The dark side of tourism: investigating the correlation between airbnb listings and their impact on crime Rate
Tourism is ramping up in many countries around the globe. Recently, researcher A. Liowska in 2017 brought together several research experiments on the the subject of crime and tourism. Conceptually, she argues a logical relation between tourism and presence of crime either done by or to tourists. However, there is insufficient research on how criminalistic statistics are related to this phenomenon.
Our team is intrigued in conducting a similar analyses and further contributing to this field by researching data retrieved from Airbnb and the Central Bureau of Statistics in the Netherlands. With our research we attempt to answer the following question:
"To what extent does the number of Airbnb listings affect crime rates within Amsterdam, and to what extent is this relationship different within different Amsterdam neighbourhoods?"
Read the generated research file by going to gen/analyses.
With the data collected by the team and the analysis performed, the following (abridged) findings have been obtained:
- The main hypothesis is valid, as there is a positive correlation coefficient between the Airbnb listings numbers and crime occurances.
- The moderator hypotheis is somewhat valid, as 9 out of 20 Amsterdam neighborhoods display significant differences in the relationship between listings and crimes (although this may need to be explored further in a follow-up study).
- The R squared coefficient of the model is low (0.1104) - this means that a low percentage (11.04%) of data is explained by this model, and that the results should be taken skeptically.
├── README.md
├── makefile
├── .gitignore
├── install_packages.r #checks and installs (if neccesary) required packages on the users' machine.
├── gen #stores the PDF version of the analysis with the motivation
├── temp-data #stores temporary data, while running the make file this will be deleted
│ ├── analysis
│ ├── raw
│ └── summarized
└── src
├── cleaning #cleans the repository of temporary data by deleting the 'temp-data' folder once the analysis output is created.
├── data-analysis #stores the code for the analysis with the motivation
├── derived #stores the code for the Airbnb listings, the crime data and the joint dataset
└── download #stores the code for downloading the Airbnb dataset and the crimerate dataset and stores the make file
To run the file you must have installed the following programs:
- R and R-studio
- Make
- Git Bash (windows user) or terminal (mac user)
- pandoc is needed to convert a markdown file to PDF
The R code contains libraries that need to be installed in order to run the code.
By running the install_packages.r code, which is included in the main folder, all needed packages will be installed automatically. This script is included in the makefile. Therefore, libraries will be installed/updated automatically.
With make installed, it is possible to open a terminal (or Git Bash) and run:
make
This runs the entire project. To run certain parts of the project, it is possible to run the specific make targets specified in the makefile. for example:
make [TARGET ...]
Disclaimer: Keep in mind that it can take a while to download all data. During testing it took approximately 3 minutes to run the download part.
- Airbnb data used available at Inside Airbnb
- Police data used available at Data.Police
This is the repository for the course Data Preparation and Workflow Management at Tilburg University as part of the Master's program Marketing Analytics, used for the team project of group 8.
- Cas van Dijk, c.c.j.m.vandijk@tilburguniversity.edu
- Chokie Tang, c.k.tang@tilburguniversity.edu
- Rob Esenkbrink, r.m.esenkbrink@tilburguniversity.edu
- Mihai Vlad Serbanescu, v.serbanescu@tilburguniversity.edu
- Anoesjka Raateland, a.raateland@tilburguniversity.edu