Comparing travel behaviour characteristics and correlates between large and small Kenyan cities (Nairobi versus Kisumu).

Understanding urban travel behaviour is crucial for planning healthy and sustainable cities. Africa is urbanising at one of the fastest rates in the world and urgently needs this knowledge. However, the data and literature on urban travel behaviour, their correlates, and their variation across African cities are limited. We aimed to describe and compare travel behaviour characteristics and correlates of two Kenyan cities (Nairobi and Kisumu). We analysed data from 16,793 participants (10,000 households) in a 2013 Japan International Cooperation Agency (JICA) household travel survey in Nairobi and 5790 participants (2760 households) in a 2016 Institute for Transportation and Development Policy (ITDP) household travel survey in Kisumu. We used the Heckman selection model to explore correlations of travel duration by trip mode. The proportion of individuals reporting no trips was far higher in Kisumu (47% vs 5%). For participants with trips, the mean number [lower - upper quartiles] of daily trips was similar (Kisumu (2.2 [2-2] versus 2.4 [2-2] trips), but total daily travel durations were lower in Kisumu (65 [30-80] versus 116 [60-150] minutes). Walking was the most common trip mode in both cities (61% in Kisumu and 42% in Nairobi), followed by motorcycles (17%), matatus (minibuses) (11%), and cars (5%) in Kisumu; and matatus (28%), cars (12%) and buses (12%) in Nairobi. In both cities, females were less likely to make trips, and when they did, they travelled for shorter durations; people living in households with higher incomes were more likely to travel and did so for longer durations. Gender, income, occupation, and household vehicle ownership were associated differently with trip making, use of transport modes and daily travel times in cities. These findings illustrate marked differences in reported travel behaviour characteristics and correlates within the same country, indicating setting-dependent influences on travel behaviour. More sub-national data collection and harmonisation are needed to build a more nuanced understanding of patterns and drivers of travel behaviour in African cities.