Probability Modelling of Motorcycle and Private Car Mode Choice in Sa'dan District, North Toraja
This study aimed to develop a binary logistic regression model to examine transportation mode choice between motorcycles and private cars among respondents in Sa'dan District, North Toraja Regency, Indonesia. A quantitative approach was employed using 162 valid observations collected through a household survey. Five explanatory variables were examined: vehicle ownership, household size, number of working household members, number of school-attending household members, and household income. Motorcycles accounted for 70.4% of the observed mode choices, while private cars accounted for 29.6%. Household income was the only statistically significant predictor of private-car choice (B = 2.303, p < 0.001, Exp(B) = 10.003). The overall model was statistically significant (Chi-square = 51.014, p < 0.001), with a Nagelkerke R-squared of 0.384 and a non-significant Hosmer-Lemeshow test (p = 0.639). The model achieved an apparent classification accuracy of 78.4% and an area under the receiver operating characteristic curve (AUC) of 0.830. The mean predicted probability of private-car choice was 29.63%. These findings indicate that household income was positively associated with private-car choice in the study sample. The study demonstrates the application of binary logistic regression to examine transportation mode choice in a semi-rural area, while highlighting the need for further validation using independent data.
Authors
- Azril Tangke Sombolinggi
- Rilva Toding Bua
Publication Details
- Journal
- Andalasian International Journal of Applied Science Engineering and Technology
- Published
- 2026-11-17
- DOI
- https://doi.org/10.25077/aijaset.v6i3.375
- Primary Topic
- Transportation Planning and Optimization
- Type
- article
- Field-Weighted Citation Impact
- 0.00