Policy Problem Proposal & Problem Statement

The Problem

Problem Statement Rising property prices and need for more inland government infrastructure are two problems facing the city of Boston. Undervalued property due to location and crime rate could be identified for government purchase to promote jobs, develop the economy, build better neighborhoods, create affordable housing, and prepare for rising sea levels brought on by climate change.

Areas of Concern Boston is an expanding city with a long history of both academia and organized crime. This mix creates an interesting political environment ripe with unseen opportunities. Local universities and the Boston city government have outlined interests in finding the best neighborhoods to expand and essentially rebuild for critical infrastructure and affordable housing. Urban planning in Boston is also somewhat difficult, with many historic sites to protect and limited land area for development. Boston homes are 2.6 times higher in price than the national average. These prices push middle and lower class income citizens out of important commercial hubs, areas which ironically need individuals from these income groups to supply a workforce.

Impact of the Problem Government purchase of land and the creation of affordable housing away from expensive coastline properties, will make Boston more accessible financially to its growing population. Former commercial, heavy commercial and industrial areas have potential for transformation into more useful properties. These properties also have higher crime rates, due to homelessness, squatting and criminal repurposing of abandoned areas. Therefore with this hypothesis the areas with high crime, further inland location and retired industrial usage will be best for government influx of capital. The creation of new resident’s will also create revenue from property tax, which the City needs to fund all types of services, including public schools.

Chosen City: Boston

Innovation related Positions and Departments The city of Boston has already published two initiatives focusing on economic development of inland properties titled “The Opportunity of Growth” 1 and “Expand Neighborhoods” 2. Both initiatives outline their policy initiatives with well thought out research and useful data visualizations. These initiatives also prove that there is incentive within the city government to fix the city’s growth problem through the use of smart city analytics. They don’t specifically specify what civic tech or datasets they would like to use for this policy concern. Luckily their outstandingly useful open data hub gives us the opportunity to create a civic tech initiative of our own. There seems to be high demand for civic data projects within the city government, both through the expansion of their Innovation and Technology Department and their hiring of a Chief Data Officer. The city government seems to understand that without the use of consistent data collection their city will be missing huge opportunities in civic growth.

Open Source Data Resources My project aims to build on this growth trend by incorporating the use of 311, crime report and property value data. This data is sourced by “Analyze Boston” Boston’s open source data hub monitored by the Citywide Analytics Team. These datasets are from 2015-2022 updated regularly with new reports. The datasets also contain longitude and latitude coordinates for exact areas for 311 and crime data. The crime dataset also has both written PD descriptions of the events, and classifications for what kind of offense was committed. Property assessment data has neighbors and street addresses, which will give us the geospatial data we need for an analysis. Property assessment data has factors such as if the building is a home, commercial or government property. It has pricing values of both the structure and the land that it occupies. Both values will be crucial in understanding what buildings can be torn down and which can be repurposed for affordable housing. Best of all the geospatial variables will be extremely helpful when creating an interactive map for the project dashboard using plotly, dash and any other useful open source packages.

Finished Product

Dashboards and Analytics The finished product will most likely be some sort of dashboard. It will incorporate interactive maps, neighborhood comparison diagrams and potentially data visualizations displaying some sort of regression analysis or machine learning findings. Not sure how I would incorporate this yet, but a classification model that classifies neighborhoods as viable investments or too expensive could be used on the pricing data. Code for the dashboard and analysis will be published on GitHub and potentially sent to the Boston Office of New Urban Mechanics. I hope to create a project that will be helpful to anyone interested in smart city analytics that wants to build their community and ensure effective urban growth.

Civic Tech Assignment

Proposal

My policy initiative proposes the use of a real time analytics dashboard to monitor crime reporting and its effect on property value. The civic tech company I found to model this initiative off is called Geolitic3, which provides law enforcement with geospatial machine learning software that creates analytics on where to send patrol cars based on crime report data. This kind of software can be repurposed for other kinds of data, including property value data. This will give city governments a better understanding of where devalued property is and what the best course of action is to create community value out of it.

