*We will continue to update this page with abstracts as we receive them*
Do people hold politicians accountable for the performance of government? I test this question in a natural experiment using individual-level experiences with the performance of public transportation. I compile records of performance tracked via individuals’ fare transactions and train delays, link these data to opinion surveys, and further test the importance of the attribution of responsibility on people’s ability to hold government accountable in a survey experiment. I show that people perceive different levels of performance, but fail to connect performance with judgments of government. I find that when people are experimentally provided with information on responsibilities, they are able to connect their experiences of performance with their opinions of government. However, these judgments are susceptible to a cognitive bias favoring recent information, stunting the ability of citizens to reflect on performance over time. These results demonstrate that confusion about government responsibilities and inherent human perceptual biases can frustrate accountability.
As Americans’ trust in government nears historic lows, frustration with government performance approaches record highs. We propose that Americans’ views of government can be reshaped by increasing government’s operational transparency – that is, the extent to which citizens can see the often-hidden work that government performs. Across two studies using laboratory and field data, increasing operational transparency improved citizens’ views of and increased engagement with government. In Study 1 (N=554), viewing a five- minute computer simulation highlighting the work performed by the government of an archetypal American town – from building roads to ensuring food safety – increased trust in government and support for government services. Study 2 (N=21,786) leveraged field data from a mobile phone application through which Boston residents submit service requests to their city government. Users who viewed photos of city workers responding to their service requests were more likely to continue using the app over the ensuing 13 months, demonstrating that operational transparency led to sustained engagement with government.
CityScore is an initiative designed to inform the Mayor and city managers about the overall health of the City at a moment’s notice by aggregating key performance metrics into one number. By applying a uniform scoring methodology to key performance metrics from across the wide variety of services and functions the City provides, CityScore has been an effective tool for focusing executive attention on areas most in need of improvement. The intuitive scoring system has also been successful in increasing engagement with performance data at all levels of City staff, leading to more data-driven conversations and decisions. CityScore has been the impetus for some of the most significant performance management and analytics efforts the City has pursued over the last year, including an in-depth engagement with Boston Emergency Medical Services on response times, and BOS:311 on operational efficiencies. Interest in the CityScore tool from myriad municipalities led to the development of an open source toolkit that allows any organization to set up a CityScore dashboard using their own metrics and data. The City of Boston believes CityScore will continue to be a powerful tool to prompt meaningful questions and drive significant improvements in city services.
How can a school district optimize its allocation of limited resources? What can a district learn from other districts’ approaches to spending and staffing allocations? Can the costs of best practices be estimated? The Department of Elementary and Secondary Education is developing a new set of tools for school district leaders—Resource Allocation and District Action Reports (RADAR)—that aim to help answer questions like these. Key features of the tools are highly visual data displays, and comparisons to ten selected districts. The Department is going deep into its huge data collections to create new data indicators that better describe how districts use their people, time and funds. For example, we can show typical class sizes by subject, grade level or school, or the time 9th graders who failed the prior year state assessment spend in ELA and math compared to other students. This year the reports are being introduced to a few pilot districts for testing. The pilot allows Department staff to learn, for example, if data from separate data sets are being matched up correctly and creating useful indicators; whether the new reports really present actionable information; and whether the new formats are accessible and understandable.
Boston Public Schools created the Opportunity Index to capture the characteristics about the home neighborhoods of students attending the Boston Public Schools. This “place-based” index measures factors such as household income, student academic performance, and crime in Boston census tracts. BPS developed the index in order to provide a deeper understanding of access to and gaps in opportunity that students face in their home neighborhood to better promote equity in delivery of services and resources.
The adaptive challenges we tackle using inquiry cycles are problems that are too complicated and too important to tackle alone”. In this session, participants will examine Boston Public Schools’ model for teacher-led collaborative inquiry. The presenters, the Boston Public Schools Data Inquiry Team, work collectively with schools and other Boston Public Schools district departments to provide coaching and resources for inquiry. BPS aims to achieve sustainable and scalable school improvement; program data from our first three years of implementation shows inquiry schools have above average student growth and teacher perceptions of efficacy. Using the Data Wise Improvement Process developed at Harvard Graduate School of Education, BPS Inquiry Facilitators progressively build the capacity of teacher leaders to guide their own teams to improve student achievement. The inquiry team uses a gradual release of responsibility coaching model that challenges teacher leaders to hone skills related to facilitation of adult learning, assessment, and data literacy. Through inquiry, teacher teams use multiple sources of data to identify, understand, and resolve student learning challenges.
In Boston, 3rd grade students are tested to place into advanced-work classes. In 6th grade students opt in to test for admission to elite exam schools. However, research indicates that Boston’s young males of colour are significantly underrepresented in accelerated programs. Jennings (2012) suggests that combining societal, career and educational inequities in communities create neighborhood distress scores, representing spacial inequalities. This measurement may be exacerbated by students’ high absenteeism, mobility, homelessness and weak achievement. Many reside in high crime and poverty neighborhoods. The school library impact research studies demonstrate causality between access to active school library programs (ASLP) and academic success and strong standardized test scores. Preliminary research indicates that those without ASLP access, or the informationally underserved (IU), perform more poorly on standardized tests than their counterparts with access. An analysis of school geocodes and educational intervention variables, including school library access, may showcase how all students deserve access to accelerated learning programs. Ideally, student demographic data and accelerated program entrance data should be compared with school geocode locations and ASLP access within BPS. This presentation highlights access to educational resources by young males of color and school library access by geocode and MA State Accountability and Assessment Levels.
