Explaining travel behaviour with limited socio-economic data: Case study of Vishakhapatnam, India

• Published in: Travel, behaviour & society Vol. 15; pp. 44 - 53
• Main Authors: Jain, Deepty, Tiwari, Geetam
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.04.2019
• Subjects: Asset ownership; Modal share; Principal component analysis; Socio-economic wellbeing; Trip length; Trip length; Asset ownership; Modal share; Socio-economic wellbeing; Principal component analysis
• ISSN: 2214-367X
• DOI: 10.1016/j.tbs.2018.12.001

•Socio-economic wellbeing score (SEWS) as proxy of income for travel behaviour study.•Application of principal component analysis on asset ownership data for estimating SEWS.•Significant association between the groups defined by SEWS and reported income.•Average trip lengths are significantly different between low and very high SEWS groups in Vishakhapatnam.•Average walking distance does not vary significantly with regard to SEWS groups in Vishakhapatnam.•Dependency on non-motorized transport decreases with increasing SEWS. Travel behaviour varies with respect to the income. Directly reported incomes used in the travel behaviour studies are subject to the issues of under and non-reporting. To account for this, we propose principal component analysis (PCA) on household asset ownership data to estimate socio-economic wellbeing score (SEWS) as the proxy of income. SEWS is used to understand the variation in travel behaviour of people belonging to different income groups in Vishakhapatnam. We have used sample data of 2623 households collected in 2012–2013. Internal coherency test and chi-square tests are conducted to assess the robustness of estimated SEWS. Travel behaviour analysis highlights that both the trip length and mode choice significantly varies with regard to the SEWS in Vishakhapatnam. People belonging to the low and low middle SEWS group are more dependent on walking and travel shorter distances as compared to the middle high and very high SEWS group. Encouraging the middle high and very high SEWS group to travel short distances and use low carbon modes of transport will need interventions related to the development control regulations and infrastructure provision. Since SEWS is estimated using multiple variables and captures the consumption pattern of the households, therefore, it can be used as the proxy of income in travel behaviour studies.