Transparency and Accountability

Geolitica creates real time crime analytics dashboards that are easily shareable with community stakeholders. Heatmaps can be created to show if some areas are being over or under patrolled based on machine learning algorithms for crime hotspots. These easy-to-understand visualizations allow communities to understand how police keep their neighborhoods safe and where high crime areas are located. Geolitica manages daily patterns of patrol vehicles, so you know if officers are going to geolocations where they are most needed. These locations can be based on hotspots, intel operations or set automatically for route efficiency on the Geolitica platform. This creates a strong relationship between guidance and compliance methodologies between precincts and patrols.

Connecting Citizens and Government

There is one method of communication from a company named coUrbanize that I think would integrate very well into the Geolitica software. CoUrbanize is a community engagement platform that gives citizens an online public meeting place that even translates into their native language. Boston University4 shows that 95% of public meeting participants are white, higher income and older compared to the actual demographics of urban neighborhoods. CoUbranize’s community engagement platform allows previously marginalized citizens with limited leisure time for things like community meetings, to still contribute meaningful comments and insights on issues affecting their community. This kind of discourse would assist in getting an accurate understanding of problems affecting these communities, both civil and criminal.

Technology

Geolitica is a “software as a service” product that uses Microsoft Azure government Cloud to protect their sensitive data. From my understanding the machine learning models use some sort of supervised learning classification method to locate areas of interest. The data used comes from three types of databases. Record management systems (RMS) is used for crime data released through officer crime reports. Computer assisted dispatch (CAD) databases for collision and public safety related data. Lastly, automated vehicle location (AVL) databases for real time officer location data. RMS and CAD are for unique events, managed through addresses or latitude/longitude like the Boston open data portal does as well.

Civic Tech Conclusion

A software-based dashboard that comparatively both monitors crime reporting and property evaluations through visualizations, community maps and heat map. These will give both governments and communities an accurate understanding of where devalued properties can be turned into revitalized community assets. Once these properties are located, affordable housing plans, small business support, critical infrastructure and other important community building policy initiatives can be implemented.

Policy Literature Review

Crime and Residential Choice: A Neighborhood Level Analysis of the Impact of Crime on Housing Prices

By George E. Tita, Tricia L. Petras, and Robert T. Greenbaum 5

Empirical Analysis

This article dives into the relationship between crime (independent variable) and housing prices (dependent variable). The hypothesis is that crime is an early indicator of neighborhood transition. They use hedonic regression to quantify the effect of crime on housing prices. Hedonic regression from my understanding is a linear regression model used to predict the price of a good.

Data

The researchers collected both crime and housing data from the study’s main sample location of Columbus, OH. Crime data was provided by the Columbus Police Department (CPD). It has crime data types such as homicide, rape, robbery, assault, burglary, larceny, and automobile theft. Property value data contains `89 census tracts from 1995-1998. Characteristics of the housing data has both physical aspects of the property and the properties price value.

Results

The researchers found that the effects of crime rates on housing prices are misleading. Not only that, but they affect prices at different rates based on the income class of the community. The most interesting yet unsurprising finding was that violent crime has the greatest effect on property value.

Spatial-temporal crime predictions in smart cities: A data-driven approach and experiments

By Charlie Catlett, Eugenio Cesario, Domenico Talia, and Andrea Vinci 6

Empirical Analysis

The authors argue that as cities and the way police handle crime is becoming more complex due to growth in size and technology. The paper outlines a predictive approach using spatial analysis and auto-regressive models to detect high-risk regions to forecast crime patterns. The author’s hypothesis is that both region (independent variable) and time of the year (independent variable) can be used to predict crime rate. First step is to identify the regions with high crime density using spatial analysis. Then the crime prediction model is used for each region. Lastly the algorithm creates a spatial-temporal crime forecasting model that gives us both a summary of crime dense regions and the predictor variables associated with them.

Data

The data is of both Chicago and New York, specifically Manhattan. The Chicago data was collected on the Plenario platform, an open-source urban data resource. The New York data was collected on the NYC Opendata platform, collected by the city government.

Results

The results of this paper are not meant to give an insight on crime in smart cities, but rather show how their analytical process works. The results forecasted the number of crimes within a given urban region with high accuracy. Their conclusion shows that their methodology can be replicated, while their actual algorithms must be tailor made for each region under analysis. In future research papers they hope to implement other machine learning techniques such as spatial clustering models.

