Park　&　Ride　(P&R)　systems　are　an　effective　way　to　attract　road　users　to　rapid　transit,　which　helps　both　reduce　the　traffic　entering　the　center　of　a　city　and　improve　the　efficiency　of　urban　rail　transit　systems.　This　paper　describes　a　Stated-Preference　(SP)　survey　of　potential　users　of　a　Park-n-Ride　system　in　Beijing.　Mode　choice　models　were　developed.　The　results　show　that　inclusion　of　preferences　led　to　improvement　in　model　performance.