緒論：探討年長者的個人屬性與身心狀況對其選擇休閒活動類型的影響效果。方法：利用2011年教育部之「GOGO樂活休閒服務專案計畫之子計畫－銀髮族樂活休閒服務需求分析」所調查臺南縣、市60歲以上年長者的問卷訪談之部份資料，包括年長者的個人屬性與身心狀況等，藉由個體選擇模型中的多項羅吉特模式（multinomial logit model, MNL）建立年長者的休閒活動類型選擇模式，分析年長者參與休閒活動類型的需求。結果：有配偶且配偶健在、專注力持續時間較長、且自評身心狀況佳的年長者傾向於選擇從事「戶外活動型」的休閒活動；60~69歲之年長者相較於其他年齡層的年長者選擇從事「藝文嗜好型」休閒活動的機率較高；男性、專注力持續時間較短、有慢性病、及自評身心狀況差者選擇從事「運動型」休閒活動的機率較高；男性較可能選擇從事「益智型」休閒活動；男性、與家人同住、及有慢性病之年長者，傾向於選擇從事「休憩社交型」的休閒活動。結論：本研究囿於實證資料的限制，造成選擇模型的整體配適度不理想，但個別變數的統計檢定結果顯示，對特定社會經濟屬性與身心狀況的年長者在從事休閒活動時的確有其偏好的休閒活動類型，此結果對政府部門及社會福利機構規劃老人休閒活動具有一定的參考價值。

Introduction: The main purpose of this study was to explore the choice behavior for types of leisure activity for senior citizens. Methods: The data, regarding with individual senior citizen's demographic attributes and assessments of his/her physical and psychological condition, was part of the information surveyed by the work of the Project of GoGo LOHAS Recreational Technology Research and Development Center granted by the Ministry of Education of Taiwan on 2011, and was used to establish senior citizen's choice model for types of leisure activity by adopting the multinomial logit model (MNL) so as to analyze their choice behaviors. Results: The outcome showed that senior citizens living with surviving spouse, having longer duration of concentration, and keeping better physical and psychological condition of self-assessment tended to engage in leisure activities of outdoor activity type; those senior citizens with ages distributed in the interval of 60-69 had higher probability to choose leisure activities of arts hobby type; besides, the senior citizens of male, having shorter duration of concentration, suffering from a chronic disease, and keeping worse physical and psychological condition of self-assessment had higher probability to choose leisure activities of athletic type; male senior citizens were more likely to engage in leisure activities of puzzle type; and finally, the senior citizens of male, living with their family, and suffering from a chronic disease tended to engage in leisure activities of social type. Conclusions: Though the overall goodness-of-fit, the likelihood ratio, for the choice model was inferior owing to the restriction of the data used, the statistical test for individual variable showed that some senior citizens with specific socioeconomic attributes and physical and psychological conditions indeed preferred to some types of leisure activity, and the results would be of much help to the development of leisure activity and facility for senior citizens.