To facilitate the implementation of deep learning-based vehicular applications, vehicular federated learning is introduced by integrating vehicular edge computing with the newly emerged federated learning technology. In vehicular federated learning, it is widely considered that the raw data collected by vehicles has complete ground-truth labels. This, however, is not realistic and inconsistent with the current applications. To deal with the above dilemma, a Semi-supervised Vehicular Federated Learning (Semi-VFL) framework is proposed. In the framework, each vehicular client uses labeled data shared by an application provider, and its own unlabeled data to cooperatively update a global deep neural network model. Furthermore, the application provider combines multi-dimensional contract theory with prospect theory to design an incentive mechanism to stimulate appropriate vehicular clients to participate in Semi-VFL. Multi-dimensional contract theory is used to deal with the information asymmetry scenario where the application provider is not aware of vehicular clients’ three-dimensional cost information, while Prospect Theory (PT) is used to model the application provider’s risk-aware behavior and make the incentive mechanism more acceptable in practice. After that, a closed-form solution for the optimal contract items under PT is derived. We present the real-world experimental results to demonstrate that Semi-VFL achieves the advantages in both the test accuracy and convergence speed, in comparison with existing baseline schemes. Based on the experimental results, we further perform the simulations to verify that our incentive mechanism is efficient.