Do Affordable Housing Projects Harm Suburban Communities? Crime Property Values, and Taxes in Mount Laurel, NJ.

By Len Albright, Elizabeth S. Derickson and Douglas S. Massey 7

Empirical Analysis

This paper evaluates the claims that affordable housing developments often harm communities more than they help. Opponents of affordable housing fear an increase in crime, drop in property value and rise in taxes. To analyze these claims the authors, use a time series group control design, comparing crime rates, property values and property taxes in Mount Laurel NJ with the same variables in nearby municipalities that do not have adorable housing developments. The variables described would be the dependent variables for this controlled comparison study, while the a/b testing control and variation would be Mount Laurel and the nearby municipalities.

Data

Location of the Mount Laurel affordable housing is located adjacent to luxury, market-rate single family homes and one age-restricted retirement community. This adjacency gives the authors the percent direct comparison to similar surrounding common luxury homes. Spatial data was collected by creating a longitudinal series of outcomes for Mount Laurel and the comparison townships, before and after the opening of the affordable housing complexes. From my understanding they used a statistical Wald test so that they could create a model with multiple parameters.

Results

The authors found that adorable housing developments in Mount Laurel were not associated with crime, lower property values or high taxes when compared to the surrounding similar municipalities. This is despite what previous studies have found that use regression models to link a correlation between violent crimes and the construction of affordable housing. These previous studies suggest that location of affordable housing in areas with a preexisting history of violent crime, increases that violent crime, while affordable housing in suburban or high-income areas have no correlation with an increase in violent crimes. I suppose this suggests to us that affordable housing does not decrease property value of higher market valued homes.

Who Participates in Local Government? Evidence from Meeting Minutes

By Katherine Levine Einstein, Maxwell Palmer, and David Glick 8

Empirical Analysis

This is a research paper written by the authors of a book I am currently reading for this project called Neighborhood Defenders, an empirical study of the housing market, affordable housing, and gentrification. I will most likely use the actual book itself as a reference for my project, but I found this additional research paper by the same authors on local participation in planning and zoning board meetings for housing development. They argue that participation in board meetings is a luxury that most working-class families do not have time for. If time is money, then high income property owners have the advantage when it comes to local democratic participation.

Data

The authors compiled a data set of instances citizens spoke at planning and zoning board meetings for housing development. They matched the same individuals that spoke in these meetings to a preexisting voter file to investigate their history of political participation. These datasets resulted in a better understanding of the participatory demographics of the communities, unsurprisingly uprooting another community issue that has led to unequal participation and rising housing prices. In this analysis, time spent in public board meetings speaking (independent variable) can show us the demographics that are most likely to participate in said meetings (dependent variable), allowing us to unearth the underlying pattern in civic participation.

Results

The authors concluded that most individuals participating in the board meetings were older, male, longtime residents, voters in local elections and homeowners than renters. These individuals also opposed new housing construction, which in turn has resulted in the rising of housing costs leading to further participation inequality. I found this research paper to be an insightful and creative method of analyzing the social phenomena effecting the rise of housing costs within communities.

Part I-APIs and Open Data and Smart Data

There have already been previous policy initiatives written that identify potential growth neighborhoods within the city of Boston. My initiative aims to improve upon this research with variables not previously considering, in this case that would be the incorporation of crime data into the evaluation of property value. Both datasets have some form of geospatial indicators, whether that’s the street address or longitude and latitude. Using these with other value indicators my solution is a geospatial dashboard that help us see the effects of high crime police districts on property and land value.

Crime Data

Crime data9 is collected by the boston police department. The data is collected daily, updating with the most recent 2022 crime reports. For the purpose of this analysis we will be sticking to the completed 2021 data file, since the timeline matches up with the completed property data. The data can be downloaded in CSV, TSV, JSON or XML format. The main variables of interest are descriptions of the offenses and their locations. Below are the specific variables that will be selected in R when wrangling the data.