Individual drivers’ decisions on which routes to take lead to a traffic “equilibrium.” To make these decisions, drivers use an “implicit” congestion function indicating the time a route is expected to take as a function of prevailing traffic conditions. This function, however, is unknown. In this work, we use actual traffic data provided by the City of Boston and the Central Transportation Planning Staff of the Boston Region Metropolitan Planning Organization. The data cover the Eastern Massachusetts transportation network and contain spatial average speeds and road segment flow capacities on a on a minute-by-minute basis throughout 2012. Using the data, we have developed methods to estimate Origin-Destination flow demand matrices and driver congestion functions consistent with the data. We quantified what is known as the Price of Anarchy (PoA), a ratio of an overall congestion metric under selfish driver routing decisions over the same congestion metric achieved when all drivers cooperate to achieve a social optimum. We found that PoA can be as high as 2 at specific days and times. This indicates that cooperation could potentially halve congestion at certain times. We propose specific interventions with the potential to make such a reduction possible.
Do you drive more safely if you have a passenger in the car? What if that “passenger” is an app running on your phone, and at the end of each trip it reminds you of your performance? In the fall of 2016 the City of Boston and Cambridge Mobile Telematics set out to answer that question in an attempt to nudge “the worst drivers in the country” (according to at least one major insurance company) towards safer behaviors. Through the Boston’s Safest Driver competition, a telematics app monitored five driving behaviors and generated an aggregated score. The competition incentivizes players to compete with their friends, family, neighborhood, and region to become the safest driver. During the three months of this unique public-private partnership driver behavior among users saw a significantly positive shift and the 200,000 recorded trips generated new insights into the City’s transportation system. This presentation highlights the results of the Boston’s Safest Driver Competition and probes some of the ways this citizen-generated dataset is helping the City of Boston reach its Vision Zero goals.
Vision Zero, from its first realization in Sweden in the 90s to the new wave of municipality-driven initiatives in the States within the last five years, has revolved around breaking with traditional ethical stances on road safety. In accordance with this initiative, the City of Boston’s Transportation Department and Department of Innovation and Technology are taking a more comprehensive, system-wide view of existing and novel data to prioritize areas demanding intervention and to evaluate interventions both narrow and wide in scope. A not insignificant part of the City’s program thus far has been to better collect and surface crash data and resident safety concerns, which has improved understanding of historical crash patterns and allocation of limited resources, expanding conceptions of road safety driven to better illuminate long-term areas of risk. The city is also pursuing and integrating more temporally and spatially granular data to support crash modeling and novel metrics. Such data acquisition efforts are crucial for designing more rigorous experiments and leaner interventions. As Boston improves on existing reactive approaches to road safety and expands the scope of road safety, we are also opening up opportunities for advanced predictive modeling and estimating the effect of interventions before they happen.
back to top
Street gangs are important subjects of study and targets of policy intervention because of their members’ increased risk of criminal involvement as well as victimization. A fairly recent development in gang research has been the application of social network analysis to an urban area’s street gang setting. Much of the work on the social networks of gangs focuses on the gang as the unit of analysis, relating them to one another based on rivalries and alliances. Individual-level work often examines the risk of victimization based on the experiences of one’s social network. Because of the gap in the literature concerning the individual-level, positive relationships between gang members, I examine the co-offending networks of gang members in the Boston area. I aim to understand how members of different gangs relate to one another with respect to their involvement in joint activity. This study provides much-needed evidence concerning how these positive relationships are structured between gangs. In addition, examining the network in relation to the crimes committed by co-offenders further expands the knowledge of the behavior of urban street gang members.
This presentation uses the Boston Police Department Field Interrogation and Observation (FIO) reports from 2011-2015 to examine police-citizen interactions in the City of Boston. Specifically, I build upon the findings of Fagan, Braga, Brunson, and Pattavina (2015) and their analysis of policing patterns in Boston from 2007-2010.1 Using publicly-available data from the City of Boston Open Data Initiative and the Census Bureau I measure the relationship between race and complexion on ‘levels of intrusion’ in police practices. My research goes further in offering an additional analysis of police-civilian interactions as they pertain to motor vehicle violations. Motor vehicle violations are a category of crime that affords a high degree of police discretion in enforcement and the likelihood of a motor vehicle operator getting stopped by a police officer does not necessarily increase as the risk presented to the public by their motor vehicle violation increases. As such, it represents an interesting sub-set of crimes on which to test my hypotheses.
The Geography of Incarceration provided timely information as state leaders engaged in an unprecedented effort to find strategies to operate our criminal justice system in a more cost-effective manner, and redirect the savings toward models that decrease crime and strengthen neighborhoods. The report captured the extent to which the city of Boston is home to high incarceration rate neighborhoods by mapping novel data provided by the Suffolk County Sheriff’s Department. The data covered all individuals re-entering to Boston neighborhoods between 2009 and 2015 from either the Suffolk County House of Correction or the Nashua Street Jail (see box p. 6 for more on how these institutions fit into our criminal justice system). These release data were juxtaposed with 2014 crime data to provide a view of how the geography of incarceration compares to the geography of crime in the city.
