Introduction Social and self-identities have been conceptualised to prevent travel behaviour change, as threats to one's identity may cause resistance to change. This study focuses on the role of social, transport, place, and self-identities on commute mode choice and intention to change mode choice. Method Data were collected in June 2015 in Utrecht, the Netherlands. Invitations to participate were distributed by mail using data from the municipality, resulting in 1062 adult participants. The outcome measures were the transport mode shares based on a 14-day travel-to-and-from-work record of trips (i) involving any car use, (ii) involving any bicycling, (iii) involving any walking, and (iv) involving any public transport use. The second series of outcome measures concerned the willingness to change the amount of car use, bicycle use and walking, determined by the question ‘to what extent do you intend to change the use of …?’. Identity was measured on a seven-point disagree/agree scale for 17 items by asking to what extent the respondent ‘sees him/herself as …’. Separate multinomial regression models were estimated stepwise adjusting for socioeconomic and transport characteristics. Results Multiple identity items were associated with the use of all commute modes. In the maximally adjusted models, identities associated with the respective modes remained significant. For example, whether someone identified themselves with being a cyclist corresponded with higher likelihood of cycling occasionally (relative risk ratio (RRR): 1.84; 95% confidence interval (CI):1.47–2.30), or always to work (RRR: 2.86; 95% CI: 2.16–3.79). In addition, we found that a family-oriented identity was negatively associated with occasional commuting by car, and a ‘sporty’ identity was negatively associated with always cycling to work. Transport identities were also associated with stated intentions to change as were several social, place, and self-identities. Identifying with being a car driver decreased the likelihood of intending to reduce car use, but it increased the likelihood of intending to increase car use, as did identifying with being career-oriented. Individuals that identified with being a cyclist were less likely to have an intention to reduce bicycle use, whereas countryside-lovers had greater intentions of increasing cycling. Individuals that identified themselves as pedestrians had a lower intention of decreasing their walking levels, and a higher intention of increasing them, as did those who identified themselves as being family-oriented. Discussion The results confirm limited previous findings that identifying with users of a transport mode correspond with its use. Nevertheless, questions around causality remain. The intention to change mode choice was associated with several identities, including transport-related identities, place-related identities, social/family-related identities, and self-identities. Future research should focus on the associations between identity and actual behaviour change to further our understanding of the effect of identity on travel behaviour.