台灣位於環太平洋地震帶上，地形為山地居多，且地質脆弱，加上位於西太平洋副熱帶地區，使得山區常受到颱風的侵擾而發生崩塌，導致土石流和洪水等災害發生，進而影響人民的生命和財產安全。因此，如何有效地建置崩塌地區域資料庫，成為國土保育與災害防治的重要課題。以往利用遙測與航測技術於崩塌地萃取的研究中，大多是於幾何糾正後衛星影像或是航測正射影像上分析崩塌地，但產製正射影像或是糾正衛星影像時，都需要花費較多的時間，對於講求時效性的救災行動而言頗為不利。本研究之目的為發展一套不需使用正射影像萃取崩塌地的方法，以物件導向影像分類法，於DMC（Digital Mapping Camera）航測原始影像上直接萃取崩塌地資訊。首先採取多重解析影像分割的技術，將航測影像依像元光譜和形狀同質性分割成不同區塊（物件），接著利用影像光譜統計值搭配區域成長法，偵測影像中的雲覆蓋地區並過濾。其次，根據光譜亮度統計特徵值，將影像區分成陰暗地區、正常地區以及較亮地區之三種土地覆蓋類型，使用線性相關糾正法（Linear-correlation Correction）將陰暗地區光譜亮度值轉換至正常地區，並利用物件的特徵值，如光譜、面積、形狀以及相關性依序萃取此三種土地覆蓋類型內的崩塌地。最後，使用光線追蹤法 （Ray-tracing），將崩塌地區塊從影像坐標轉換至地圖坐標，使其可以套疊地形資料如坡度、坡向，並進行空間分析以提升崩塌地的判釋精度。研究結果顯示，崩塌地萃取之使用者精度和生產者精度，均有82%以上，並且整個實驗可大量批次處理影像，及快速建立崩塌地資料庫，本研究之方法和崩塌地資料庫將有助於國土保育與崩塌地的災害防治。 / Being located in a subtropical and seismic zone of the West Pacific, the geology is fragile and topography is mountainous in Taiwan. Landslides, floods and other disasters induced by typhoons occur frequently, and it cause the life-threatening and property loss of human beings in Taiwan. Therefore, how to establish landslides data effectively become an important issue of land conservation and disasters management. In recently years, most of the researchers used aerial ortho-images or satellite georeferencing images to detect landslides sites. However, it spent a lot of time generating aerial ortho-images and rectifying satellite images, and it also reduced the efficiency of landslides analysis. Thus, this study developed an object-oriented classification method, which can be directly applied in raw image data, to detect landslides sites. Firstly, this study used multi-resolution image segmentation technique to segment images acquired by Z/I DMC（Digital Mapping Camera） into individual regions （objects） according to the homogeneity of spectral and shape features, and then removed cloud areas by using brightness features depended on the spectral information of images. Secondly, the study divided the entire image into three areas, which are darker area, normal area and lighter area, according to brightness value. Next, Linear-correlation correction （LCC） method was used in this study to transform darker area to normal area so that it can easily detect the landslides sites in darker area, and the object features, such as spectral, area, shape and space correlation indices, were used to extract landslide sites in images. Finally, in order to enhance the accuracy of landslide, the initial landslides were converted from image coordinate system to map coordinate system by ray-tracing method, so the initial landslides data can be further extracted by using topographic data, including slope and aspect data. The results of this study showed that the user and producer accuracies of detecting landslides can reach up to 82%. Moreover, the entire experiments process of this study can batch analyze automatically and establish landslides database quickly. It is expected that the method and landslides data of this study may have contribution to land conservation and disasters management.