library(readr)
bcrime<-read_csv("https://data.boston.gov/dataset/6220d948-eae2-4e4b-8723-2dc8e67722a3/resource/f4495ee9-c42c-4019-82c1-d067f07e45d2/download/tmp7_f32p54.csv")
## Rows: 71721 Columns: 17
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr  (6): INCIDENT_NUMBER, OFFENSE_DESCRIPTION, DISTRICT, DAY_OF_WEEK, STREE...
## dbl  (8): OFFENSE_CODE, REPORTING_AREA, SHOOTING, YEAR, MONTH, HOUR, Lat, Long
## lgl  (2): OFFENSE_CODE_GROUP, UCR_PART
## dttm (1): OCCURRED_ON_DATE
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
#Gives us a simple view of first 10 values within one of our indicators.
head(bcrime$DISTRICT,10)
##  [1] "B2"  "B2"  "B2"  "B2"  "E13" "C11" "A1"  "E13" "B2"  "A1"

Crime Indicators

  1. Offense Description describes the type of crime committed in the offense

  2. Shooting notes if the offense was a shooting and how many shots were fired.

  3. Offense Code is a simple foreign key used to identify the type of offense.

  4. Incident Number is a simple primary key to identify the specific offense, not its type.

  5. Police District is the jurisdictional region where the offense took place.

  6. Longitude and Latitude is used to identify the geospatial location of the offense for mapping purposes.

Property Data

Property data10 is collected by the Boston Assessing Department. The data is collected yearly, but the most recently completed submitted data is from 2021. The data can be downloaded in CSV, TSV, JSON or XML format. Types of land use vary from residential, commercial, condominium, agricultural, etc. The main variables of interest used in my analysis are the property value indicators and the address variables.

library(readr)
bhousing <- read_csv("https://data.boston.gov/dataset/e02c44d2-3c64-459c-8fe2-e1ce5f38a035/resource/c4b7331e-e213-45a5-adda-052e4dd31d41/download/data2021-full.csv")
## Rows: 177091 Columns: 63
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (45): PID, CM_ID, GIS_ID, ST_NUM, ST_NAME, UNIT_NUM, CITY, ZIPCODE, LUC,...
## dbl (18): BLDG_SEQ, NUM_BLDGS, RES_FLOOR, CD_FLOOR, RES_UNITS, COM_UNITS, LA...
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
head(bhousing$TOTAL_VALUE, 10)
##  [1] "$719,400.00"   "$744,800.00"   "$730,500.00"   "$667,900.00"  
##  [5] "$714,200.00"   "$1,181,900.00" "$1,151,600.00" "$1,045,200.00"
##  [9] "$804,000.00"   "$358,700.00"

Property Indicators

  1. Land Value shows us the value of the land the property is located on.

  2. Building Value shows us the value of the physical structure on the property.

  3. Total Value is land and building values. Together we can divide by total value. This can show us which holds the higher percentage of value for a given residential non apartment based property.

  4. Gross Tax shows us the government revenue the city government collects in property taxes.

  5. Living area is the area in square feet that is actually habitable within a residence, meaning space not taken up by physical structure of the house or building.

  6. Overall Condition is the rating of the property through categories.

  7. Address helps located the property for geospatial purposes.

Part II-APIs and Open Data and Smart Data

Crime Data Visualizations

Recording Police District as a Smart City Services indicator allows us organize locations categorically in a simpler way than actual address, neighborhood or community. This makes both spatial data easier to organize, and one variable visualizations like this easier to display. This bar chart hopefully gives us an understanding of where in the city of Boston the most police reports are being filed. In future research I will add more cotnext to where these police districts are actually located within the city.

Property Data Visualizations

The two indicators I choose for the property data fall under the Smart City Services category. The two indicators I choose for the visualization are the recorded living area and total value for single family homes. I filtered my data to only single family homes, because it is easier to compare uniformly through the above visualization. Single family homes are not representative of the larger housing availability in the city of Boston, but they do help narrow down the scope of this specific assignment Living area is important because it shows us how much space in the property is actually habitable, which is important when understanding the quality of life of a cities residence. Overall home value is important from a fiscal standpoint, understanding how much the market values a particular property. Together these two indicators are important because they show us how strong the correlation is between living area and property value. Hopefully this visualization brings some insight into the data I have available. If not, any readjustments or suggestions moving forward would be appreciated.

