In this study, we assume that socio-economic development is the underlying mechanism for population changes. In the past 50 years, Taiwan has experienced drastic socio-economic changes and rapid population change. Simultaneous equations are thus used to integrate the effect of Taiwan's socio-economic factors on the projection of its age-specific fertility rate (ASFR). The estimated ASFR is then used to make population projections for Taiwan in the period 2004-2033. The results of our low and medium projections are close to the official (CEPD) medium-high projection for the same time period. The results of the high projection for both studies are rather similar too. Specifically, if Taiwan's socio-economic development grows at a high variant, its fertility would rebound to replacement level by 2033. If it follows the medium variant, fertility would rise to 1.6 births per woman. But if Taiwan's future socio-economic development grows sluggishly as in the assumed low variant case, Taiwan's fertility would remain at a lowest-low level.

本研究假設社經發展是人口數量改變的潛在機轉。在過去50年，台灣經歷了急劇的社經變化和快速的人口變遷。同步方程式因此被引用來整合社經因素探討對年齡別生育率之影響。估計所得的年齡別生育率再用來推計台灣2004-2033的人口。本研究的低與中推計結果接近行政院經建會同期間的中高推計，而兩者之高推計結果也接近。質言之，假如台灣社經發展呈快速成長總生育率有可能在2033年反彈到替代水準；假如社經發展呈中度成長，總生育率有可能回升到1.6；假如社經發展持續下挫，總生育率有可能維持在超低水準。

中國推行計劃生育政策的目的之一就是增進兒少福祉。然而，過往研究多注重生育政策的人口後果，因此，政策是否達到了改善兒少福祉的目的尚不清楚。本文利用2000年「中國健康與營養調查」資料，突顯出政策的地區差異和在很大程度上取決於政策的兄弟姊妹結構，探討生育政策與青少年教育機會之間的關係。本文發現，獨生子女的失學概率低於非獨生子女，尤其低於有一個哥哥或兩個以上(2+)兄弟姊妹的孩子，而獨生子女在教育方面的優勢在後義務教育階段尤其明顯。由此可見，通過教育來衡量的青少年的素質因姊妹數量的增多和兄長的存在而降低。因此，繼續宣導「優生優育」的生育理念（但這並不等於提倡「獨生子女政策」）、促進兄弟姊妹之間的胎次和性別平等、完善現行的教育制度及延長義務教育年限將有助於改善青少年的教育福祉。

One of the purposes of China's one-child policy is to improve child wellbeing. However, past studies overwhelmingly focus on policy effect on fertility, and thus, whether the policy has achieved this goal remains unclear. This paper explores the relationship between the one-child policy and school dropout among adolescent children at the turn of the century, using the 2000 CHNS data. Policy is measured directly by local variations of policy strength (strict one-child policy, girl-exception policy and two-child policy) and indirectly by sibship composition. Analytical findings suggest that policy strength does not yield an independent effect on education. However, among children who have been educated beyond compulsory education, only children have a consistent, strong advantage in schooling relative to those with an older brother or two or more siblings. The results provide evidence to support the idea that promoting sibling equality in schooling and expanding compulsory education will necessarily help increase adolescents' educational opportunities.

本文利用民國81年與91年「國內遷徙調查報告」的個體資料來研究遷移行為。我們同時考慮遷移決策與遷移地點選擇並以兩巢層巢式羅吉特模型進行聯合估計。如此對於遷移行為有更全面的掌握，估計的結果亦可涵蓋過去研究的成果，同時考慮文獻上所討論各種影響遷移行為的因素，個體與總體資料同時使用，不但考慮個人社經特性也考慮了總體環境變數，較過去的研究更為全面。
本文的估計結果可以發現台灣的遷移行為在十年來有很大的變化。在影響遷移決策的影響變數方面，民國91年與81年的主要不同在於教育程度對於遷移的影響程度大幅的擴大。顯示出在知識經濟的潮流下，高教育程度較容易在各不同地點就業的優勢。在遷移地點的選擇上，可以發現十年來選擇遷入台北、高雄地區的機率下降，而遷入桃園、基隆、台中的機率增加，遷入其他縣市的機率也有小幅增加，顯示出台北都會區的外擴、台中都會區的興起以及城鄉的差異性逐漸縮小。另外我們也發現，台北、高雄、台中地區的遷移者以留在原地區的機率最高，但是其他地區居民則主要選擇遷往與原居地相近的都市地區，顯示為了加速適應環境以及回鄉的便利性，遷移地點仍然偏好選擇與原居地相近的地方。

From survey data we can find that the rate of relocation has declined significantly in the past decade in Taiwan. To understand the nature of this change, a conceptual model is developed to identify all factors that affect the joint decision regarding moving and location choice. Then a two-level nested logit model is used to empirically estimate the model.
The result shows the effect of education level on whether to move or not significantly increased from 1992 to 2002. This demonstrates that the advantage in employability in different locations with a higher education level has increased in this past decade. The result also shows that the probability of moving into the Taipei and Kaoshiung areas greatly decreased in the past ten years, while the probability of moving to Keelung, Taoyuan and Taichung increased. This shows the enlargement of the Taipei metropolitan area and the rise of Taichung metropolitan area.

本文使用「華人家庭動態調查資料」，探討在不同的經濟發展階段之下的五個出生世代（1930-39、1940-49、1950-59、1960-69、以及1970-79），家庭背景對個人教育成就的影響。文中發現對出生於1930-39、1940-49、以及1950-59的世代而言，性別是影響個人的教育成就最重要的變數，其次為家庭社經地位（以父母親教育年數與父親職業為代表）。到了1960-69與1970-79的出生世代，家庭社經地位對個人教育成就的影響已高於性別的影響。雖然家庭社經地位一直是解釋個人教育成就的重要變數，但其解釋能力卻是逐代下降。此外，手足人數之負向影響，以1950-59與1960-69出生世代最為顯著。出生序則隨台灣的經濟發展階段而對個人教育成就有不同的影響。一般而言，手足人數與出生序，對女性有較大的影響。再者，姊姊有助於個人教育成就的現象主要存在於1940-49之前的出生世代，對於1950-59以後的出生世代，不管是哥哥、姊姊、弟弟、或妹妹皆扮演家中資源競爭者的角色。最後，控制其他家庭背景因素解釋變數後，族群差異主要存在於1940-49與1950-59的出生世代，其後差異已不顯著。

This paper examines the impact of family background upon a child's educational achievement over five cohorts: 1930-39, 1940-49, 1950-59, 1960-69, and 1970-79. The data are taken from the Panel Study of Family Dynamics. Our empirical results suggest that (1) among all the variables related to family background, parents' social economic status plays an important role in determining their children's educational achievement, yet its influence declines over time; (2) the inverse relation between sibship size and a child's schooling accentuates for the cohorts 1950-59 and 1960-69; (3) the relationship between birth order and a child's schooling varies along different stages of economic development; (4) the effect of older sisters upon a child's educational achievement becomes negative after the 1950-59 cohort; and (5) ethnical differences mainly exist among the 1940-49 and 1950-59 cohorts.