本研究採用「台灣教育長期追蹤資料庫」（TEPS）調查結果，對台灣地區高中職學生個人特徵、家庭背景以及學生生活進行次級資料分析，探討各種因素與學業表現之關聯性。 　　本研究選取資料庫中之9,750樣本數，以資料庫綜合分析能力測驗成績作為應變數，選取性別及手足數目等12個自變數，以Logit Model為實證模型分析，再以Ordered Probit Model進行實證結果穩健性檢驗（Robustness Check），發現兩者結果呈現一致性，本研究發現： 一、男性成績較女性成績為佳，曾經或現在就讀資優班者成績較為優異。 二、家庭手足數目對學生學業成績表現有負向影響，雙親教育程度、家庭收入愈高及家庭文化資本愈豐富對子女學業表現有正向顯著影響；家庭親子關係、父母閱讀頻率對於子女學業表現則無顯著影響。 三、補習、參加社團活動以及校隊培訓對學業成績有正向影響；打工及結交行為偏差朋友則對學業成績有負向影響。 　　本研究提出教育政策方向建議如下： 一、建議文創產業主管機關以差別取價之價格誘因方式提高親子參與藝文活動之意願，並持續鼓勵文創產業提供學校學生及家長參與藝文展演活動機會。 二、教育主管機關應持續執行對（中）低收入戶子女學雜費補助之政策，並建立對高中職學校社團活動之管理制度，另應加強學生品性教育管理。 / This study analyzes the association between personal characteristics, family background, the life of students and academic achievement of students in senior high school. The data have adapted Taiwan Education Panel Survey (TEPS). The research has used 9,750 samples in the database. The dependent variable is the test scores of students while the independent variables are gender, the number of sibling, etc. This study has used Logit Model for empirical analysis and Ordered Probit Model for robustness check of the results. The research findings indicate that both models have similar conclusions, which are shown as follows. Firstly, the academic achievement of males is better than that of females. Moreover, students who are in the gifted and talented class have a better performance than those who are not in. Secondly, the number of sibling has a negative effect on academic achievement; whereas the education background of the parents, income and cultural capital of the family are as the opposite. The frequency of reading and the relationship between parents and children have no significant influence on academic achievement. Thirdly, the student who goes to cram school or participates in extracurricular activities has a positive influence on his/her academic achievement. However, the student with the part-time job or bad-behavior friends has a negative influence on his/her academic achievement. Therefore, the authorities should encourage parents and their children to participate in cultural activities through offering economic incentives and the tuition subsidies for low-income households continuously.

台灣和中國大陸於2010年六月正式簽署「經濟合作架構協議」（ECFA）。無疑地，這是兩岸交流有史以來規模最大、最具官方性質的制度性協商。在象徵意義上，意味著兩岸經濟整合邁入一個嶄新的階段。在實質意義上，透過早期收穫計畫的制度安排，使得兩岸之間的部分貨品和服務享有關稅調降的特殊待遇，對於台灣的整體經濟和部分產業具有立即性的影響。因此，本文旨在探討兩岸簽署ECFA前後，台灣民眾的統獨態度有無發生變化？在影響選民統獨態度因素中，有長期穩定與短期變動，也有感性與理性面向，選民對於兩岸簽訂ECFA的經濟效應評估又扮演了何種角色？最後，隨著兩岸經濟整合的腳步加速，理性層次的麵包效應又是否可能抵銷情感認同的作用？ 本研究使用2008年到2012年「台灣選舉與民主化調查」的定群追蹤資料（panel data），並以「固定與隨機效果並用法」（hybrid method of fixed and random effect model）來分析兩個時間點民眾統獨立場的動態變化。研究結果發現從08年到12年這段時間，民眾的統獨立場呈現往現狀/統一方向移動的趨勢，儘管變化的量不大，但在統計上卻是顯著的。在兩個時間點的動態架構中，選民對於ECFA經濟效益評估的態度變化，不僅與統獨立場的變遷模式與變化方向互相連動，在控制其他變數之後，ECFA經濟評估的態度變化對於統獨立場也有獨立性的影響效果。最後，當短期經濟利益和情感認同相互牴觸時，ECFA經濟的效果甚至會削弱感性認同的作用力。這意味在給定台灣人認同沒有改變的情況下，民眾仍有可能因為簽署ECFA的經濟因素而移動其統獨立場。因此，隨著兩岸經濟整合日益加深，影響個人統獨態度中理性層面利害考量的因素應該予以重視。 / The Taiwanese government has signed the ECFA with China in June 2010. Undoubtedly, ECFA is one of the largest and most official institutional negotiations in the history of cross-strait interactions and exchanges. Signing ECFA with China not only represents that cross-strait economic integration has entered into a new stage, through the arrangements of early harvest program, its impacts on Taiwan’s economy and industry are also immediate. Accordingly, this study aims to explore whether Taiwan people’s attitude toward the independence/unification changed or not after signing ECFA? How the economic inducement from China affect Taiwanese voters’ policy stances on independence/unification issue? Finally, as the accelerated pace of cross-strait economic integration, whether economic factors such as ECFA evaluation may offset the effects of emotional identity on the issue of independence/unification? By using individual panel data from ‘Taiwan Election and Democratization Study’ (TEDS), and taking advantage of hybrid method of fixed and random effect model, the empirical results show that respondents in 2012 are statistically significantly more inclined to maintain status quo or unification in comparison with their attitudes in 2008. Moreover, the attitudinal change of ECFA evaluation are not only systematically associated with the change of policy stance on independence/unification issue, it also reveals independent effect in the statistical model after controlling for other variables. Finally, although emotional affective identity is an important factor to determine public’s policy stances on the issue of Independence/Unification, its effects have begun to weaken especially when the economic interests are large and visible. The implication is that we shouldn’t underestimate the logic importance of political economy played in the trend of regional economic integration, and short-term economic fluctuations may have influence on long-term affective identity.

