In order to explore the relationship between land use and the occupational and the job-housing function of subway stations, residential passenger flow index and employment passenger flow index were proposed as site clustering indicators. Firstly, the subway stations were classified to describe the type of job-housing function of the station. Then, the spatial logistic regression model was constructed to analyze the influence of land use around the subway station on its job-housing function under the influence of spatial effect. The results show that compared to employment-oriented stations, residential-oriented stations exhibit a more significant spatial dependence, for each additional station of the same type around residential-oriented stations, the probability of forming such stations increases by 4.1%. Taking comprehensive stations as a reference group, for residential concentration-oriented stations, for every additional educational or cultural or parking facility point of interest (POI) around them, the probability of forming such stations decreases by 2.7% and 3.2%. In contrast, for employment concentration-oriented stations, for every additional life service or accommodation-related POI around them, the probability of forming such stations decreases by 2.1% and 0.8%, facilitating their transformation into comprehensive stations. The research results can provide reference for station area land use planning or optimization strategies.