Policy Postition

Unfortunately, in the United States access to housing is not a right. You are not guaranteed a home to rest at night, or a safe place to raise your children. Marginalized communities remain marginalized, and the rich stay rich. Sadly, after a while homeless encampments surrounding our cities begin to just fade into the landscape. The worst part of it is that financial institutions want this status quo to persist, because it guarantees their position as the dominant social class within our ironically democratic society. In fact, in recent years, we have seen an acceleration of housing property being used as an investment method by Hedge Funds and Investment Banks. Houses have become assets to keep on the books, left empty, rather fulfilling their intended purpose of providing shelter. On top of this, we have witnessed collateralized debt obligations (CDO) turn mortgages into a destructive investment weapon that destabilized the world economy. There is a deep need in our nation’s cities to understand how to better utilize housing properties, and where there could be potential for construction of affordable housing.

As single-family homes become harder to purchase for average Americans, we must figure out unconventional methods to identify opportunities to create affordable housing. Luckily thanks to city government innovation offices we have access to open data that can be used for affordable housing data analysis. The Assessment Department within the city of Boston for example gives us a dataset of around 32,000 observations of single-family homes, with their exact locations, quality ranking and their total value. This source is rather conventional when it comes to housing policy analysis, so to create a new insight I believe we need to pair it to another rather unconventional dataset. The Boston Police Department crime data base is one unconventional data source that pairs nicely with Assessment Department housing data. The Boston Police Department datasets give us both descriptions of the crime and exact coordinates through longitude and latitude. The Assessment Department data on the other hand contains street addresses, the city, the zip code, and the state but no longitude and latitude. We can change these addresses to coordinates through the help of google cloud platform (GCP) and the GGMAP package in R. Creating an API key in GCP allows me to insert it into R studio and convert addresses into coordinates using the GEOCODE function. Pulling requests of this size was time consuming, but in the end, we are left with a dataset both describing the value of single-family homes and their exact geospatial locations. Together with both the crime and housing assessment database we can create digital maps and dashboards that give us an understanding of where undervalued property in high crime areas are located. With this data we can understand what current conditions are and how we can best empower the government to take back the affordable urban housing market.

Opposing arguments to this policy position might suggest that this policy is just another way to identify new ways to gentrify underprivileged neighborhoods. This might be true if these communities were targeted by private entities, but in this case, we are advocating for homes to be rejuvenated for affordable housing options and other city government run housing initiatives. This policy position is advocating for the analysis of housing opportunities to increase these communities’ quality of life, not push them out through rent hikes. There are 3 main types 11 of affordable housing options in the city of Boston. First there is subsidized rental housing where you pay a set percentage of your income every month, meant for individuals at very low to nonexistent income. Sadly, with this option the demand out ways the supply with over 10,000 12 new applicants every year and a limited number of available homes. Second option is income restricted housing that is limited to individuals under an income threshold. These income restricted properties can be for either sale or rent, giving potential applicants some leeway in their options. Lastly, we have voucher programs, where individuals are given a certain amount of rent to pay for rent in private apartments. There is only one voucher program in Boston currently receiving applications which is the Veterans Affairs HUD-VASH vouchers. Out of these 3 options subsidized rental housing seems the best option, but again with the extreme demand the government needs to find undervalued property that can be used by the program.

In summation, I am advocating unconventional methods of analysis such as the effect of crime on housing prices, to identify urban regions of opportunity for city government affordable housing initiatives, specifically subsidized rental housing. Unconventional methods are the only way city governments will be able to beat the hedge funds and investment banks that hold limited properties hostage; spiking increases of monetary value in the housing market. Pricing working class Americans out of their own neighborhoods is fundamental. Protecting government funded subsidized rental housing in low-cost high crime neighborhoods is the only way that is fiscally realistic to meet the extreme demands of affordable housing applications.

Google Data Studio

(https://datastudio.google.com/embed/reporting/40ee0858-46ef-404b-8f8e-6c595172130f/page/uFmxC)

As we can see from this visualization some neighborhoods have higher gross area, total value and sheer number of properties than others. The two categorical dimensions I added to the scatter plot were neighborhood and year built. By adding year built I got a better sense of the age of the city, and what decades had the most number of both low and high value properties. As we cn see there is a clear positive correlation between gross area and total value presented by the scatter plot. The tree map on the bottom was just to give a larger perspective on the record count of single family homes in different neighborhoods. Space is a high priced commodity in overcrowded cities, a cost of which affordable housing initiatives will help lower.