The web platform was developed at the Zofnass Program for Sustainable Infrastructure at the Harvard Graduate School of Design. The platform is the outcome of data-driven research at the intersection of urban planning and water infrastructure systems, and highlights urban solutions based on sustainability policies. The City of Chelsea, in Boston area, was chosen as a partner for development of the platform. The result is a city web application that consists of two parts: an information sharing platform for non-expert awareness, and a tool for interactive city modeling. Through data visualization and system-based storytelling, the first part provides information about the system architecture, the performance data, and the geospatial synergies. The second part introduces intervention opportunities that could improve the system and promote sustainability. Our algorithm identifies places in the city as opportunities for green infrastructure solutions, prioritizes them based on best practices, and quantifies their potential impact to the water system. The web platform is a tool to assist non-expert city stakeholders, such as city officials, policy makers, real estate developers, and the community.
Note: the platform referenced in this talk will also be featured during the Screen Session
In response to the threats of our changing climate, neighborhood groups, state agencies, and municipal governments have been assessing their vulnerabilities and creating action plans. The Metropolitan Area Planning Council (MAPC), the regional planning agency for Metro Boston, has been assisted a number of municipalities and multi-municipal groups in preparing for changes in our region’s climate. These efforts are necessarily data-driven. Scientists create models that describe our climate and environmental conditions and project changes expected in coming decades. Planners then have the challenge of using the best available data on natural resources, infrastructure, land use, and demographics to translate those projections into specific threats to a given community, and to assess mechanisms that may already be in place for addressing those threats. This planning leads to recommendations on how to prioritize changes to infrastructure, policy, bylaws, or practices to make communities more resilient to extreme weather events or other effects of climate change. This presentation reviews MAPC’s methods for conducting these analyses and for effectively communicating the results through maps and other data visualizations.
Cities today are increasingly pervaded by systems, networks, and devices that incorporate digital technology, are networked, and generate data in function of their operations. This data is incidental to human activity, generated and captured in the everyday activities of urban life, and in this way reveals the diverse and complex facets of the city and its inhabitants in detail and in real-time. Work with this data can contribute to revealing novel insights into human activity in urban space, support methods to optimize system operations or inform other correlated contexts opportunistically. The Northeastern Energy Flows project aims at developing novel ways of visually representing urban energy systems real-time data to support systems optimization as well as the usage of this data for novel perspectives on campus dynamics. As part of this project a number of interactive data visualization systems provide access to fine grain data on energy consumption, building occupancy, climatic conditions and social activities. This helps in obtaining a deeper understanding of the dynamics related to energy consumption, to foster engagement across disciplinary boundaries and to support decision making. The project currently uses the Northeastern University campus in Boston as a ‘city-in-a-city’ testbed.
We describe the ongoing design and development of a prototype cloud-based data repository and platform that can assist in the implementation of algorithms, applications, and services that utilize urban data sets that have been manually assembled, retrieved from third party services (e.g., the City of Boston Data Portal), or generated algorithmically by other internal or external applications. The platform is being developed within a pedagogical context: students enrolled in a graduate-level urban data science course address urban challenges by assembling data retrieval, analysis, optimization, and service modules within a platform framework that allows interdependent data sets and interoperating components. The platform’s architecture is an exploration of solutions for common challenges of leveraging data-driven approaches in policymaking, research, and business efforts: dependence on transient or constrained third-party APIs and data sets, scalability of data processing algorithms, interoperability between distinct modular units, long-term maintenance, and provenance tracking. These are addressed by platform features that enable (1) caching and staggering in data retrieval processes, (2) merging of data sets with certain common schema elements, (3) scalable testing, and (4) exposure of capabilities via APIs. Example applications developed by student contributors using data sets relevant to the City of Boston illustrate these platform features.
Boston’s Citywide Analytics Team is partnering with the Boston Fire Department (BFD) and Boston EMS (BEMS) to improve their services and efficiency. First, we used data to improve how BEMS manages its resources in response to a growing call volume. We analyzed multiple years of incidents and outcomes, focusing on a subset of calls that they do not require acute medical care. These analyses have led to an FY18 pilot proposal for alternate response plans to reduce unnecessary ambulance dispatches and provide appropriate care to patients with mental health, addiction, and homelessness issues. This program could increase efficiency while also improving coordinated care to the City’s most vulnerable residents. Second, the Analytics Team is working with BFD to improve response times. We analyzed GPS tracks of every fire truck to measure driving speeds throughout the city and determine the quickest responder to any location. BFD can now re-structure dispatch responsibilities to reduce response times, identify problematic intersections, and suggest street design changes to allow for better flow of public safety vehicles. These projects highlight the value of latent information within administrative data, and demonstrates a successful research process for enabling data-driven innovation in government.
The City of Cambridge instituted its open data program to build on its commitment to transparency, efficiency, and innovation. By providing city data to the community, Cambridge empowers volunteer programmers, data scientists, and designers to create new applications that could benefit the municipality or its residents. However, an enduring problem for open data programs around the country is that “civic hackers” often do not have firsthand knowledge about a city’s true needs. Likewise, city staff may be unaware of the full menu of technological solutions for city problems or the full range of use cases for open data. A new model of collaboration is needed—one in which city staff, the civic innovation community, and everyday residents work together to identify and implement new uses for the open data. During the summer of 2016, hired a research fellow to investigate this issue. Over the course of eight weeks, she spoke with city staff, civic hackers, and open data experts to begin designing the new collaboration model. This presentation will detail the research findings and discuss Cambridge’s progress toward improved open data collaboration. The full research report is available at https://data.cambridgema.gov.
