人口轉型導致生育率下降的原因分別有「社經結構調適過程」與「（創新）傳播擴散過程」兩種解釋途徑，前者認為生育率的下降是社會邁向現代化過程中對於結構轉變的調適結果，強調社經結構對於行為者的生育態度及決策的影響，而後者則認為社會網絡與人際互動等因素，形塑大眾減少或延後生育的觀念，以及對避孕知識與方法的認識並採納，是促使生育率普遍下降的主要原因。本研究透過空間迴歸模型，以臺灣358個鄉鎮為研究範圍，同時考慮社經結構，以及空間傳播擴散因素，檢驗社經結構、族群以及鄰近效應對生育率的空間變異可能造成的影響。綜合而論，從1980、1990、2001三個觀測年度的迴歸分析結果發現，社經結構始終維持在百分之五十的解釋力，族群與鄰近效應的解釋力則有增加的趨勢，尤其是鄰近效應的解釋力增加尤其明顯。基本上，鄰近效應有補強功能，提高模型的解釋力，但不影響原有變數的解釋力，證實的確有鄰近效應存在，間接支持傳播擴散理論，使社經結構調適理論在解釋生育率上更為完整。

In past studies, there are two theories explaining the process of decline in fertility. The first is the socioeconomic adjustment theory, which indicates that change in fertility is related to motivational forces stemming from changes in socioeconomic conditions; the second is the innovation diffusion theory, which admits the importance of motivational forces as well, but argues that social networks and interpersonal connections, rather than socioeconomic factors, are the reasons for making people accept contraception. In this paper, by using the methods of spatial econometrics, we consider both the factors of socioeconomic structure and spatial diffusion to explain the decline in fertility rates in Taiwan across 358 townships and within three years, 1980, 1990 and 2001. The spatial lag dependent variable, representing neighborhood effects, is used to verify the application of the innovation diffusion theory. The findings suggest that socioeconomic structure factors could explain about half of the variance in fertility rates; in addition, ethnic factors are significant, showing that some cultural reasons played an important role in the process leading to a low fertility rate. Finally, the existence of neighborhood effects in our regression models implies that the innovation diffusion theory has been verified, and its influence has been increasing with time.