Introduction

Neighborhood data counts. 

Neighborhood data counts people and properties, vouchers and vacants, murals and mobility, crime and connectedness. These counts become indicators through data aggregation, analysis, and normalization, allowing for comparison year over year and decade-over-decade. 

But what precisely is an indicator, and how does it differ from data? What data gets used to create indicators? What does that process look like? What do indicators tell us about neighborhoods and communities? For long-time readers of Vital Signs, these questions may be old hat. But, for others, the answers to these questions might not be entirely clear.  

This introduction for Vital Signs 21 serves as a brief refresher on how BNIA-JFI creates this report. Vital Signs compiles numerous datasets to develop indicators. The datasets used in this report include U.S. Census Bureau counts, American Community Survey estimates, business databases, 311 call records, 911 call records, home sales databases, snapshot reports from City and State agencies, and many others. All these datasets share three common themes – they create equivalences across different cases, they develop categories, and they code individual phenomena into categories. ^[1]  

For example, the primary dataset used in [Chapter 5: Crime and Safety] comes from Part I crime numbers. Incidents of crime reported in Part I are considered the most serious offenses. By organizing serious offenses into a single type, this dataset makes many forms of crime equivalent. There are two categories within Part I criminal offenses that BNIA-JFI analyzes, property crimes and violent crimes. Different types of crime are coded to each category. For instance, burglary and auto theft are subcategories of property crime. 

Why does data classification matter to readers of Vital Signs? First, while a data point represents a ‘count’ of an occurrence, the systems that classify data shape what we understand about the event. These classifications inform what indicators BNIA-JFI can develop and track annually. Second, classification systems are created through interpretive work. Sorting data into categories requires human judgment. 

Ultimately, BNIA-JFI collects and warehouses data from public and private sources. We can only develop indicators by using the data classifications provided to us. Throughout Vital Signs 21, you’ll encounter numerous descriptions of the data sources we use, their limitations, and their strengths. We include this for transparency and to recognize that while our indicators point to events within neighborhoods, ultimately, we do not control how that occurrence gets counted, classified, or categorized.  

Indicators, on the other hand, are developed and maintained by BNIA-JFI.  As stated in the FAQ on our website , “indicators are a measure or set of measures that help to quantify express progress towards a goal or outcome”. Another definition describes indicators as a “knowledge technology that can be used to quantify, compare, and rank virtually any complex field of human affairs” veers into the domain of academic jargon. ^[2]  But describing indicators as a technology that aims to sort out and compare complex local occurrences through quantification also speaks to the heart of the Vital Signs project.  

While BNIA-JFI is a data intermediary, we also recognize that creating indicators requires interpretation. For instance, you’ll notice that many of our indicators provide a rate followed by the phrase ‘per 1000 residents’. This unit of analysis is a result of normalization where BNIA-JFI makes all our Community Statistical Areas comparable to one another. While using population provides a way to compare neighborhoods consistently over time, there are other ways that we could have chosen to normalize the data for our indicators.  The count of people, events, or properties can be divided by total population of people or properties to create a rate.

We want to make our choices as transparent as possible so that others feel confident using our resources. BNIA-JFI also acknowledges that as we make choices about data and indicators, we necessarily introduce bias into our work. Amongst media, technology, and machine learning scholars such as Ruha Benjamin, Safiya Noble, Margaret Mitchell, and Timnit Gebru, there is a consensus that rather than being value-neutral, algorithms and other forms of data-driven technologies often reproduce sexist and racist ideologies and burden those most vulnerable.  

The calculations BNIA-JFI uses to create indicators are in no way as complex as machine learning tools. But, given our institutional mission to provide meaningful and actionable data, we have an obligation to provide transparency. 