With support from the Knight Foundation’s Knight News Challenge on Libraries, the City of Boston recently released Analyze Boston, Boston’s new open data platform in beta (available @ data.boston.gov). This new platform is the central open data hub to find facts, figures, and maps related to our lives within the city. We are working to make this the default technology platform to support the publication of the City’s public information, in the form of data, and to make this information easy to find, access, and use by a broad audience. Underlying this platform are key relationships, processes, and an agile mindset that we believe can transform open data to open knowledge and catalyze the next generation of citizen engagement.
Often public data has been undiscoverable and unusable; methods of analysis undecipherable and not replicable; knowledge for constituents unattainable. Whether it has been buried in a dense report, put in an archaic file format, or just impossible to find; data and analytics have not achieved their vast potential. But how do we take data files and go beyond mere discoverability? How do we incorporate data into the world we live in as to create dynamic experiences, both digital and analog, for constituents? Can we take analytics and predictive modeling and apply open science principles in order to share broadly and with an eye towards collaboration? What is the future of knowledge in a world of open data, open science, internet of things, machine learning, and artificial intelligence? How do we take these vast technological changes that society is facing and harness them to serve the public good?
Previous examinations of the social process of gentrification have been predominantly influenced by traditional methodologies of neighborhood ethnographies and qualitative comparisons. While these studies have produced fruitful findings, an argument exists for the integration of new data resources in the study of social processes like gentrification. In recognition of this new argument, this exploratory analysis uses unique “big data” resources in the examination of gentrification in urban neighborhoods. The use of these resources is a new opportunity to both test and provide further insight to the traditional qualitative findings of previous gentrification studies. The power of these new data resources is demonstrated in this analysis by using the City of Boston’s business license data in conjunction with other pertinent data to measure and map commercial gentrification. The use of these diverse data in this exploratory analysis captures unique forms of entrepreneurial activity in Boston over a seven year period. Unique business clusters have been identified using business license classifications as proxy measurements for business activities. While this analysis validates previous discussions on the importance of the invasion and succession of business types in gentrifying neighborhoods, it also raises new questions and points of inquiry for future gentrification researchers.
In an effort to proactively work towards preventing residential displacement, the City of Boston has employed geographic information systems and a multitude of data to spatially understand risk of displacement across the city. Residential displacement—particularly for low- and middle-income residents—requires increased attention in light of Boston’s recent development boom and rising housing prices. While these trends are evident across the city, they yield further questions: Which areas of Boston are most at risk of displacement? How do we quantify and measure such risk? To answer these questions the City has created a Displacement-Risk map. The map is based on an index of 17 indicators of displacement, including demographics, market changes, and proximity to desirable amenities. Overlaying these 17 variables allows us to easily visualize the areas most in-need of focused resources, outreach, stability assistance, and anti-displacement planning. This data-driven output can be used to improve existing programs and inform broader policy decisions. This presentation will explain our methodology for measuring displacement risk, and show the resulting map.
Urban dwellers tend to rank the quality of their neighborhood by the amount and variety of accessible amenities from their home. Although urban accessibility has been widely studied for years, no cohesive answers arise from the literature to questions such as (1) what is an accessible walking distance to an amenity; and (2) what are the convenient times for driving and public transport rides. Measuring the spatial distribution of amenities and services in cities across countries can shed light on the inequalities amongst urban neighbourhoods and provide data-driven metrics to support location decisions in future planning efforts. Combining a large scale dataset extracted from Google Maps with data about the spatial distribution of population across the world (NASA, 2015) we measure exact travel durations by walking, driving and public transportation from population concentrations. Our results provide fine-grained insight to where are the ‘truly disadvantaged’ communities located in cities. Initial results measured in the city of Boston show that more than 85% of the urban community has a closest park, supermarket, school, bank or pharmacy located within less than 15 minutes walking. Subway stations however, are significantly less accessible in Boston, with a mean walking duration of 25 minutes accounting for only 45% of the city’s population.
With Boston aggressively pursuing pedestrian-friendly streets in its Complete Streets plan and Imagine Boston 2030 and Cambridge’s creation of the Pedestrian Plan, the conversation of walkable communities has gained prominence. This interest cuts across many domains: the concern for active mobility in public health, positive effects on real estate value in economics, increasing social concern in urban planning, among others. Moderating human behavior is the nature of place. As such, what are the characteristics most associated with pedestrian activity in the Boston area? This presentation discusses an approach for measuring spatiotemporal macro-behaviors of walking activity using individual, locative, passively-collected, mobile phone data. The scale and precision of this data offers an opportunity to move beyond the limitations of previous studies. We analyzed 250,000 walking trips passively collected from mobile phones from over 5,000 users in Boston and Cambridge over one year, at the breadth of Boston and Cambridge. This allows for a novel awareness of aggregate pedestrian behaviors that speaks to the social and physical characteristics unique to a street or city as a whole, with insight into behavior influenced by dynamic conditions.
Service-learning is a pedagogical framework that integrates sustained community engagement with for-credit academic courses and critical reflection. We conducted an analysis of a service-learning program at a mid-size, urban university to assess the impact of a service-learning program. Our research indicates that facilitated reflection on community engagement helps the development of civic-minded students in several key areas. This data helps inform avenues of prosocial student development through increased integration of service-learning courses. While impacting universities in the Boston area, this data reveals benefits for the local community with areas demonstrated to strengthen service endeavors and the development of more service-oriented community members. As a pedagogical tool that aids in the development of civic-mindedness in students, service-learning provides an opportunity for community organizations and academic institutions to develop collaborative, mutually beneficial relationships.