擴散權重影像與神經纖維追蹤可以用來探討腦區域之間的連結性，目前透過網路分析方式已經證實腦網路是有小世界的特性，最近也有研究不同受試者或者是病人之間的網路連結量測集中程度，但是擴散權重影像所運算出來的網路參數中間要經過很多步驟，這些中間步驟可能會影響到網路參數。所以有必要對於量測網路參數的受試者間變異性和重複量測重現性進行研究。本研究的目標是利用機率式神經纖維追蹤術量測大腦網路參數的重現性，探討三個會影響計算網路參數的重現性的變因，分別是，路徑定義方式、有無損耗正規化、受試者群體的網路連結篩選機制。變異係數定義（Coefficient of Variance, CV）為標準差除以平均值，分別計算二次量測之間的變異係數（CVwithin），以及受試者之間的變異係數（CVbetween），另外也計算組內相關係數（Intraclass correlation coefficient, ICC）。 掃描30受試者（15男，15女，年齡20~26）。每人掃描二次，並利用機率式神經纖維追蹤術計算網路連結，網路節點則是使用AAL標準模板定義的節點。若使用Wij = 1 – Pij定義長度，三項網路參數（區域效率、全域效率及損耗）重現性皆可接受（CVwithin<1.08%, CVwithin ≤ 10% and ICC > 0.7）。如果使用Wij=1/Pij定義長度，其損耗的CVwithin相較於Wij = 1 – Pij的大。如果長度的全距大，區域效率會不尋常地增加。如果二次掃描分別實施連結篩選，全域效率的CVwithin會較大。 本研究探討不同的網路建構方式將會影響測試內重現度，不同的研究團隊，縱使是採用相同的受試者群體和相同的儀器，所發表出來的網路參數可能會因為纖維追蹤術造成的誤差而不一致，因此實驗必須謹慎的分析資料以及闡述結果。 / Diffusion tensor imaging (DTI) with associate tractography can be used to access the connectivity of cortical regions in brain. Network analysis applied to connectivity matrix has demonstrated that brain has small world property. Recent studies also use network analysis to study the variation of concentricity among different group of subjects and patients. However the estimation of network metrics from DTI takes sophisticated processing steps. These intermediate steps may influence the estimation of network metric. It is therefore needed to investigate the potential variation of estimated network metrics using reproducibility test. The goal is to study the reproducibility of network properties derived from diffusion connectivity matrix constructed using probabilistic tractography. The effects of three factors on the reproducibility of network metrics estimation were studied. They are definition of path lengths of network matrix, path with and without cost normalization, the application of threshold to subjects groups. Coefficient of Variation (CV) defined as standard deviation divided by mean is used to test the intra-session (CVwithin) and inter subject (CVbetween) variability. Intra-class correlation coefficient (ICC) was also calculated. Images were acquired from 30 healthy participants (15 male, 15 female, aged 20-26 years). Each subject was scanned twice, denoted as N1 and N2. Probabilistic tractography was performed to mapping of cortico-cortical anatomical connections between regions defined from an anatomical atlas. All three of the tested network metrics (local efficiency, global efficiency and cost) were identified as acceptable (CVwithin < 1.08%, CVwithin ≤ 10% and ICC > 0.7) using path length defined as Wij = 1 – Pij. When the path length is defined as Wij = 1/Pij, cost showed higher CVwithin compared to Wij = 1 – Pij. It is unusual that local efficiency increase when the range of path length of edges is large. Global efficiency showed higher CVwithin as threshold is applied to N1 and N2 separately compared to both scans together. The present study revealed that different ways to construct cortical network had an effect on intra-session reproducibility. Our study also showed that despite evaluation of identical subjects using the same MRI system, variation of network metrics may be found by different research groups due to the potential errors from tractography. Replication of the experiment need to be carefully analyzed and interpreted.

