台灣腎臟病問題嚴重，不論是洗腎發生率或盛行率，在2008年均高居世界首位。許多討論認為一個主因在於民眾聽信地下電台買藥，因濫服來路不明成藥導致腎臟功能衰竭，連帶造成健保洗腎醫療費用節節高升，於2008年達到318億，成為健保中花費最高單一治療項目。 本論文主要目在釐清地下電台是否造成洗腎發生率和盛行率增加。由於地下電台普遍設於中南部且絕大多數以閩南話播放，這些節目對於外省籍榮民影響相對有限。本論文使用國衛院發行的健保資料承保檔與重大傷病檔，採DDD模型(difference-in-difference-in-difference)，比較同一地域但不同承保人口 (農民、榮民，與地區人口) 中，五十五歲以上洗腎發生率和盛行率是否因投保人口而有明顯差異，藉以釐清地下電台對腎臟病影響。估計結果顯示在2000年後高高屏地區農民洗腎發生率有略高於榮民，而雲嘉南地區農民和榮民的洗腎發生率不存在明顯差別。

身為大量消費的現代人，不得不關切垃圾何去何從。以往垃圾問題源自於焚化或掩埋兩法皆會損耗大量的環境成本及引起利益團體抗爭，造成社會無謂的損失，受世界環保潮流影響，我國政府決定從降低垃圾量著手。 垃圾減量從垃圾分類做起，方便後續的資源回收。我國自民國94年1月1日起，強制台灣10個縣市的民眾進行垃圾分類，成為我國少數未全國同步施行的中央政策。本研究引進多用於公共衛生、醫療等自然實驗的計量方法—差異中的差異(Difference-in-Difference，簡稱DID)來探討屬於社會科學領域的政府政策，檢驗該政策是否達到其制訂之目標。 本研究將台灣第一階段便實施強制政策的10個縣市當作實驗組，4個第二階段才全面實施的縣市作為控制組，在控制了各縣市的平均每月家戶所得、高等教育程度、清運人員、垃圾車數量等變數後，分別去分析政策對每人每日垃圾清運量、每人每日資源回收量、每人每日資源回收量相對於垃圾清運量比例三者的影響，觀察政策對於降低垃圾清運量、提高資源回收量的政策目的是否有效。 結果發現，我國隨時間經過，垃圾清運量降低、資源回收量提高，但由於主要探討的政策係數並不顯著，所以我國垃圾清運、資源回收情形的改善，是否起因於強制垃圾分類政策造成，沒有明確證據。 / Modern-day life consumes far greater resources than ever and, thus must be concerned about the “problems of trash.” Traditionally, government officials have adopted landfill and incineration policies to dispose of this municipal waste. Yet these two measures come with large social and environmental costs. In keeping with international trends toward environmental protection, the Environmental Protection Administration of the Republic of China has decided to solve the problems by fundamentally reducing the volume of municipal waste. In order to decrease the volume of waste and increase the recycled, the first phase of the “Compulsory Trash-Sorting Policy” was implemented in 10 localities on Jan. 1, 2005, with additional counties brought on board in later phases. This policy is one of the few policies that are not implemented island-wide in Taiwan. This study adopts the “Difference-in-Difference” approach—a popular research method in the field of public health, medical caring and natural sciences—to evaluate the effectiveness of this policy. The 10 counties where the first phase of the policy was successful is the “treatment group.” Another 4 counties where the second phase of policy was successful is the “control group.” This study takes into account monthly per family income, average level of higher education, number of street-sweepers and trash-haulers, and the trash trucks as the exogenous variables. This study calculates the changes in per capita daily volume of municipal waste and per capital daily volume of recycled materials to determine the success of the policy. This study concludes that the volume of the municipal waste and increases the recycled materials decreases by time; however the actual effect is still vague because the coefficient of policy in this study remains insignificant. Therefore, it is only certain that the campaign for “trash-sorting” works better with time. No evident linkage exists between recycling and the “Compulsory Trash-Sorting Policy.”

本研究探討美國最低稅負制與個人非現金捐贈行為的關係。1993年時美國對於最低稅負制做稅制上之修正，將非現金捐贈之扣抵稅額以購入成本計算修改為以公平市價計算；此一稅制上的改變，正提供一個相當適合研究的時機，可以對非現金捐贈部分進行深入了解。所以本研究以美國內地稅務局在1989年及1995年的橫斷面個人申報所得稅資料庫，並採用差異中的差異分析法，來探討最低稅負制度對於非現金捐贈行為的影響，並加入可能會影響非現金捐贈的解釋變數，包括：所得、財富、年齡、婚姻、扶養親屬人數等變數。 經實證結果發現，最低稅負制度對於非現金捐贈行為效果顯著且正向，即代表1993年美國稅制從以購入成本計算扣抵額修正為以公平市價計算扣抵額，此政策能夠誘發民眾對非現金捐贈金額增加，至於解釋變數上經實證結果發現，當所得及財富越高之納稅人，對非現金捐贈金額越高，在婚姻及年齡方面，已婚者及年齡較大者在捐贈金額上為多；至於扶養親屬人數方面，對於捐贈金額影響是不顯著的。 / This thesis is focus on the relation of the Alternative Minimum Tax (AMT) and Non-Cash Donations (NCD). The America amended the AMT in 1993. It changed the NCD-deduction calculation from cost to fair value. This revolution gave a good time for researching about NCD. So, this report took the individual tax database from Internal Revenue Service (IRS) in 1989 to 1995 for example. And used Differences-in-Differences (DID) way to discuss the infect of NCD from AMT. Besides, this research added some NCD variables including of income, wealth, age, marriage, numbers of dependent. The empirical result showed that the AMT had obvious and positive effect on NCD behavior. It represented that the AMT policy change in 1993 could cause the NCD increase. The more income and wealth had more NCD amount. People married and elderly people offered more donation than single and young. The numbers of dependent had no obvious effect on donation.