Mel Kranzberg, a historian of technology, developed a fitting truism applicable to thinking about indicators and their use as knowledge technologies. He writes, “technology is neither good nor bad; nor is it neutral”. ^[3]  Indicators presented throughout these chapters point to neighborhood characteristics, but they are not neutral descriptors. Ultimately, this is the key strength of Vital Signs. Our indicators are place-based and informed by over 20 years of community feedback, input, and engagement. BNIA-JFI indicators reflect diverse perspectives and values from Baltimore neighborhoods – making them powerful tools in the quest for change. 

In 2020, the United States conducted its 24th census. Although the decennial census occurs with the certainty typically allotted to only death and taxes, the 2020 count was one for the history books. The 2020 enumeration coincided with the COVID-19 pandemic and marked the first time all US households responded to the census online. At a national level, we also witnessed how  counts  or even the  questions  asked on the census have profound political consequences. 

For Baltimore, the biggest takeaway from the initial release of census data was the confirmation of a long-feared trend. Between 2010 and 2020, Baltimore experienced a population loss of -5.7% from 620,691 to 585,708. This decline is particularly noteworthy as all cities along the I-95 corridor gained population during this period. More recent releases from the Census Bureau in 2023, show that Baltimore lost an additional 7,000 residents over the past year. It’s uncertain if this loss is a permanent or temporary effect of the Covid-19 pandemic. Such drastic population loss presents an existential crisis for Baltimore, its residents, and its economy.   

As BNIA’s  Community Change 2010-2020  project demonstrated, the pattern of population change in Baltimore has remained consistent since 2000. Between 2000 and 2010, many communities in the east and west parts of the city declined by more than 10% while communities along I-83 and around the Inner Harbor grew. Between 2010 and 2020, much of the same pattern persisted. Communities that grew in the first decade continued to grow in the second decade; communities that declined continued to decline. Over the two decades, 44 out of the 55 Community Statistical Areas (CSAs) in Baltimore maintained the same pattern.  

Population loss also showed significant variation by race and ethnicity. The city declined by Black/African American (-15%) and White/Caucasian (-11%) populations yet gained Hispanic (+77%) and Asian (+46%) residents in the last decade.  

Despite the declining populations, Baltimore neighborhoods find ways to change, adapt, and respond. These qualities are evident in Vital Signs 21 as the city began to recover from the first year of the pandemic. The Baltimore City Health Department began the administration of the  Moderna Covid-19 vaccine in January 2021  to priority groups. In March 2021,  Mayor Brandon Scott also eased capacity  restrictions for fitness centers, recreation centers, religious institutions, libraries, and other facilities. By the end of April 2021, all residents 16 years and older were eligible for the vaccine. On July 1st, 2021, Former Governor Larry Hogan ended the statewide COVID-19 state of emergency.  

While these milestones signal a return to normalcy, Baltimore City remains deeply impacted by the pandemic. Throughout Vital Signs 21, you will find evidence of disparity in recovery. In Chapter 3: Housing and Community Development you will see that as housing prices in the city started to soar in 2021, many residents struggled to afford housing. In Chapter 8: Education and Youth, chronic absenteeism amongst elementary, middle, and high school students soared. The communities with the highest levels of chronic absenteeism also have the highest percentages of children living below the poverty line.  

A city divided cannot stand, nor can it thrive. In BNIA’s Community Change project, we used the phrase 'enduring divergence' to describe the historical and increasing disparities between Baltimore neighborhoods. Enduring divergence, or pulling apart, identifies a trend requiring justice-informed policy solutions that redress historical harms and disinvestment. Such solutions have the potential to enhance all of Baltimore’s neighborhoods.      

For example, 55% of Baltimore City residents commute to work outside of the city. While this percentage fluctuates slightly year-to-year, it has not dipped below 53% in the past ten years. For some CSAs like Brooklyn/Curtis Bay/Hawkins Point and Morrell/Park Violetville, the number is above 70%. Even in Downtown/Seton Hill, a CSA where a quarter of the population walks to work, 45.6% of residents work outside of the city (this is also the lowest percentage of workers who work outside the city of any CSA). Almost half to just below three-quarters of the population leave the city to work, making this a city-wide concern. These are troubling figures when you consider that cities are frequently the center of economic activity – not the periphery.    

