Preface

Does response time have a significant effect on ratings? What are some popular description words that lead to a higher demand in availability? What is the relationship between geographical location and average price? ‘

All of these questions can be answered through Inside Airbnb’s comprehensive database on the Airbnb listings and reviews of cities around the globe. The site’s data sources provide information on current and historical price, availability, ratings, review sentiment, and more.

Based on our findings from the dataset, our final project will focus on comparing the Airbnb data between Paris and Los Angeles within the past 90 days, two large cities that have similar statuses in attracting tourists. We evaluated variables such as pricing, minimum nights, reviews, neighborhood, and room types from the Inside Airbnb dataset to evaluate the variability between travel patterns, cyclical booking patterns in a given year, and the dynamics of Airbnb clustering within the two cities. We have hypothesized that given the prominent tourist status of both cities, we may find similar patterns in the Airbnb rental availability and booking patterns — however our data has instead shown large variability. To reach this conclusion, we did extensive research on the Inside Airbnb dataset, which yields insights towards the customer perceptions of Airbnbs, travel patterns in booking Airbnbs, as well as mapping visualizations of where the Airbnbs are located relative to the popular city attractions. This dataset comes directly from a collection of Airbnb data, and thus helps us to gain valuable insights into how and why customers choose among Airbnb properties and hotels as well as how such behavior influences the hotel industry in two big cities. 

Since the rise of Airbnb in the lodging industry, scholars have been interested in determining the influence of Airbnb on traditional hotels. Many studies have aimed to investigate the causes of people choosing to stay in Airbnb properties instead of hotel rooms from the perspectives of customer behavior patterns, special Airbnb attributes, and environmental factors. Using various statistical methods such as weighted regression, sentiment analysis, and unsupervised LDA model, these studies have shown that neighborhood elements, host behaviors, and physical characteristics of Airbnbs are all factors that contribute to the preferability of Airbnbs over hotels (Ding, Kai, et al, 2021; Ju, Yongwook, et al, 2019; Xu, Feifei, et al, 2019). Furthermore, other studies have focused on quantifying the impact of Airbnbs in the lodging industry. Using the Difference-in-Difference approach and time series analysis, these studies have found that even though Airbnbs became more and more popular over the years, they have not diminished the hotel industry by a significant amount, especially the well-established hotel brands (Byers, John, et al, 2017; Heo, Cindy Yoonjoung, et al, 2018). These studies have provided us with ample evidence that there exists a strong relationship between the various aspects of Airbnb and the behavioral patterns of Airbnb customers. However, although most studies used one or more cities as their case study, there is surprisingly little literature focusing on systematic comparison between cities. Thus, in our project, we chose to focus on two major tourist cities, Los Angeles and Paris, that are similar in their Airbnb industry and tourism, comparing and contrasting the above mentioned aspects of Airbnb properties and customer behavior in relation to the impact of Airbnb on the lodging industry in general. 

 

Why Our Project is Significant

This topic is important as it looks at Airbnb activity in two major cities and how key components, such as location, room types, and minimum stay, compare between these two major tourist destinations. Contrary to prior assumptions that these components may be similar between the cities, our data actually showed significant variability. For example, while both cities display clusters of Airbnb rentals around popular neighborhoods, the cluster in Paris spans for over 8 miles while the clusters in Los Angeles are much smaller. However, Los Angeles has multiple pockets of major Airbnb rental activity while Paris only has one major cluster. These results could also be explained by geographical differences between the two cities. Whereas Los Angeles sees a scattered Airbnb rental activity, it also has a large amount of coastline in which many listings are congregated. Paris does not have this natural advantage, and thus could lead to an influenced consumer behavior to congregate towards major attractions. These results can connect to the larger question of why and where a customer chooses to rent out an Airbnb. In Los Angeles, a customer may need to be more particular in their area of stay while a customer in Paris has a greater selection across a large, singular region. Another key difference is the type of rooms that customers book in each city based on their average minimum nights stay in the city. In LA, the minimum nights stayed at each type of room is roughly consistent (between 6-24 nights). In Paris, however, there is a segment of customers who rent an entire home or apartment for a long period of time, creating a greater average minimum nights stayed between 3-110 nights. This data allows customers to have a better understanding of what type of room to select for their length of stay. It also communicates that long stays in Paris through Airbnb are a viable option that is less so in Los Angeles. This could be due to the infamous “nomads” seen on social media and other news outlets, who travel whilst working remotely after the COVID-19 pandemic created a work from anywhere culture, or it could be due to other phenomenon/user trends.

Overall, these insights not only help customers in their rental strategies and selection choice, but also helps Airbnb understand their market in each city and better cater to different travel purposes and demands. Ultimately, understanding patterns within customer’s traveling patterns and Airbnb bookings unveils the many drivers that cause people to travel but also the influences that outsiders travel can have. 

 

Research Question

1.  How does word choice in listing descriptions influence the likelihood of a customer renting an Airbnb?
2. What room type do customers prefer to rent in LA and Paris?
3. Is there a relationship between neighborhood popularity and the average prices in these neighborhoods?

Overall, we are investigating relationships between the different variables in the Airbnb dataset to answer the overarching question: What do consumers look for in an Airbnb?

By understanding this information, we can inform others, specifically hosts, about the characteristics that make a successful Airbnb. We can inform them what words to use in their listing descriptions to make their listings much more attractive. Moreover, we can suggest what room types tend to be more popular, where the popular neighborhoods are, and at what prices hosts in these areas tend to lease their listings. Ultimately, we are aiming to help hosts increase their chances of leasing out their Airbnb.