Trusting a data set or an analysis always requires a leap of faith. Beyond an acceptance of margins of error, the imprecision of a regression, or the assumptions of a projection, all data-driven decisions in government necessitate what William James once called a “will to believe.” When it comes to data that impact or justify government decisions, there first needs to be a will to believe not only in government’s ability to be honest and rigorous with data, but in the very authority itself of data to tell us something meaningful about the world. In an era of “alternative facts” and fear-based advocacy, this is a sad truth that we must contend with; but it may also sometimes be a symptom of data tunnel vision, of forgetting to attend to certain aspects of citizen engagement that involve the sometimes irrational, sometimes inefficient, but always human need for something more than facts to act from. How can we be better at designing the conditions for people to develop faith in our (and their) ability to do good things with data?
How do you prepare a city for changing climates and rising seas? The challenges from climate change are substantial and complex but can be addressed through bold and creative actions that support the city’s vitality and livability. More cohesive and informed communities are more resilient in climate events, so many proposed initiatives have benefits today and in the future. In considering the impacts on people, the interdisciplinary Climate Ready Boston team assessment focuses on socially vulnerable populations, people who are more vulnerable to climate hazards because they already experience stressors, such as poverty, poor health, and limited English proficiency. The vulnerability assessment considers potential damages to property and expected impacts on Boston’s transportation, power, water and sewer, emergency response, and environmental systems. Finally, it evaluates the potential economic impacts of flooding, such as the loss of jobs and disruption of business operations.Boston can thrive in the coming decades if it takes action to adapt its people, its neighborhoods, and its economic and cultural assets, starting now.
The overarching objective of this study was to conduct an analysis of data from the emergency shelter system in the City of Boston to address key gaps in existing knowledge about unaccompanied homeless adults who experience repeated episodes of shelter use over time. We used these emergency shelter data to identify a cohort individuals who entered emergency shelter between January 1, 2011 and July 29, 2012. We then used optimal matching and cluster analysis methods to identify distinct subgroups within this cohort based on their patterns of emergency shelter use over a three-year period. We found that repeated shelter users accounted for 34% of the study cohort and we identified four distinct subgroups of repeated shelter users on. The largest of these four groups (74% of all repeated users) were comprised of “sporadic users” who had infrequent and brief shelter episodes that totaled an average of 102 days over the three year observation period. Age was the most salient predictor of sub-group membership. These findings can help inform decisions about the scope and type of assistance needed to help repeated shelter users obtain stable housing.
For the first time a Massachusetts based multi-agency program dedicated to serving adjudicated youth embarked on an internal evaluation of its Arts Initiative. The main objective of the Arts Initiative is to integrate arts into academic and career-readiness programming for all adjudicated youth throughout the state of Massachusetts. Recognizing that non-cognitive skills are at the root of delinquency, and are imperative to gaining and maintaining employment, the evaluation focused on measuring the impact of programming on these skills. Despite the difficult nature of evaluating non-cognitive skills, high rates of attrition due to varying juvenile sentences, and limited capacity the evaluator was still implement the evaluation. As a result, attitudes around evaluation and views of program implementation have changed drastically and a new perspective on data-drive practice was born.
The opioid syndemic has lead to major public health challenges in Massachusetts. Fatal overdoses are considerably higher than the national average. However, tabular overdose data alone cannot fully describe the impact of the syndemic. Opioid prescription rates, infectious diseases, mental health disorders, and access to harm reduction services reveal the syndemic’s complexity. The spatial distribution of such data is highly informative and yet largely unknown in Massachusetts. We used a geographic information system (GIS) to analyze primary and secondary data focused on opioid crisis outcomes between 2002-2016. Using survey data, we characterized health status, injection habits, geographically-based socio-behavioral factors. Analyzing surveillance data, we identified the geographic distribution of opioid overdoses, prescription drug monitoring, and infectious disease. Through descriptive maps and geostatistical analyses, we identified regions with high densities of selected outcomes. We also identified municipalities with significant HIV and HCV clusters (p<0.05), neighborhoods with high densities of injection-mediated risks, and inadequate access to harm reduction services. Using GIS, we identified and characterized regions of high risk, burden of disease, and unmet need related to the opioid syndemic in Massachusetts. Spatial analyses are important to understand the broad impact of the opioid epidemic and to inform targeted public health interventions.
The opioid epidemic affecting Massachusetts and many other states has taken a record number of lives in each of the last five years. Since 2000, opioid-related deaths have increased in Massachusetts by nearly 400%. As part of a multi-faceted effort to address this unprecedented public health crisis, the Baker Administration signed a law (AKA Chapter 55) that permitted the linkage and analysis of existing data across state government in order to better guide policy development and programmatic decision-making. The results have been impressive not only from a policy perspective but also from the partnerships that have formed. Not exactly open data – but openly accessible data. Not exactly big data – but a vast array of 15+ government data sets linked at the individual level. Not exactly a public health data warehouse – but the framework for how one can be built and maintained. Now in its second year, the Chapter 55 project has brought together more than 40 government, academic, and private partners who work side by side with minimal resources on a problem that is sweeping the nation. The Chapter 55 partnership model is likely a harbinger of how public policy work will be conducted in the future.
This project uses new technology to address the deepening housing affordability crisis in Boston. Using an originally designed GIS interface, I measure the effect of both spatial sensitivity (NIMBYism for “Not In My BackYard”) and socioeconomic biases on where new housing is politically feasible. Finally, I measure how these attitudes vary between the most politically active citizens and the city as a whole. This granular image of housing attitudes will not only empower those unable to attend city planning meetings, but also inform elected officials of how best to respond to the affordability crisis.
