Correlates of formal and informal public transport usage in Windhoek, Namibia

Abstract

Despite the growing body of literature on public transport systems in African cities, there remains limited empirical research on how neighbourhood characteristics, land use patterns, and commuting behaviour collectively shape the choice between formal and informal public transport modes. This study addresses this gap by examining the correlates of transport mode usage in Windhoek, Namibia, with a focus on municipal buses (formal), traditional taxis, and e-hailing services (informal). It investigates three key questions: How do neighbourhood characteristics, land use patterns, and commuting behaviour correlate with formal public transport usage? How are these factors associated with taxi usage in Windhoek? And what drives the adoption of e-hailing services in the city? The hypothesised correlations between these factors were tested by collecting data from a household survey of 1000 respondents in Windhoek and developing three multiple linear regression models to analyse the relationships. Results indicate that accessibility-related variables, particularly proximity to formal public transport stops and service frequency, are the strongest predictors of formal public transport usage, while affordability and service quality are key drivers of taxi usage. Population density and car ownership show divergent effects across modes: taxis and e-hailing services are more commonly used in low-density areas and among car-owning households, suggesting that informal modes often complement, rather than substitute, formal transport services. The findings reveal context-specific patterns of transport behaviour in a hybrid mobility system, contrasting with trends in high-income cities. This study contributes new empirical evidence from Windhoek’s under-studied transport context and offers policy-relevant insights for improving accessibility, affordability, and integration in rapidly urbanising African cities.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 2 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.