User Intentions for Adopting Public Transport and Micromobility Through Incentives

Incentive-based mobile applications are increasingly considered tools for encouraging public transport (PT), walking, cycling and micromobility. However, their potential depends on whether travellers are willing to use the applications and whether reward mechanisms reflect existing travel habits and user preferences. This study investigates the factors associated with intentions to adopt a digital incentive application designed to support combined public transport and micromobility use in Thessaloniki, Greece. 

An online survey conducted in March 2024 gathered 301 responses covering demographic and socioeconomic characteristics, existing mobility behaviour, familiarity with technology, willingness to use the application and preferred incentives. The data were examined through descriptive statistics and an ordered logistic regression model, with intended application use measured on a five-point ordinal scale. The modelling considered technological familiarity, use of transport-related applications, public transport quality perceptions, bicycle-use frequency, age, income and household size. 

Results show that greater technological familiarity and frequent use of mobility applications were positively associated with willingness to adopt an incentive application. Respondents who never cycled were considerably less willing to use it than daily cyclists, suggesting that rewards alone may not overcome barriers associated with cycling. Those rating existing public transport as very good also reported lower adoption intentions than respondents giving it a very bad rating. Household size and some age and income categories were negatively associated with intended uptake relative to their respective reference categories. Descriptive responses indicated preferences for supermarket and mobility-service vouchers, as well as rewards for personal achievements rather than team-based competition. 

The findings highlight the importance of tailoring incentive schemes to different user segments and combining them with accessible app design and supportive mobility infrastructure. They can inform the design and targeting of gamified or reward-based initiatives seeking to make multimodal sustainable travel more attractive. The analysis is based on self-reported intentions, rather than actual application uptake, verified trip behaviour or demonstrated modal shift. 

Supporting evidence

The study used 301 responses to an online questionnaire distributed through local websites in Thessaloniki in March 2024. Respondents provided information on mobility habits, technological familiarity, demographics, willingness to use an incentivization application and reward preferences. An ordered logit model was fitted in R using the MASS package, with the dependent variable measured on a five-point scale. The final model reported AIC=730.641 and McFadden’s pseudo-R²=12.6%. 

Compared with respondents reporting neutral technological familiarity, those reporting good (coefficient 2.527; p=0.001) or very good familiarity (2.502; p<0.001) were more willing to adopt the application. Always using transport apps was also positively associated with intention (1.473; p=0.001), relative to never using them. 

Reporting very good public transport quality was negatively associated with adoption intentions (−1.795; p=0.035), compared with a very bad rating. Never cycling, compared with cycling daily, was negatively associated (−2.134; p=0.002). Larger household size also showed a negative coefficient (−0.219; p=0.050). 

The descriptive analysis reports a preference for supermarket and mobility-service coupons and for individual achievements over inter-team competitions. The sample included very few respondents over 65, limiting the generalisability of the results to older travellers. These findings describe stated preferences and statistical associations rather than observed effects of an incentive programme. 

Key findings

1. Technological familiarity is an important correlate of adoption intentions. Respondents reporting good or very good technological familiarity were substantially more willing to use the proposed incentive application than those in the neutral reference category. Frequent users of existing transport apps also expressed stronger intentions. These results support testing accessible interfaces and digital onboarding for people who are less familiar with mobile technology.
2. Existing travel habits matter. Respondents who reported never using bicycles were less willing to adopt the application than daily cyclists. Incentives may therefore be more readily accepted by individuals who already practise some sustainable mobility, while non-cyclists may require additional support. The study discusses cycling infrastructure and perceived safety as potential barriers but does not directly quantify their effects.
3. Public transport satisfaction shows a counterintuitive association. Participants rating current public transport quality as very good reported lower intention to use the incentivization app than those rating it very bad. This may indicate that rewards appeal differently to satisfied and dissatisfied travellers; it does not establish that incentives improve service satisfaction or retain passengers.
4. Demographic differences suggest that a single incentive design may not suit everyone. Some older and higher-income respondent categories showed lower intentions than their reference groups, and larger households were less receptive. Given the limited representation of older adults and people with disabilities, broader evidence is needed before generalizing these patterns.
5. The preferred reward format is practical and individualized. Descriptive responses favoured coupons for supermarkets and mobility services and recognition of individual achievements rather than team competition. Designers could investigate loyalty points, discounts and personal challenges as alternatives, while evaluating which actually affect sustained engagement.
6. Actual behavioural effects remain untested. The ordered logit model explains variation in stated app-use intention; it does not measure app adoption, verified multimodal journeys or reductions in car use. Field trials and longitudinal monitoring are needed to establish whether incentives produce lasting modal shift.