[[alternative]]A Study on Domestic Travel Constraints for The Elderly in Taiwan

[[abstract]] 根據內政部（2010）公布統計資料顯示，1993年起我國銀髮族佔總人口的10.60%，已經成為「高齡化社會」。台灣銀髮族群於社會中所占比例逐年攀升，相關銀髮族休閒產業的發展對社會整體經濟層面影響力與日遽增。2013年「國人旅遊狀況調查」資料顯示，銀髮族不從事國內旅遊的比例占42.14%。若能探究其中原因並克服旅遊阻礙因素，提升銀髮族適宜遊憩活動之參與，除了可為其帶來健康之效益，亦可使其成為國內旅遊市場與促進經濟發展的重要族群。 本研究利用2001年至2013年「國人旅遊狀況調查」資料，分析台灣銀髮族國內旅遊參與阻礙的因素。為了解決進行阻礙因素分析可能帶來的樣本選擇性偏誤問題，本文以Heckman所提出之兩階段估計法進行估計。第一階段，以probit model探討銀髮族人口特徵變項與是否旅遊之間的關係。第二階段，加入樣本選擇性偏誤修正項，以multinomial probit model探討銀髮族人口特徵變項與阻礙因素之間的關係。 第一階段估計結果顯示：教育程度愈高且所得愈高之者銀髮族，不旅遊的機率愈低；住在非都會區的銀髮族相較於都會區者，不旅遊的機率愈高，且城鄉差距有擴大的趨勢；長期而言，台灣銀髮族不旅遊的機率則是愈來愈低。第二階段估計結果顯示：台灣銀髮族因「沒有興趣」和「無法負擔」這兩個因素而不去旅遊的機率愈來愈低。且進一步發現教育程度愈高、所得愈高的銀髮族女性，會因為「沒有興趣」而不去旅遊的機率會愈低。至於因為「沒有時間」、「無法負擔」和「身體健康因素」這三個理由而無法旅遊者，過去13年來並沒有明顯的變化趨勢。[[abstract]] According to the Ministry of Interior (2010), Taiwan’s elderly population has reached 10.60% since 1993, which defined the country as an ageing society. The rising proportion of the elderlies in population has a greater impact on the overall social economy. The Survey of Travel by R.O. C Citizens of 2013 showed that the elderlies not engaging in domestic travels accounted for 42.14%. If the reasons causing such phenomenon can be investigated and solved, the involvement in proper recreational activities of the elderlies would increase gradually. The consequence would not only bring benefits to their health, but make them become an important role in the domestic travel market and social economic development. Thus, the study on the constraints in elderlies traveling has become an important issue. This study utilizes data from the 2001 - 2013「Survey of Travel by R.O.C. Citizens to analyze the factors affecting domestic travels of the elderlies and their constraints. In order to resolve the issue of selection bias, I uses Heckman’s two-stage estimation method. In the first stage, the probit model is used to examine the relationship between socio-demographic variables and the probability of not traveling in the elderly population. In the second stage, the multinomial probit model is used to investigate the relationship between socio-demographic variables and constraint factors. The results of the first stage show that those with higher education and income level had a lower probability of not traveling. Those living in non-metropolitan areas have a higher probability of not traveling compared to those living in metropolitan areas, and the gap between these two is increasing over time. Overall, the probability of not traveling for Taiwan’s elderly population has been decreasing over time. The results of the second stage show that the impacts that “lack of interest” and “lack of budget” have on not traveling have gradually decreased; in addition, the impact of “lack of interest” as a reason on not traveling has decreased even further on elderly women with higher education and income level. As for elderly people who cite the “lack of time”, “lack of budget” or “health conditions” as reasons for not traveling, there has been no significant change over past 13 years