People often make social-related trips to perform activities with their friends. An individual's group of friends can be characterized by his or her social network. While traditional social network data collection is time-consuming and dependent on memory recollection, new online social networking sites may address these shortcomings. This research focuses on the use of tie-strength, the strength of an individual's relationships in his or her social network, to characterize friendships and how this influences an individual's air travel behavior. Four candidate weighting schemes were developed using data collected from a web-based survey which included demographic information, an air travel diary, and friendship information retrieved from Facebook.com. The candidate weight matrices were then tested in a spatial Durbin count model (social model). The results of this study are threefold. First, candidate weighting schemes which consider mutual friendship (i.e. the number of mutual friends two people have in common) exclusively produced higher log-likelihoods than weighting schemes which also consider whether individuals are direct friends (i.e. whether the two individuals are friends themselves). Second, the results of the social model were compared with those of a non-social model. These results suggest that there exist major flaws in using a non-social model to represent variables which may be socially dependent and correlated. Finally, results suggest that individuals tend to have friends who, on average, make more trips than they do. With a growing number of people using online social networks, exploring and understanding friendship influences on travel behavior will help the transportation industry better recognize future travel needs.
Identifer | oai:union.ndltd.org:GATECH/oai:smartech.gatech.edu:1853/52312 |
Date | 27 August 2014 |
Creators | Zhang, Bingling |
Contributors | Watkins, Kari E. |
Publisher | Georgia Institute of Technology |
Source Sets | Georgia Tech Electronic Thesis and Dissertation Archive |
Language | en_US |
Detected Language | English |
Type | Thesis |
Format | application/pdf |
Page generated in 0.0022 seconds