Perceived safety and first/last mile mode choices. A dataset from image-based double stated preferences experiments

This dataset originates from an image-based double stated preference experiment conducted to assess perceived safety across four different first/last mile transport modes: car, e-bike, e-scooter and walking. The experiment involved 129 respondents from Athens, Greece who were (mostly) unfamiliar with micromobility transport modes. Later, 50 respondents from Munich, Germany were integrated in the dataset (v2.0). It contains 8171 perceived safety ratings and 2060 choices.

Each scenario was represented through an image of an imaginary road environment, accompanied by explanatory text. These images depicted various road infrastructure types, pavement conditions, static objects (e.g., pedestrian crossings, obstacles), and moving objects (e.g., vehicles, pedestrians, bicycles). Participants rated perceived safety on a Likert scale from 1 (very unsafe) to 7 (very safe) in response to the question:

“How safe would you feel on this road? To drive a car? To ride an e-bike? To ride an e-scooter? To walk?”

Additionally, an extra question was included to explore mode choice under realistic traffic conditions:

“Your daily route consists of urban roads with traffic conditions like above. Which transport mode would you select, if you were aware of the travel time and cost?”

The dataset provides valuable insights into how urban road environments influence perceived safety and first/last mode choices, highlighting the significant safety concerns associated with car-dominated road infrastructure as a barrier to adopting micromobility services in cities

For a detailed breakdown of the dataset variables, please refer to the accompanying README file.