Gentrification, one of the most debated phenomena of urban change, has proved hard to quantify. The expected result of gentrification is profound neighborhood change, but actual measurement of its effects has been elusive. Research based solely on Census indices has long been considered insufficient. Some recent efforts to tap into new data sources are reductive, focusing on a single novel measurement. Other projects are too hard to replicate, requiring access to restricted data or intense labor. I’ll show how open public data (also called administrative data), can be used to “read” the changes in local behavior that reflect shifts in socioeconomic status through time, with techniques designed to be easily reproducible. Results obtained after applying this methodology for Boston will be presented: How urban activity patterns captured in the City’s official open datasets predict socioeconomic status change in different neighborhoods?
A wide array of scholars and policymakers—including most recently the Obama White House in a widely released report—have underscored the important role that zoning plays in limiting the development of affordable housing. While a robust line of research has investigated the origins and use of exclusionary zoning initiatives and public attitudes surrounding zoning policy, there has been comparatively little scholarship exploring how the accumulation of regulations, zoning codes, and city institutions renders it easy for an individual to stop projects or substantially increase the cost of development. In this paper, we explore this issue by developing a theory of institutional NIMBYism. We combine a formal model in which individuals and developers bargain over affordable housing projects with case studies and statistical evidence from the Boston metropolitan area to reveal how contemporary zoning creates gridlock by allowing a small minority of citizens to stymie development. Our empirical analyses will center on a novel data set exploring the sale and development of Catholic Church properties in Greater Boston.
Boston After School & Beyond (BASB) is a nonprofit intermediary that seeks to ensure every child in Boston has the opportunity to develop to his or her full potential. To illustrate BASB’s approach to data-driven program evaluation, this presentation focuses on the Boston Summer Learning Community (BSLC), a citywide network of summer learning programs unified in their commitment to closing the opportunity gap and working together towards continuous improvement. In 2010, the BSLC featured 5 summer learning programs serving 232 students. By 2016, this network grew to 127 programs serving 10,084 students in diverse, non-traditional settings around Boston. BASB convenes this network of summer sites regularly throughout the year to conduct trainings on student skill development, to brainstorm strategies for addressing persistent problems, and to share best practices from the field. Importantly, a common suite of assessment and observational tools implemented across a diverse array of programs provides information on student-level outcomes and measures of program quality, so that programs can identify what is working and what needs to be improved. BASB collaborates with the Boston Public Schools, Cityspan, and research partners the National Institute on Out-of-School Time and the PEAR Institute to implement this continuous improvement model at scale.
Children’s Savings Accounts (CSA) programs are small dollar programs that raise expectations for children’s future; they plant a seed of hope, and research suggests that hope is a stronger predictor of college success than test scores (Elliott et al. 2013). CSA programs typically pair a financial product with technology, which allows families to monitor their savings progress. Boston Saves, the City of Boston’s CSA program, is leading the field in its use of an innovative online platform, developed by InvestCloud, which decouples the need for a single financial partner (allowing families choice in both the financial institution and the financial product they use for deposits) as well as the need to retain control over families’ accounts (placing full control of savings behavior in the hands of families). This amount of control and trust is rarely seen in government programs, especially those targeted at low-income families. Use of the InvestCloud platform, therefore, realigns the power paradigm typical of targeted programs, potentially breaking new ground and setting precedent for a more equitable distribution of power between program developers and program participants. We share preliminary qualitative findings demonstrating the use of technology as a means for government to build mutual trust with families.
The well-established connection between college completion and economic prosperity drives Success Boston—Boston’s ambitious cross-sector college completion initiative. Together, the Boston Foundation, the Boston Public Schools (BPS), the City of Boston, local colleges, and nonprofits work to increase Boston students’ college completion. Transition coaching for the first two years of college—a key element in Success Boston—seeks to ease BPS graduates’ transition into college, particularly for traditionally underrepresented students. A longitudinal study (with five successive student cohorts) examines both transition coaching program implementation and program impacts on student outcomes (college persistence, credit accumulation, GPA, FAFSA completion, degree completion). This presentation describes the evaluation, which initially focuses on SBC-participating students from BPS Classes of 2013 and 2014. The study data collected from BPS, Massachusetts Department of Elementary and Secondary Education, and the National Student Clearinghouse, includes student demographics, academic experiences, high school characteristics, and college enrollment and completion. The evaluation also collected and analyzed quantitative and qualitative data about program implementation. This presentation will provide an overview of the program evaluation and its methodology, and describe key outcomes.
Research has shown that living near major roadways or highways is associated with adverse health outcomes including respiratory, cardiovascular and neurological harm. Evidence has grown that at least part of these health risks are attributable to traffic-related air pollution released by motor vehicles. Ultrafine particles (UFP) which are a component of this pollution are of particular concern. Our research has shown associations between UFP and cardiovascular disease risk in older adults in Boston and Somerville. Working with community partners in Boston Chinatown and East Somerville, we have turned out attention to ways to reduce exposure to UFP and other traffic pollutants. We conducted two pilot studies of in-home air filtration that revealed numerous problems with making it effective. We reviewed available tactics for reducing pollution exposure near highways and assessed the evidence for their efficacy. We also conducted design activities focused on specific developments and came up with proposals about how to improve building and urban landscape design to be protective. In addition, we wrote a zoning ordinance for the City of Somerville that would require use of protective measures near major roadways and highways. We will share what we have learned with reference to specific developments in Boston.
