographic area.

S T E P 4 : F I N D YOU R D A T A T A B L E .

A map of our county now appears on the screen — we are getting closer to seeing the data. We initially searched for the keyword “computer.” This keyword appears in multiple ACS data tables, and we must now select the most suitable table for answering our question. The left-hand sidebar shows all relevant tables. We can scroll through them to find one that includes age, as we would like to see results that pertain to children and youth, computer access, and internet access. In thi s case, our list will contain the table “age by the presence of a computer and types of internet subscription in a household.” We select this table, shown in Figure 6. The information included under the table title tells us that the data in this table comes from ACS, is available for years 2017 through 2019, and has the unique identifier B28005.

Figure 6: Selecting a data table with relevant information.

Figure 7 shows the result of selecting our preferred table: the map of our county (in this example, Patrick County in Virginia) at the census block group level. The map shows the default row from this table or the total number of individuals living in a given census block group. Our next step is to change our variable selection and map variable from this table, which shows the total number of children and youth who may have trouble accessing online education.

Figure 7: Populating the area map with data.

S T E P 5 : F I N D YOU R V A R I A B L E .

To select our variable of interest, we click the drop-down menu that starts with the currently displayed variable name, “ Total:- Estimate”, on the navigation bar above our map. The resulting drop -down menu displays all table rows. You can explore all the options; in this example, we will map the number of individuals in a census block group within Patrick County who have access to a computer but do not have an internet subscription. See Figure 8 and scroll down to the row that refers to that inform ation, labeled “Total - Under 18 years-Has a computer-Without and internet subscription- Estimate.”

Figure 8: Selecting the variable displayed on the map.

Figure 9 shows our new map. The updated top bar title now informs us that this is a map of ACS data estimating the number of persons under eighteen living in households with a computer but without an internet subscription. The bottom-left legend tells us how map colors correspond to the estimated number of such persons in each census block group, with darker colors indicating a higher number of children and youth facing barriers to remote education. In this example, we can see that the eastern parts of Patrick County have no children and youth living in households with a computer but no internet access; conversely, census block groups in the west of the county have up to approximately eighty such persons. Our map is interactive and clicking on a census block group will show detailed information about it. For example, clicking on block group 1 — the block group with one of the darkest colors — shows us that forty-nine individuals under eighteen years old with a computer but no internet live in the block group. Our map can now tell us where to target our initiatives and efforts. Because we are mapping areas where school-aged children have no internet, we may think about providing them with free wireless internet access hotspots for their computers or reaching them via cell phone instead. We could also explore other ACS tables, or other

variables included in our table, to find areas where youth have no computers, areas where internet connection may be slow (for example, with youth only having satellite or dial-up internet connection), or where they have no computer access at all.

Figure 9: Reading the map.

S T E P 6 : CU S TOM I Z E YOU R MA P .

After we’ve arrived at the map displaying the data we need, we can click the “customize map” button on the top-right of the portal to change its look, as shown in Figure 10, followed by clicking the new gear icon that appears on the bottom-left of the screen, as shown in Figure 11.

Figure 10: Navigating to map customization options.

Figure 11: Navigating to map customization options.

The customization options that appear and are displayed in Figure 12 allow us to change the map color scale, the number of categories in our map legend, and the size of the categories that we map. In this example, we select a red color scheme, keep the number of data categories at five (experiment with increasing and decreasing the number of categories!), and change the classification type. Selecting “equal interval” provides category ranges of the same range.

Figure 12: Customizing the map.

S T E P 7 : V I EW D A T A AN D E X P L O R E MO R E .

Perhaps you are curious about the data you are mapping or wanted to know the exact number of children and youth under eighteen living in households with a computer but without internet. Click the “view table” button in the customization panel, shown in Figure 13, which will display the relevant information. The new window that appears lists the number of such persons for every block group in our Patrick County example.

Figure 13: Viewing the raw data.

Finally, the “notes” button, shown in Figure 14, provides technical details about our data, including data collection, links to documentation, a brief description of ACS survey procedures, and information about the questions asked on the survey to obtain our map estimates.

Figure 14: Learning more about the data.

