The Nested Logit Model was used to analyze the questionnaire investigation based on users’ travel modes choices. The factors influencing on users’ travel modes choices were researched. Users’ selection probability and sensitivity of travel modes were computed. The results show that cost, station distance, trip distance, driving-age, ownership of private cars and users’ age make a difference on selection probability of electric car sharing. Selection probability of electric car sharing based on RP(revealed preference) data is 2.36% while based on SP(stated preference) data is 3.30%. If cost of electric car sharing increase 1 yuan/kilometer, based on RP data selection probability of electric car sharing will decrease by 8.6% while based on SP data decrease by 8.9%. If station distance of electric car sharing increase 1 kilometer, based on RP data selection probability of electric car sharing will decrease by 43.7% while based on SP data decrease by 21.1%.