少子化下，全臺學齡人口時空分布差異與影響因素研究

A Study on the Temporal and Spatial Distribution Changes and Influencing Factors of the School-Age Population in Taiwan

本研究希望透過分析民國99到105年村里學齡人口的資料，瞭解少子化下學齡人口的時空分布差異，並推估學齡人口分布變遷的影響因素。民國99年至105年臺灣0至14歲之學齡人口分布主要集中在六都和新竹縣市，以北北桃的熱區比例最大，臺中市次之，新竹縣市與北苗栗縣居第三，高雄市和臺南市則為五個主要熱區中的最後兩名。據研究中的學齡人口結構發現，高雄市區之學齡人口數雖多，但實際占總人口之比例是低的，以人口結構來說，高雄市為臺灣六都中少子化現象較嚴重的地區。本研究也發現以國小入學前的遷移量較多；學齡人口入學後，其遷移的狀況減少，此外，各地移出狀況較平均，但想移入的地區有集中特定區域的情形，此與明星學區有關。本研究最後從空間分析的角度探究社會經濟因素與傳播擴散效應對少子化相關應變數的影響。總結來說，社會經濟因子和空間的鄰近效應都同時影響臺灣的少子化相關變數，以解釋力最高的國民教育階段學齡人口比例和較有代表性的出生率之迴歸分析來看，出生率及學齡人口比例與戶量，山地原住民比例呈正相關，出生率及學齡人口比例與碩士學歷以上比例、人口密度呈負相關，此四個社會經濟變數與研究假設相符；但出生率及學齡人口比例與社會增加率、綜合所得中位數呈正相關，出生率及學齡人口比例與死亡率、已婚人口比例呈負相關，此四個社會經濟變數與一般研究假設之都市化地區出生率較低的情況不同，但符合研究發現的趨勢，出生率和學齡人口比例集中於臺灣的都市地區。本研究發現臺灣出生率和學齡人口分布情形正在轉變，雖鄉村地區仍有高生育率的情形，但在少子化的趨勢下，高出生率和學齡人口漸漸集中於資源較豐富的都市地區。

Through the analysis of the school-age populationon the village from 2000 to 2016, this study aims to understand the space-time difference of the distribution of school-age population in alow birthrate society. This study also aims to examine the influencing factors that cause changes in the distribution of the school-age population. From 2000 to 2016, the school-age population from the age of 0 to 14 years old were mainly distributed in the six municipality cities, Hsinchu county and Hsinchu city. Among them, cities that occupy the largest proportion of school-age population is Taipei, New Taipei and Taoyuan city, while the second being Taichung, the third being Hsinchu and northern Miaoli, and the last two being Kaohsiung and Tainan. From the demographic of the school-age population, we found out that despite there are many school-age children in Kaohsiung, the actual proportion to the total population is the lowest among the six cities. Regarding population structure, Kaohsiung is the area that suffers the most from low birthrate among the six municipality cities. This study reveals that immigrating and emigrating processes among Taiwanese households occur most frequently before students start going to elementary schools, while both immigrating and emigrating processes occur least frequently after school-age children are enrolled in school. Furthermore, the number of emigrants from each area is even, while the areas that they wish to immigrate to are concentrated in specific regions, often being popular school districts. Finally, this study discusses the influence of social economic factors, and spreading and diffusion effects on the corresponding low birthrate dependent variables from aspatial analysis perspective. In conclusion, social economic factors and spatial contingency effects both influence the corresponding low birthrate variables. Taking the most statistically significant regression analysis of education level and birthrate as an example, shows that the birthrate and school-age population will be higher if the number of households and proportion of indigenous peoples are higher, and if the number of population with an education level higher than a master degree and population density are lower. Meanwhile, with higher social increase and individual income rates, and lower death rate and lower proportion of married population, birthrate and school-age population will be higher. These four social and ecological factors contradict the common assumption that urbanized areas have lower birthrates. However, it matches the pattern that we found in the study, which is that birthrate and the proportion of school-age population concentrate in urban areas in Taiwan. This study finds that the distribution of birthrate and school-age population in Taiwan is changing. While the birthrate in rural area is still higher, within the trend of low birthrate, high birthrate and school-age population gradually concentrate in urban areas with more ample resources.