Case Study 2: Where in my County Could Residents be at Risk for Eviction, and How Does that Compare to Surrounding Areas? The coronavirus pandemic heavily affected the financial well-being of people whose income was tied to in- person service. Many individuals who were laid off, furloughed, or faced other income disruptions, also held lower-paying jobs and rented rather than owned property. With little savings and loss of income, many of these persons were at risk of eviction. While the U.S. Centers for Disease Control issued an order to halt evictions, and some state governments passed legislation of their own to slow evictions, these orders are temporary. As eviction moratoria are lifted, some communities may face a greater risk of evictions than others. As an Extension professional, you may be concerned about hardship in your community. Assessing your community’s risk of evictions could help you provide your constituents with information on rent laws, resources to alleviate the rent burden, and information on social programs and temporary shelter. You may rely on your extensive knowledge and experience serving your community to assess eviction risks, but how could data help you enhance your decisions on outreach and resource management for evictions? The ACS has a wealth of data on renting, income, and other potential indicators for housing instability. This case study will guide you through how to use ACS data to learn where in your county residents may be more likely to face evictions given factors like renter status, low income, and unsteady employment. We will walk step by step through how to answer this question and look at how ACS can help compare the situation in your area with those in surrounding counties.

S T E P 1 : S E A R CH F O R YOU R A R E A A N D D A T A T A B L E

We start our search at the Data Profile subsection of the ACS website. Data profile tables are created with selected indicators on four factors: social, housing, demographic, and economic. In this portal, we can select the geography level and l ocation we need and then select the topic of the data profiles we’d like to see, as in Figure 15. For our example, we will use California’s Los Angeles County. To select this location, we first choose “county” under Geography Type and then find California and Los Angeles County in the drop-down menus that appear. We click the “get data profile links” button to move to the next screen.

Figure 15: Selecting location and geography for the data.

Because our focus is on housing characteristics that may give us an idea of where residents may be at risk for eviction in our area, we select the link to pertinent housing data, as in Figure 16. The housing characteristics data profile will give us information on factors like property values, crowding, and rent burden. On your own, you can explore other data profiles for your area (including social, economic, and demographic) that can provide you with more context. Try it!

Figure 16: Selecting a data profile subject area.

S T E P 2 : F I N D AN D U N D E R S T AN D TH E R E L E V AN T HOU S I NG I N D I C A TO R S

After selecting “Housing Characteristics,” the next screen will display our results: the requested table “Selected Housing Characteristics”, labeled DP04. By default, this table will display data from the latest available ACS five- year estimates, noted as “2019: ACS 5 - Year Estimates Data Profiles” next to the table name and shown in Figure 17. If we were interested in comparing data over time or seeing data for a different year, we would use the drop-down menu to select a different year. (Try it on your own. What story do housing characteristics for your area from this year tell, compared to what may have been happening in 2010?)

Figure 17: Table DP04 information.

Let’s review our results table. The screen we see provides a quick overview of housing character istics for our area. The table contains labels, estimates, percentage estimates, and margins of error for each estimate. We care about our area’s residents and their risk of housing loss due to the pandemic. Therefore, we want to know four things about our location: what percentage of our area residents are renters (i.e., what percent of

housing units in the area are renter- occupied), whether our area has high resident “turnover” (i.e., what percent of residents moved into the area in recent years), whether our area has housing units with crowding (i.e., more than one occupant per room), and whether our residents are rent-burdened (i.e., how much rent do our residents pay as a percentage of their income). We can scroll through this table to find clues about the answer to each of these questions. First, we can look for what percentage of our area residents are renters. The “housing tenure” row describing occupied housing units can answer this question. Housing tenure can be of two types: either owner-occupied or renter-occupied. The percent column tells us what percent of Los Angeles County residents fall into each category. As we can see in Figure 18, renters occupy an estimated 54.2 percent of housing units in Los Angeles County. The margin of error for this estimate is +/- 0.3 percent, which tells us that the percent estimate could vary between 53.9 percent and 54.5 percent. This is a minor variation, and in either case, the data tells us that over half of the county residents are renting. The population of renters, with low-income renters, in particular, could be at heightened risk of eviction.

Figure 18: Learning about housing tenure from the housing characteristics data profile table.