Policy Position’s Geospatial Data Visualizations

As we can see from these two maps there are a shortage of single family homes in police districts A1 and D4, but with exceptionally high crime rates. We can also see that the single family homes in these districts are relatively on the low end of price compared to the neighboring policy district E13. While these maps are clearly not perfect I think they give use a baseline understanding of where the areas of opportunity are within the city of Boston. The final product for this policy initiative will most likely be a leaflet or ggplot map, not ggplot2.

Spatial and User Generarted Data

The above map of Boston is of the 10 police districts within the city. The blue dots are for the locations and pricing of single-family homes by the assessment department in 2021. The yellow dots being locations of assaults recorded by the Boston PD also in 2021. The shading of blue dots is lower to higher pricing, with lighter blue being lower cost and darker blue being higher. The yellow dots are both simple and aggravated assault. I filtered out only assault data because all the crime data was too large to load into Mapbox, also assault data seems like it would have a significant impact on neighborhood housing value. As we can see from the map some districts have more yellow assault dots with lighter blue home dots, specifically B2 and B3 police districts. E18 also has lower housing costs but less assault records, indicating that there a plenty of lower priced single-family homes further in land from the cost. B2, B3 and E18 make a sort of corridor where homes are cheaper, further inland and have higher crime rates, all of which are factors that create an opportunity for the city government to buy single family homes for affordable housing projects. On the other hand, Assault location is not a perfect indicator, because police could be biased on where they patrol due to racial and ethnic neighborhoods being perceived as higher crime. A better indicator might be theft locations, because those are more likely to be reported by victims rather than witnessed on the street. In the next map we will view the same housing data, but this time with theft locations on top.

The green dots on the map above represent locations of reported theft in the city of Boston in 2021 by the Boston PD. As we can see the locations of reported theft are very similar to the locations of assault depicted in the first map. There does seem to be more reported cases in police districts A1 and D4, but there are few single-family homes reported there, therefore less likely for there to be homes on sale at reasonable prices for city government purchase. I would make the same conclusions as with the first map, that the most affordable opportunities for affordable housing purchases are with the B2, B3, and E18 corridor. I am very surprised to find the maps having such similar clusters. I assumed because the nature of the crimes they would have different target locations, but this was not the case as seen in their mirrored similarities.

Metrics for Outcomes

Average Single Family Home Values

First metric used would be the average single family home cost (under 1 million) within each police district within the city of Boston. To create this metric we would stratify the single family home total values according to their police districts. We would then take the average of each one of these strata, creating 10 average home values. Now that home values are created into averages we can use this to possibly create a choropleth map, rather than a dot map.

  1. How much service was delivered?

The service delivered was the assessment of home values for every single family home, apartment building, and other residential buildings within the city of Boston.

  1. How well was it delivered?

The dataset on home assessment value was very well delivered. It has thousands of values with almost every category completely filled out and usable for analysis.

  1. How much change was produced?

The change it produced was giving the public a consistent standardized way to value homes within the city of Boston. This would be much more difficult if it was done by a private company or entity, which may be intrinsically biased for personal gain.

  1. What was the quality of change produced?

The quality of the change was high. The home values seem to be consistent based on their locations. Only problem with the assessment data is that some homes recorded are not within the actual city limits of Boston, making them irrelevant and hard to get rid of if a non geospatial analysis is done.

Civic Tech Accessibility

Adding to the choropleth map we would use the “Viridis” package in R for the shading scale of the map. Viridis is a color blind friendly R package, it is a common equitable standard to use when teaching color scaling in R. I had a colleague at a previous job who didn’t let the other data science staff know he was color blind due to personal embarrassment. Luckily, long before I knew of his disability I had been using the viridis package for all the choropleth maps I produced in R. He was able to follow along without having to talk about his disability before he felt comfortable sharing that information. I have severe allergies to several types of food, which I sometimes do not share with team members before team lunches due to fear of jokes. As someone with a chronic condition, it feels good to be accommodating to others who fear the same “fun poking” that usually comes with expressing your weaknesses.

