This study presents an integrated model to shed light on the factors influencing individuals’ likelihood and frequency of usage of bus transit in Bengaluru, India, with a focus on the role of individuals’ subjective perceptions of service quality. Typically, subjective perceptions of transit service characteristics such as comfort, cleanliness, reliability, and safety are measured using Likert rating scale questions in travel surveys. A shortcoming with many such surveys is that the Likert rating scale questions do not include a “don't know” response category for the respondents to express their unfamiliarity and lack of opinion on the transit service. For this reason, some respondents who are not familiar with and do not have an opinion about the transit system are likely to choose the neutral response to Likert scale questions. At the same time, travelers who are familiar with and/or informed about the transit system may also choose the neutral response to state their opinion neutrality. As a result, some travelers’ unfamiliarity with (and lack of opinion about) transit services may be confounded with the informed perceptions of those who are familiar with transit. This is because those who are unfamiliar with the transit system are less likely to use it and more likely to state neutral responses than those who are familiar with the system. Ignoring such influence of travelers’ unfamiliarity can potentially distort the ordinal scale of Likert variables, result in biased parameter estimates and distorted implications about the influence of perceptions on transit usage. To address this concern, this study uses a generalized heterogeneous data model (GHDM) that allows a joint econometric analysis of the influence of individuals’ perceptions of transit service quality on their likelihood of transit use and frequency of use and at the same time disentangle unfamiliarity from informed perceptions. The empirical results shed light on: (a) the role of individuals’ demographic variables and subjective perceptions on their use and frequency of use of the bus transit system in Bengaluru, (b) the importance of separating unfamiliarity from informed opinions on transit service quality, (c) the need to include an option for respondents to reveal their unfamiliarity in Likert rating scale survey questions on perceptions, and (d) demographic segment-specific strategies for attracting new riders and enhancing ridership of current users of the bus transit system in Bengaluru.
- Latent variable model
- Likert scale responses
- Perceptions of service quality
- Public transit use in India
ASJC Scopus subject areas