Other estimates included in the table can help address our remaining housing questions. Suppose we navigate further down the page to the indicator telling us about the number of residents who moved into their housing unit during a particular period of years. In that case, we can see that most units — 27.4 percent — became occupied between 2010 and 2014, shown in Figure 19. Only 8.8 percent of housing units became occupied between 2017 and today. This suggests that Los Angeles County has not recently experienced a surge in moves or new residents that could indicate housing instability.

Figure 19: Learning about the year Los Angeles County residents moved into their homes.

Next, let’s look at whether Los Angeles County residents are experiencing crowding. Crowding suggests that families may not have sufficient space in their home for all members; usually, we describe a household as crowded if it has more than one person per room and overcrowded if it has more than 1.5 persons per room. As shown in Figure 20, 88.7 percent have one or fewer occupants per room within occupied housing units.

Crowding may be an issue for the remaining housing units; 6.6 percent, or 218,863 housing units, have between 1.01 and 1.50 individuals per room, and the remaining 4.7 percent, or 157,049 units, have 1.51 persons per room or more. This reveals some crowding in Los Angeles County housing units, but it is not severe.

Figure 20: Learning whether crowding is an issue in Los Angeles County housing units.

Finally, we investigate whether Los Angeles County residents are rent-burdened. This means that they spend more than thirty percent of their household income on rent, making it difficult for these households to cover other expenses and build savings. ACS data tells us, as shown in Figure 21, that in Los Angeles County, of all 1,711,020 occupied housing units where residents pay rent, 9.5 percent of these units are occupied by individuals who are spending 30 percent to 34.9 percent of their household income on housing. Alarmingly, occupants in 48.1 percent of these housing units pay thirty-five percent or more of their household income on rent. This means that a large share of Los Angeles County residents — a combined 57.6 percent — spend over thirty percent of their income on rent. These individuals may be at high risk of eviction if their income drops significantly, as has been the case during the pandemic.

Figure 21: Learning about rent burden in Los Angeles County.

S T E P 3 : COM P A R E YOU R A R E A TO ANO TH E R G E OG R A P H Y

After we have learned about potential housing issues that residents face in our area, we may be interested in understanding how we fare compared to other counties. Table customization options allow us to add a nother area to our results table. Let’s try doing so. We can add a county by selecting the “Customize Table” button on the right-hand side of our screen, as shown in Figure 22.

Figure 22: Customizing the results table.

In this example, we are curious about renting, rent burden, crowding, and housing instability in Los Angeles County compared to its neighboring San Bernardino County. We can click the “2 Geos” button from the top options bar to find our answer, shown in Figure 23.

Figure 23: Relevant selection tools in the customize table section.

Select the level of geography (county), state of interest (California), and county of interest (San Bernardino County) in the options screens that appear and are displayed in Figure 24. We could make the same comparison at other levels of geography — try comparing states or census tracts — or select other areas, including those outside of California.

Figure 24: Selecting an additional location for comparison.

To make the comparison clearer in the table, we will de- select the “Margin of Error” button, highlighted in Figure 24. This does not mean that the margin of error is not important, but it will allow us to see estimates of Los Angeles and San Bernardino side by side; we can always click the “Margin of Error” button again to display the additional information. We are also able to more easily compare both counties to estimates for the whole state of California. After selecting these options, we can collapse the options screen and review our results. Using the example of housing tenure that tells us about these areas’ renter population, Figures 25a, b, and c show estimates for our three geographies. We can learn that in the state of California, renters occupy 45.2 percent of all housing units. In Los Angeles County, as we know from our previous steps, this percentage is slightly higher, at 54.2 percent. This means that Los Angeles County has an approximately 10 percent higher renter population than the state average. In comparison, renters occupy 40.2 percent of housing units in San Bernardino County. San Bernardino County therefore has a lower percentage of renter-occupied housing units compared to both Los Angeles County and the state. This comparison suggests that Los Angeles County may have a higher population of renters and a potentially higher risk for more evictions during the pandemic than the entire state and San Bernardino County. This concludes our comparison, and we’ve learned how to work with tables. In the next case study, we’ll get familiar with a different table type, learn how to download a table, and how to filter table information to simplify our view and make the table easier to read.

