Lab Four

The impact of NYC museums on neighboring property sales and rental prices

With over 130 museums located in New York City, I was curious to know if museums have an impact on economic development such as on real estate and tourism. I framed my research question as: Does the proximity to a museum affect property sales and rental prices nearby?

The four datasets I have used for my analysis include:

1. NYC Museums, 2018

Longitude & Latitude of NYC museums

2. NYC Census Tracts, 2010

A polygon layer with land-based boundaries of census districts and neighborhoods of NYC

3. DOF: NYC Real Estate Sales, 2003-2018

This Geodatabase is comprised of point layers created from Detailed Annual Sales Reports by Borough published by New York City Department of Finance. It represents locations of all properties sold in New York City between the years of 2003 - 2018.

4. NYC Airbnb Open Data, 2019

This dataset describes the Airbnb listing activity and metrics in NYC for 2019.

The datasets were cleaned on OpenRefine, uploaded as shape layers into QGIS, then visualized on ArcGIS. Some of the detailed steps I took to clean and join the data are:

1. Separating museums geocodes to longitudes and latitudes so a point layer can be created in QGIS
2. Changing the Coordinate Reference Systems (CRS) of layers to make sure they all match
3. Joining real estate data with census tract to get an aggregated count of properties using count of shapes in polygon analysis in QGIS
4. Joining Airbnb data with census tract to get an aggregated count of Airbnbs using count of shapes in polygon analysis in QGIS
5. Joining real estate data with census tract to get an aggregated mean price of property sales in QGIS
6. Joining Airbnb data with census tract to get an aggregated mean price of Airbnb rentals in QGIS
7. Creating a new real estate layer with columns on aggregated count of properties and aggregated mean price of property sales by joining the census tract id
8. Creating a new Airbnb layer with columns on aggregated count of Airbnbs and aggregated mean price of Airbnb rentals by joining the census tract id

A final map was then created on ArcGIS. The first layer (black dots) shows the location points of NYC museums and their names. The second layer (pink) is a choropleth map of the average property sales in NYC areas. The third layer (green) is a choropleth map of the average Airbnb rental prices per night in NYC areas. These three layers will give us insight into how property sales and rental prices vary when considering the proximity to museums.

Some insights include:

• The average property sales and Airbnb rental prices are higher in Soho, Midtown, and Upper East Side. These are also the areas where most museums are located.
• Museums tend to be found in clusters in Manhattan.
• The most expensive area of properties sold was in Midtown-Midtown south, with an average price of $2221,000,000 for the two properties. This is also where the Discovery Times Square Exposition and International Center of Photography are located.
• The most expensive area of Airbnb rentals were in Bayside, with an average price of $1,338 per night for the two Airbnbs. No museums are located in the vicinity. However, the second most expensive area is Upper East Side-Carnegie Hill, with an average price of $1,099 for the four Airbnbs. This area is where the Frick Collection, Asia Society, American Society, and the Museum of American Illustration are located.

Conclusion

Even though museums are also found in less expensive areas, but I would say that for the most part, there is a positive correlation as museums located in Manhattan contain some of the most expensive property sales and rental prices.

However, there are many other factors that could influence the local economic development as well. Perhaps, a more effective study is to measure the change in rates before and after the museum had opened, and compare it to properties and Airbnbs further away but still in the same neighborhood. This would require analyzing a few case studies than looking at all 130 museums at once.