In previous weeks, manifest variables of residential and public unit square foot-per-person (USFPP) were created to measure neighborhood overcrowding. An OLS regression analysis of residential USFPP suggested that census tracts with higher percentages of homeownership and median assessed value percent change tend to have larger amounts of residential space, holding all other factors constant. However, … More Residential and Public USFPP and Neighborhood Segregation
This week I began exploring the relationships between variables from the 2016 Tax Assessor dataset. My measures are concerned with characteristics of residential areas of Boston and the relative presence of “cultural” institutions: churches, schools, colleges, or libraries. I began by testing correlations between measures of cultural parcels and those of residential ones. TADCT2 <- … More Cultural sites in residential areas?
In previous weeks, manifest variables of residential and public unit square foot-per-person (USFPP) were created to measure neighborhood overcrowding. Yet the question remains as to how these two variables relate to other measures of neighborhood characteristics. For this week’s assignment, variables from the longitudinal and cross-sectional Boston Tax Assessor’s database were selected based on a … More Predicting Residential USFPP
To continue making sense of the city of Boston using data from the Tax Assessor department, we can work to make the data more useable by creating measurements out of the data. This process can make order of administrative, passively collected datasets and help us better understand the urban social world. In this case, I … More Creating a measure of neighborhood homgeneity
As we continue to work with 2016 data from the Boston assessment department, we can begin to create new variables to make better sense of the city of Boston. The first step I took was to calculate a variable that gives the total value per square foot of a property, by adding together the … More New variables of the 2016 assessment data
As we continue our initial exploration of Boston’s 2016 Assessor Department database, we can look for the most interesting categories of data. To begin, let’s look at the initial year of construction for buildings in Boston. Using the summary command in R, we can gain some first insights: > summary(TAD$YR_BUILT) Min. 1st Qu. Median … More Pulse of the City through Boston 2016 Assessment Data
This semester I will seek to gain insights into the city of Boston, its residents, and its governance, by conducting analysis of a set of “big” data. These data will come from the 2016 database of the City of Boston’s Assessing Department. These data include specific points on a wide variety of data categories about … More Beginning a data story with the 2016 Tax Assessor database
This week we’ll put R’s inferential statistics functions to work. Our question is: “Do areas with different building value dispersion exhibit different levels of ethnic diversity?” To begin, we’ll load Tax Assessor’s and Census data: TAdata <- read.csv(“TAdata.csv”) demographics <- read.csv(“Tract Census Data.csv”) Now we’ll estimate a building value dispersion index, intended to represent how … More Do areas with different building value dispersion exhibit different levels of ethnic diversity?
This week the effort is to analyze the Energy Efficiency Index (EEI) using t-test and ANOVA. As the index is built for residential units only, a comparison across land use type will not be meaningful. Hence I studied the EEI mean across planning districts and compared it with average building value per square feet. EEI … More Energy Efficiency Index – t-Test and ANOVA
For this week’s data assignment, I looked at the statistical relationship between the most common decade built for a parcel and it’s distance to Boston Common. This is interesting to me because I have noticed a “concentric circle” temporal pattern in development in Boston (i.e. for the most part downtown areas have older buildings) and … More Examining the Relationship between Most Common Decade Built of Geographic Aggregates and their Respective Mean Distances to Boston Common