Figure 25a: Renter-occupied housing unit estimates for all of California.

Figure 25b: Renter-occupied housing unit estimates for Los Angeles County.

Figure 25c: Renter-occupied housing unit estimates for San Bernardino County.

Case Study 3: Who in My County is at Risk of Food Insecurity and How has That Changed Over Time? Job loss and fewer economic opportunities during recessions reduce individuals’ income. These factors can compromise food security, forcing persons and families to choose between paying for housing expenses and paying for food. As the coronavirus pandemic wreaked economic havoc in 2020, an estimated forty-five million Americans, including fifteen million children who may have lost access to school meals, faced food insecurity. Similar projections for the year 2021 suggest that many communities across the United States will continue to experience financial struggles that compromise their food access and put them at greater risk of poorer long-term health. Areas with higher shares of children and older adults, low-income or unemployed, and Black or Hispanic residents may be particularly vulnerable. The Supplemental Nutrition Assistance Program (SNAP), also known as “food stamps,” is one federal program that can help alleviate food insecurity. SNAP provides lower-income households with economic resources to buy food. It is also an indicator of food insecurity — those who receive SNAP benefits are already at risk — and can provide us with insight into food access barriers within communities. As an Extension professional, you can often rely on intuition in assessing potential food hardship within your area. Data can bolster your “hunch” about food insecurity in your community and inform your idea of who is at risk and may benefit from a food bank, food pantry, or another type of informal food assistance outreach or program. Data on SNAP receipt is available in the ACS. In case study 1, we learned how to investigate where in your county residents with particular characteristics live, and you can apply the same steps to identify areas where your constituents are receiving SNAP. This case study will guide you through how to use ACS data to learn who in your county is receiving SNAP and may have been at even greater risk for food insecurity during the pandemic. We will walk step by step through how to answer this question. We will also showcase how to use ACS to investigate change over time, looking at whether receiving SNAP benefits in your community has decreased or increased over the past several years.

S T E P 1 : S E A R CH F O R YOU R K E YWO R D .

By now, you are already familiar with the initial step of looking for data: We first search for a broad keyword using https://data.census.gov. In this case study, we want to understand who in our county receives SNAP benefits, so we use “SNAP” as the keyword. Figure 26 shows the suggestions that the data portal gives us as we type. The first suggestion refers to the SNAP/Food Stamps topic. We select this option to get a complete list of data tables that describe SNAP receipt.

Figure 26: Performing a search using the data.census.gov search bar.

S T E P 2 : S EL E C T TH E D A T A T A B L E .

Because we selected a topic, the data portal will show us a listing of table, map, and page results that fall within the SNAP/Food Stamps category. To find the information on the characteristics of SNAP recipients in our county of interest, we must next filter t he results. We will do so by first clicking on the “filter” option shown in Figure 27.

Figure 27: Accessing the advanced result filtering option.

Our next step will be to select an appropriate table. ACS subject tables can quickly offer us a summary of information across multiple factors like family size and poverty or work status when looking for a resident group’s social and demographic characteristics. You may recall that we used detailed tables in case study 1; detailed tables are well-suited for exploring geographic areas and display information for one indicator (or one variable) such as age or employment status. As opposed to detailed tables, subject tables show information about a collection of statistics on a topic. They are thus more helpful in understanding a broad issue or substantive (rather than geographic) area. Even if you do not use the advanced filtering feature to look only for subject tables, you can recognize them by the p refix “S.” Here, let’s access the subject tables using the filter as shown in Figure 28. We select the “Survey” filter on the left - side pane and look for “ACS 5- Year Estimates Subject Tables” on the right side. When we are done with our selection, we can c ollapse this panel.

Figure 28: Filtering results to view subject tables.

Figure 29 displays the only table that results from our filtered search, and its title matches our interest in SNAP recipient characteristics. As we mentioned, you can tell that this is a subject table because its unique identifier number begins with an “S.” You may also notice that the search results suggest the years for which this subject table is available. We will use this information later. For now, let’s click on the table name to review its contents.

Figure 29: Filtering results to view subject tables.

