Recently, I do a research about mode choice behavior. Since I am a graduate student of transportation engineering, there are many many topics can be conducted in a transportation research such as freight transport, travel behavior, road pricing, safety riding, etc. I found myself very keen with travel behavior topic.
What is travel behavior?
It is a study that discuss how people (would) act towards several conditions that happened in transportation, especially human transport. Most of the study needs a primary data which means the data should be collected directly from the resources. The resources are, of course, people.
One of the form of travel behavior is mode choice. The data in mode choice study can be divided into three main groups which are socio-demographic data, utility data, and compiled data between both. The output could be a model of mode choice, variable definition towards the mode choice propensity, or specific relation between variables that influence the mode choice.
The main challenge is the method of data collection. If the data sampling has no robust reason or exact boundary, the data could be failed. It could not build a correct model, even it would not describe anything. The data needs to fulfill some statistical requirements before it could be developed to be a model. Moreover it is also needs to have a realistic and logic relationship in the real world (not just theory), so it can be applicable.
The study object is people. It is obviously not quantitative since the values depend on people's opinion and perspective. Sometimes I ask myself: why am I doing this? Why don't I do something more scientific like people do in other engineering field?
But it makes me feel more human (what am I, a cat? :p)