Further, 25.6% of Baltimore City has a travel time to work between 30-44 minutes, and 20.8% has a travel time to work of 45 minutes and over. These two indicators point to how challenging it is for residents to live and work in the city. As demonstrated in our Community Change 2010-2020 project, a higher percentage of residents traveling more than 45 minutes is associated with population loss, higher percentages of African American households (and neighborhoods with lower diversity overall), higher unemployment rates, and higher percentages of no internet access at home.  

Therefore, we can also understand long commute times, and nearly half of the workforce seeking jobs outside the city's borders, as a legacy effect of segregationist policies and practices that neglected Black and poor neighborhoods to enhance the predominantly white urban core. Retaining and gaining population requires a commitment to deconstructing this legacy. Better and closer jobs, reliable transit, maintained civic infrastructure, and enhanced connectivity are parts of this solution – but only if they materially improve the lives of those who experience the greatest burdens.  

Over two decades ago, several Baltimore nonprofit organizations, city government agencies, neighborhood leaders, and foundations, developed the concept of a “data intermediary” dedicated to developing and maintaining a community-based data system open and accessible to all neighborhoods. BNIA also became an early partner in the Urban Institute’s  National Neighborhood Indicators Partnership  (NNIP), which today is a network of organizations with similar missions in more than 30 cities across the United States. 

In 2006, BNIA moved to the University of Baltimore’s Jacob France Institute and was renamed the Baltimore Neighborhood Indicators Alliance – Jacob France Institute (BNIA-JFI). 

Since 2002, BNIA-JFI has been producing the Vital Signs report annually to provide outcome indicators that "take the pulse" of Baltimore neighborhoods’ progress towards a better quality of life in every neighborhood. The goal of this effort is for neighborhood residents, organizations, and other stakeholders to use data and the Vital Signs report to strategically and effectively foster new ways of thinking about improving our City, neighborhoods, and government over time. In 2012, Baltimore City Council passed a resolution that endorsed the use of Vital Signs in local policy-making to “reflect the diverse conditions of neighborhoods and provide the basis for a system of tracking progress toward a shared vision” for Baltimore. 

Over the years, the Vital Signs report and analysis community-based data have supported decision-making in Baltimore City and its neighborhoods. 

Indicators provide relevant tools to Baltimore’s neighborhoods, communities, and institutions. BNIA-JFI is committed to assisting our stakeholders use these tools to achieve their goals and advocate for change. Through the NNIP Network, BNIA-JFI is connected to nearly three dozen other cities for learning and advancing research on neighborhood-based data. Below are a few ways to use BNIA-JFI indicators and data. Our staff is also available to provide technical assistance, grant writing support, and other resources to communities in Baltimore.  

• Local planning processes: Several neighborhood, local, and regional plans include specific indicators to monitor, track, and evaluate the effectiveness of plan implementation.  

• Grant writing: Community-based organizations and non-profits rely on Vital Sig indicators to help make a data-driven case for leveraging resources.  

• Advocacy: Vital Signs indicators may be useful for community advocates when preparing public testimony, submitting comments, or in movement-building activities. 

Wherever possible, Vital Signs uses Community Statistical Areas (CSAs) as the geographic level for which data is provided.  CSAs are clusters of Census Tracts  that correspond to Baltimore’s neighborhood boundaries and are consistent statistical boundaries for which data can be acquired. Neighborhood lines often do not fall along CSA boundaries, but CSAs are representations of the conditions that occur within those particular neighborhoods. The CSAs were created in 2002 and were revised for Vital Signs 10 using new 2010 Census Tract boundaries. CSAs were not revised in 2020 following the Census, but some names were changed based on community input.