S T E P 3 : R E V I EW T A B L E CON T E N T S AN D S E L E C T TH E G E OG R A P H I C A R E A .

We have now accessed the data table containing information about the sociodemographic characteristics of SNAP benefit recipients, shown in Figure 30. Take a moment to review its structure and content. The left side of the table lists all the person, household, and family characteristics available. These include household structure, the presence of children in the family and their ages, household poverty status, whether any persons with disability live in the household, racial and ethnic composition, and family work status. The columns that appear to the right display raw number estimates, margins of error (or variation around the estimates), and percentage calculations for each group that appears in the column headers: total U.S. population, households receiving SNAP benefits, and households not receiving SNAP benefits.

Figure 30: The Food Stamps/SNAP subject table.

Here, we are interested in the group of food stamp recipients, and we can filter the table to show only these data. We can use the “columns” button on the right -hand side of the screen to select the statistics we would like to see. Figure 31 shows the options that appear, and we can uncheck boxes that we don’t need. Let’s select the options to view raw numbers and percentages that pertain to the group of SNAP recipients.

Figure 31: Selecting table columns.

We’ve now obtained information on our resid ent group of interest but have not yet narrowed down the data to our county. We are viewing the default display that describes the entire United States. We can see that, for example, of all households receiving SNAP benefits in the US, 47.3 percent lived below the poverty level, and 52.7 percent had household income at or above the poverty level in 2019. If we instead filtered our table to households not receiving SNAP benefits (Try it!), we would see that only 8.4 percent had income below the poverty level among the latter group. In comparison, 91.6 percent had income at or above the poverty level in the same year. Looking at this data point alone tells us that SNAP recipients are a vulnerable group.

As the last step, let’s now use the “customize” but ton shown in Figure 30 to view data that describes our county. We can do so by selecting the “geos” (which stands for “geographies”) option in the customization panel that appears and is displayed in Figure 32.

Figure 32: Customizing the table to display information about a selected geographic area.

After we select “geos,” a familiar geography filtering screen appears. We used the same filter in case study 1. We repeat the procedure we conducted in the case above and sequentially click through filtering panels to identify our geography of choice. Here, let’s retrieve information about Carroll County, Virginia. Figure 33 shows the steps to obtaining this information, but you can select any state and county. After you are done with your selection, click “Close” or collapse the cu stomization panel.

Figure 33: Selecting a geographic area for the subject table.

S T E P 4 : U ND E R S T AN D TH E D A T A .

The result of our filtering and geography selection, shown in Figure 34, is the data we were looking for: We now have information on the social and demographic characteristics of households that received SNAP benefits in 2019 in Carroll County, Virginia (or in your selected county). Let’s look at what we can learn about this group of country residents. For clarity of presentation, Figure 34 shows percentages for households that received SNAP and for those who did not receive SNAP benefits in Carroll County, which we toggled using the “columns” option on the right side as we learned in step 2. We removed the display of margins of error by clicking and de- selecting the “margin of error” button on the top navigation panel. (Try it!)

Figure 34: Subject table with results.

The resulting table contains data for Carroll County, Virginia, residents. It shows percentage breakdowns for each selected socio-demographic characteristic for two groups: households that received SNAP (left center column) and households that did not receive SNAP (right center column). The left-most column lists all the indicators, or variables, available in the SNAP subject table, along with its categories. First, notice that different labels are more or less indented. This gives us a clue about our “total” category or the category that represent s 100 percent for our estimates. You can see “households” is the least indented, and you can read this line as follows: Out of all (100 percent) households in Carroll County, 11.2 percent

Page 1 Page 2 Page 3 Page 4 Page 5 Page 6 Page 7 Page 8 Page 9 Page 10 Page 11 Page 12 Page 13 Page 14 Page 15 Page 16 Page 17 Page 18 Page 19 Page 20 Page 21 Page 22 Page 23 Page 24 Page 25 Page 26 Page 27 Page 28 Page 29 Page 30 Page 31 Page 32 Page 33 Page 34 Page 35 Page 36 Page 37 Page 38 Page 39 Page 40 Page 41 Page 42 Page 43 Page 44 Page 45 Page 46 Page 47 Page 48 Page 49 Page 50 Page 51

impact.extension.org