The world’s urban population is expected to exceed 70% and urban land area is expected to triple by the year 2050. The rapid rates of urbanization have raised concerns about declines in air and water quality due to increases in nitrogen and phosphorus in runoff from fertilizer inputs and local sources of atmospheric deposition in precipitation. We measured atmospheric nitrogen and phosphorus deposition and carbon dioxide losses from soils across the greater Boston area. We find that mean rates of nitrogen deposition in the greater Boston area are almost double rates at a rural reference site and that rates of nitrogen deposition are highly variable within the greater Boston area. Ammonium deposition composes two-thirds of total inorganic nitrogen deposition and comes from a combination of spring fertilizer application and emissions of ammonia from road vehicles. We also find high rates of atmospheric phosphorus deposition, as well as large fluxes of carbon dioxide emitted from residential soils that are similar in magnitude to summertime carbon dioxide emissions from fossil fuel combustion. Together, these results demonstrate the need for atmospheric monitoring networks to incorporate urban sites and to take into account biological fluxes that produce or sequester carbon dioxide.
2.7% of Massachusetts’ natural gas (McKain et al, 2015) is leaked to the atmosphere before it can be used. Since pipe-quality gas is 95% methane (CH4), a powerful greenhouse gas, the resulting emissions have a disproportionate impact on the climate. Recent research (Hendrick et al, 2016; Brandt et al, 2016) has shown that just 5 to 7% of the leaks are “super-emitters,” emitting half of the total emissions. A new state law has mandated that these high-volume leaks must be fixed. Unfortunately the utilities, having always concentrated on safety, not volume, don’t have a proven method of locating super-emitters. An unusual partnership of academic researchers, nonprofits and for-profits studied 39 gas leaks in Greater Boston, quantifying the emissions and various metrics, to find a utility-friendly method of locating high-volume leaks. Preliminary results show the surface area of gas-saturated soil over a leak is most correlated with volume. The coalition also created an easy method to measure emissions off of leaks using common utility tools so the utilities have a learning mechanism.
The continuous subzero temperatures and unrelenting Polar vortex turned the middle of the 2014-2015 Boston winter into a disaster. The amount of snowfall which fell between January and February of 2015 released devastating payloads of snow upon the city, which resulted in transportation paralysis and amplified the vulnerabilities of communities already at risk. Building on empirical evidence that resilience to shocks and disturbances depends heavily on the social networks and trust between relevant players and local communities (Aldrich 2012; Dynes 2006; Chamlee-Wright 2010; Prasad et al. 2014), we investigate how communities with deep reservoirs of social capital overcome shocks and crises. More specifically, we examine the impact of the Boston snow storms and how the subsequent bus and rail shutdowns interacted with vulnerable populations in the Boston area. Geographically, we focus on the communities in the areas of Mattapan, Roxbury, and Dorchester as past research has identified many residents in those areas as vulnerable. We developed and administered an original survey in these communities. We apply GIS spatial analysis, factor analysis and matching to our new, sui generis dataset to explain the impact of the Boston snowstorms on urban residents and on their perceptions of preparedness for future disasters. Controlling for a number of potential confounding variables, including demographic, economic, and transportation factors, we find that density of bonding social capital serves as a mitigating force. We also illuminate how levels of efficacy – the belief that one has an impact on policies and practices in one’s community – connect strongly with perceptions of preparedness. These findings bring with them important public policy findings for residents, NGOs, and decision makers alike as we struggle to adapt to climate change and increasing urbanization.
This project involves foundational research around collective trauma in response to neighborhood violence. While violent crime rates have followed a downward trajectory in Boston over the past two decades, particular neighborhoods continue to experience pervasive and chronic violence. The serious mental health effects of exposure to violence are well known. In addition, pervasive violence may affect neighborhood social-level processes, such as social cohesion, ability to engage in collective action, and collective identity. Although community-level trauma from violence has potential durable impacts on social functioning in neighborhoods, it has received little scholarly attention or formalized policy response. Through this study, we attempt to develop a more precise description of how neighborhoods collectively respond to ongoing violence. We also aim to inform the development of methods that community groups might use to promote collective healing. In partnership with the Boston TenPoint Coalition, we collected face-to-face surveys of a randomly selected sample of residents of several violence hotspots in Boston in January-March 2017. We will discuss methodological considerations in community partnerships and present preliminary results from these surveys.
Community-based interventions and research is well established in the public health fields and is beginning to gain wider recognition in the social sciences. Social researchers are working directly with communities to produce knowledge, communicate ideas, and design digital objects or services. It is clear that the knowledge and attitudes of community partners can influence the integrity, legitimacy and outcome of academic projects. However, very little research has focused on the perspectives of communities being engaged in such efforts. Our research examines community-academic partnerships from the perspectives of Community Partner Organizations (CPOs) in Boston. CPO representatives were interviewed about their experiences and attitudes toward current and/or recent academic collaborations in the social sciences. Specifically, we address issues of communication and data. Results reveal gaps and barriers to equal decision making, open communication, and access to digital data. We conclude by offering a process of collaboration facilitated by a Memorandum of Understanding (MOU) template. The concepts and topics covered in the template were generated through the research, and the MOU is offered as a tool to adopt and adapt in collaborative research.
back to top
Massachusetts Department of Elementary and Secondary Education (ESE) developed an Early Warning Indicator System (EWIS) for all students in grades 1-12, making use of the wealth of data school districts provide. EWIS systematically identifies students that may need additional support to reach a particular academic milestone, from reading by end of third grade to persisting in postsecondary) . EWIS information is available to schools and districts through interactive reports in a secure system, and ESE provides trainings to support the effective use of EWIS as part of ongoing data work through a data driven-cycle of inquiry. While Massachusetts’ consistently outranks other states on the National Assessment of Educational Progress (NAEP) and has seen its graduation rate increase for the 9th consecutive year, urgency exists to close persistent proficiency gaps and increase the percentage of graduates enrolling in postsecondary without the need for remediation. EWIS data and tools – in conjunction with local context and appropriate interventions – can help leaders improve academic outcomes, close performance gaps, and increase the percentage of students succeeding in postsecondary. This session will describe EWIS, highlight ways schools and districts are using this data to drive student-level school reform, and engage participants in using the data.
