Quality of Social Life Comparison in Wilkinsburg, Hazelwood and Homewood

Intro

The goal of the research is to explore social life in three communities within Pittsburgh Metro Area -  Wilkinsburg Borough, Hazelwood and Homewood neighborhoods. These communities are perceived as lacking of social life due to many reasons, such as lacking of amenities. Social data could reflect certain degree of quality of social life in these communities.

Approach 

After selecting the communities, I defined three criteria, including accessibility to community amenities, intensity of usage of certain amenity spots and finally social generosity(please see below), to measure the quality of social life.

Dataset

The data for analysis is from Foursquare's venue data, the original datasets were provide from professor Daragh. The final data for analysis purpose was filtered by geolocation. All venue data within Wilkinsburg Borough, Hazelwood and Homewood were exported as shapefile from ArcGIS Pro for importing use. 

Additionally, datasets like venue popularity rating, venue attribute are joined with original dataset for exporting as shapefile.

Process

After all datasets are geocoded with longitude and altitude data, it could be packaged and imported into Carto. 

Accessibility to Local Amenities

The selected amenities include food services, night life spots, shops, art & entertainment place, and outdoor & recreation place. The data in this part was only used to visualize the geolocation of amenities and see where the amenities are concentrated in the selected communities.

Intensity of Usage

Using the user count data to measure the intensity of usage at commercial and leisure sites in the communities. Additionally, using popularity rating data to create an underlay layer to explore whether there are any other popular spots not captured by commercial spots, art and entertainment, and outdoor recreational sites.

Social Generosity

Similar with previous analysis but changing the underlay layer with tips generated at the sites, which I assume could reflect the emotion connection during the service exchange process, such as dining, shopping or other services, then the tip behavior could reflect the appreciation of the services and quality of services.

Insights

Three communities has great neighborhood assets, which generated great population flow at the certain parts of the neighborhood, and majority of the activities happen in the main street district and major arterial streets in the communities. 

Limitations

1. The data wasn't filtered correctly by average period of time, which only reflects the data after Foursquare's release. Also, some spots are opened within a short period of time, so may cause some variation on the visualization.

2. Due to limitation of Carto's heat map ability, the underlay layers were not working very well as an actual heat map.

3. Also, many of the activities might be generated from outside visitors, for instance, a heavy percentage of commercial activities along Penn Ave in Wilkinsburg were generated from passing by traffic as many auto oriented commercial establishments were designed for them. As a result, the quality of amenities for community oriented social life might not be very accurate.

4. For Generosity, it should include amount of tips or percentage of total bill would be more accurate.