Numerous methodologies measuring walkability have been developed over the last years. This paper reviews the Walkability Index (WI) literature of the last decade (2009–2018) and highlights some limitations in the current approaches. Only a few studies have evaluated walkability in Latin America, mainly in big cities but not in medium and small-sized cities in the region, which present their own urbanisation dynamics, security issues, sidewalk invasion problems, and poor planning. Furthermore, most WIs in the literature use objective mesoscale variables to assess walkability in a given area. This paper contributes to filling these gaps by generating new evidence from a medium-sized city in Latin America to question if characteristics of the built environment encourage walking trips, as found in the literature, are transferable among regions. The study also proposes a novel index comprised of microscale and mesoscale built environment variables to assess walkability using virtual tools and considering users’ perceptions. The WI estimation relies on ranking probability models. The results of the case study suggest that subjective Security and Traffic Safety are the most crucial factors influencing walkability in these kind of cities, which is different from what is found in the literature from cities in developed countries where Sidewalk Condition and Attractiveness are the most important factors. Security appeared to be strongly associated with a subjective dimension, represented by the fear of crime or perceived risk for crime, instead of the actual occurrence of crimes. This result evidences the importance of the physical attributes of the real world and how they are captured, judged, and processed by pedestrians. Then, regional transferability of WIs needs to be done carefully. Finally, results in this paper highlight the importance of microscale built environment characteristics in the WI formulation in these cities. Results are in line with other research in some cities of the region, which found that microscale variables such as pavement quality and presence of obstacles on the sidewalks are relevant components to promote walkability.

Studies conducted in cities from developed countries reveal associations between travel behavior and built environment. A priori we would expect that most of these associations hold for cities from developing countries but we do not have any basis to suppose whether or not they will be quantitatively the same because of differences in the nature of urban form between cities of developed and developing countries, and potential differences in attitudes and travel preferences from their travelers. This article analyzes the relationship between walking patterns, travel attitudes, and neighborhood's urban structure in Porto Alegre, Brazil, under the assumption that built environment affects travel attitudes, instead of the more typical assumption that goes the other way around. We develop a two-step ordered choice model enriched with attitudinal variables to integrate travel attitudes, built environment, and travel behavior. A multiple indicator multiple cause (MIMIC) model is used to identify travel attitude as latent constructs that feed a frequency-of-walking ordered logit model with both the built environment and travel attitudes as explanatory variables. Our results support the a priori assumption that we would replicate many of the results already reported in the literature. However, we obtained the same differences in the magnitudes of some key elasticities. The results of the elasticities computed for the number of walking trips with respect to the built environment variables were smaller than in other studies. Only the effect of population density on walking frequency seems to be marginally higher.