Crime Count for Theft and Assault

The second metric is just a sum of all the counts of theft and assault individually per police district. This will give us a direct value to match with the home value data so the end dataset result would be the names of the 10 police precincts, 10 single family home averages, 10 theft counts, and 10 assault counts. With this new data set comparative analysis will be easier, using both data visualizations and linear regression. These new visualizations will make it easier to have a one-to-one analysis of crime’s effect on home value. The mapbox maps will remain using the original data, because it gives us a good understanding of spatial clustering that is still useful.

  1. How much service was delivered?

The service of data collection was delivered every time a police report was filed by the Boston PD.

  1. How well was it delivered?

The quality of which the dataset was delivered is very good, all categories are completely filled out and consistent. Only thing I would change would be better descriptions of how the individual events occurred, so that we could do some type of text mining process such as a word cloud, to understand how the police perceive each individual reported crime.

  1. How much change was produced?

I think there is an immeasurable amount of positive change from recording police reports in an open data set for the public. It allows average citizens to understand what is happening in their communities and how their local police describe local crimes committed. It also gives citizens a data supported way to learn what areas are safe and what areas are dangerous within their communities.

  1. What was the quality of change produced?

Quality of the change seems high. Having the power to view police reports through easy open accessibility has immeasurable value for communities. It allows us not only to know what areas are safe or not, but also if police are overpatrolling a certain district due to racial or demographic biases. This gives the citizens an understanding of how and where police interact with their communities.

  1. “The Opportunity for Growth .” Imagine Boston 2030 . Boston Government, n.d.https://www.boston.gov/sites/default/files/embed/i/imagine-boston-opportunity-of-growth.pdf.↩︎

  2. “Expand Neighborhoods.” Imagine Boston 2030 . Boston City Government , n.d. https://www.boston.gov/sites/default/files/document-file-06-2018/taking_action_-_expand_neighborhoods_1.pdf.↩︎

  3. “Public Safety.” Geolitica. Accessed June 14, 2022. https://geolitica.com/public-safety/. ↩︎

  4. Einstein, Katherine Levine. “Who Participates in Local Government? Evidence from Meeting Minutes.” Politics of Housing . Department of Political Science at the University of Boston , June 29, 2018. https://www.politicsofhousing.com/research/who_participates_in_local_government.pdf.↩︎

  5. Tita, George E. “ Crime and Residential Choice: A Neighborhood Level Analysis of the Impact of Crime on Housing Prices.” Springer. Springer Science+Business Media, Inc., July 28, 2006.↩︎

  6. Catlett, Charlie. “Tita, George E. ‘ Crime and Residential Choice: A Neighborhood Level Analysis of the Impact of Crime on Housing Prices.’ Springer. Springer Science+Business Media, Inc., July 28, 2006. .” Elsevier, January 17, 2019.↩︎

  7. Albright, Len. “Do Affordable Housing Projects Harm Suburban Communities? Crime, Property Values, and Taxes in Mount Laurel, NJ.” City & Community. American Sociological Association , June 12, 2013.↩︎

  8. Einstein, Katherine Levin. “Who Participates in Local Government? Evidence from Meeting Minutes.” Housing Politics Lab. Department of Political Science Boston University, June 29, 2019. https://www.politicsofhousing.com/.↩︎

  9. “CRIME INCIDENT REPORTS - 2021.” Analyze Boston . Boston Police Department , August 2021. https://data.boston.gov/dataset/crime-incident-reports-august-2015-to-date-source-new-system/resource/f4495ee9-c42c-4019-82c1-d067f07e45d2.↩︎

  10. “PROPERTY ASSESSMENT FY2021.” Analyze Boston . Boston Assessment Office, April 2021. https://data.boston.gov/dataset/property-assessment/resource/c4b7331e-e213-45a5-adda-052e4dd31d41.↩︎

  11. “Affordable Housing Options.” Affordable Housing in Boston . City of Boston , November 18, 2020. https://www.boston.gov/affordable-housing-boston#where-to-start.↩︎

  12. “Housing Availability.” Boston Housing Authority, 2022. https://www.bostonhousing.org/en/For-Applicants/Do-I-Qualify-for-BHA-Housing.aspx.↩︎