Boston Day and Evening Academy (BDEA) is an alternative high school within Boston Public Schools that serves students at high risk of dropping out. Over the last year, the school has developed a customized Student Information System (SIS) and competency-based learning platform. The platform, called Connects, is built on top of Salesforce. BDEA uses Connects to collect and organize a wide spectrum of data about its students, with the ultimate goal of serving its student population more effectively. Unlike other SIS models, Connects allows the school to collect data on risk factors, social-emotional skill development, and interventions – all of which serve to contextualize the student. BDEA uses this data to inform school level decisions. Alison Hramiec, Brian Connor and Arpi Karapetyan will lead participants through a participatory workshop focused on the use of Connects to inform school level decisions. Participants will engage in a meeting modeled off of meetings held at the school. They will analyze and use data to make decisions aimed at helping improve outcomes for students. This participatory workshop will bring up questions of what data is important and why, the best ways to sustainably collect data, and what the data actually means for practitioners.
Infrastructure Ecology is an emerging paradigm that helps cities understand the nature of interactions between co-located, interdependent physical infrastructure, and the mapping of physical infrastructure to the political ecology of institutions and agents responsible for managing urban infrastructure. Boston has innovated a system that flags conflicts and opportunities for coordinated infrastructure management on its streets. How can other cities and towns learn from Boston? How can Boston take its leadership to the next level? This workshop brings together stakeholders from multiple utilities and agencies to explore best practices in infrastructure ecosystem management to improve Boston and cities across the nation.
Discourse around government transparency has begun to pervade civic tech and civic innovation circles. Pushing toward Open Data–making data accessible to all–has been seen as one solution both to ensuring government accountability, but also to sparking innovative uses of public datasets in the research com-munity. Led by the City of Boston’s Mayor’s Office of New Urban Mechanics in partnership with the City-wide Data Analytics Team, this set of focus groups is aimed at learning about how you conceptualize and use data so that we can better serve your needs. Capacity is limited.
Boston has an average of 32.23 annual crimes per 1,000 residents, which makes it less safe than 82% of all the cities in the US. One in every 140 people in Boston is a victim of violent crime.
I designed an interactive Boston Crime Map to show crime incidents of all kinds of theft, assault, vandalism, burglary and arson cases happened in Boston in 2016. The data of crime incidents come from [goog_1273940638]City of Boston.gov, provided by the Boston Police Department. Each circle shows a crime case happened in the exact location in Boston in 2016. Boston users can locate to the place they are in to see all kinds of crime cases happened during the year of 2016 and click the spot to see detailed information. Other users could zoom in to see all the crime incidents. Exploring the graphic, certain rules could be observed. The assault problem seems more serious in Downtown area, South End, Roxbury and Dorchester. Burglary cases are more frequent than theft.
Interactive visualizations can promote the communication and understanding of injustice. We used open data to show racial and geographical disparity in stop-and-frisk events in Boston. Throughout our design process, we considered both top-down approaches (presenting context prior to data) and bottom-up approaches (discovery of patterns through interaction). In this talk, we will cover the challenges encountered while working with open data, and how our work is situated within data visualization, journalism, and anti-oppressive endeavors.
In an effort to help policymakers better address the issue of human trafficking in the City of Boston, the Citywide Analytics Team has developed a set of metrics that gather and analyze data from backpage.com, Google, and the Boston Police Department to better quantify the supply and demand of human trafficking in the City. The information has been complied onto a dashboard that is updated daily and used to help policymakers and the Mayor better monitor human trafficking activity happening within the City. The dashboard is split into three sections: supply, demand, and intervention with each displaying metrics associated with the different parts of the industry. The tool helps decisions makers including the Mayor, cabinet chiefs, and police officials continue to stay informed and aware of patterns associated with trafficking activity. Through aggregating metrics from a variety of different sources, stakeholders can quickly gain an understanding of the overall status of the trafficking industry within the City and more easily adapt policies and decisions that better respond to specific issues.
The research team at Boston University is building a mapping platform called MAGICS (MApping Green Gray Infrastructure Coupled Systems). Underground utility maps are not in widespread circulation due to “security” concerns. But this data should be open source and available to the public for their decision making as well as in urban planning. As part of the research, our team is putting a novel methodology for collecting data related to gray infrastructures in Boston (or anywhere), using an open source platform. We demonstrate the technical integration of Google Maps API, SketchUp and ArcGIS technologies to create, characterize, map, and analyze the network of gray utilities underground and their spatial proximity to trees, buildings and road networks above ground. This enables us to derive insights into many infrastructure problems. Our online platform would be useful for collecting or disseminating data or data-derived information. The research presented here is part of a larger funded NSF (National Science Foundation) project that aims to characterize interactions among co-located physical infrastructure systems and green infrastructure(urban canopy) and the diverse human sociopolitical networks in Boston.