跨國公司與開發中國家官員之賄賂關係被稱為骯髒聯盟。非國家行為者是形塑與改變國際規範之重要參與者，本研究檢視重點在於國際透明組織為切斷骯髒聯盟之供需關係，如何運用其獨特之跨國倡議網絡模式，透過建立聯盟(coalition-building)途徑，促成國際公約並且改變簽署國之國家行為。該非政府組織影響國際關係之具體成果為1997年簽署之OECD反賄公約，該公約是第一個從供應方(supply-side)切斷跨國商業賄賂鏈之國際規範，其立法歷程與會員國之國家偏好改變足以證明國際透明組織之影響力。 主流學派之結構現實主義者主張國際關係是結構決定論，以強權國家為主要的、具決定性之行為者，國家偏好是外生給定的利益排序。本研究主張以建構主義為基礎之全球反貪運動則是反其道而行，國際透明組織建構之跨國倡議網絡，擅用人氣政治與切身性政治，以合作取代對抗，有效促成已開發國家採取集體行動，簽署OECD反賄公約。當強權之間拒絕改變現狀時，非國家行為者之理念說服行動可以逐一改變國家偏好，進而改變國際規範。在倡議國際反貪議題上，國際透明組織的確是國際新規範之催生者。 本文採案例研究，主體為國際透明組織，客體為OECD反賄公約，交叉運用深度訪談與調查式過程追蹤研究，密集而深入取得相關當事人提供之第一手資料，彌補既有文獻靜態分析之不足。 / Non-state actors are essential participants in shaping and changing international norms. There has been a long-standing practice of a tacit “dirty alliance” between multinational corporations and developing countries officials. This research focuses on Transparency International(TI), which seeks to sever the supply and demand chain relationships of the dirty alliance by mobilizing a unique model of transnational advocacy network (TAN), engaging coalition-building, lobbying for a new set of international norms and persuading the changes of signatory states’ behavior. The fruits are the adoption of the 1997 OECD Anti-Bribery Convention, which is the first supply-side norm to cut off transnational business bribery chain. Both the key events of the legislative process and the changes of state preferences among the signatories validify the argument for the TI’s significant influence. Contrary to mainstream international relations theories such as structural realism claiming structural determinism, predominant role of powerful states, and exogenously given nature of state preferences, this research proves constructivism prevails. TI’s TAN model artistically exercises popularity politics as well as proximity politics, advocates cooperation instead of confrontation against bribe-givers and bribe-takers, and effectively persuades developed countries to take collective actions by adopting and ratifying the OECD Anti-Bribery Convention. While powerful states resist changing the status quo, non-state actors indeed demonstrate their capability of persuading states to change their preferences. TI is entitled to be the midwife in formulating new international norms against transnational corruption. This research applies case study methodology, making TI as the subject and the OECD Anti-Bribery Convention as the object. Diligent analysis and interpretations are based on the cross-references of in-depth interviews and investigative process-tracing method to extract the primary data from relevant parties so as to supplement the deficiencies of the available literatures.

偵測新興議題對於研究者而言是一個相當重要的問題，研究者如何在有限的時間和資源下探討同一領域內的新興議題將比解決已經成熟的議題帶來較大的貢獻和影響力。本研究將致力於協助研究者偵測新興且具有未來潛力的研究議題，並且從學術論文中探究對於研究者在做研究中有幫助的學術智慧。在搜尋可能具有研究潛力的議題時，我們假設具有研究潛力的議題將會由同一領域中較具有影響力的作者和刊物發表出，因此本研究使用貝式估計的方法去推估同一領域中相關的研究者和學術刊物對於該領域的影響力，進而藉由這些資訊可以找出未來具有潛力的新興候選議題。此外就我們所知的議題偵測文獻中對於認定一個議題是否已經趨於成熟或者是否新穎且具有研究的潛力仍然缺乏有效及普遍使用的衡量工具，因此本研究試圖去發展有效的衡量工具以評估議題就本身的發展生命週期是否仍然具有繼續投入的學術價值。 本研究從許多重要的資料庫中挑選了和資料探勘和資訊檢索相關的論文並且驗證這些在會議論文中所涵蓋的議題將會領導後續幾年期刊論文相似的議題。此外本研究也使用了一些已經存在的演算法並且結合這些演算法發展一個檢測的流程幫助研究者去偵測學術論文中的領導趨勢並發掘學術智慧。本研究使用貝式估計的方法試圖從已經發表的資訊和被引用的資訊來建構估計作者和刊物的影響力的事前機率與概似函數，並且計算出同一領域重要的作者和刊物的影響力，當這些作者和刊物的論文發表時將會相對的具有被觀察的價值，進而檢定這些新興候選議題是否會成為新興議題。而找出的重要研究議題雖然已經縮小探索的範圍，但是仍然有可能是發展成熟的議題使得具有影響力的作者和刊物都必須討論，因此需要評估議題未來潛力的指標或工具。然而目前文獻中對於評估議題成熟的方法僅著重在議題的出現頻率而忽視了議題的新穎度也是重要的指標，另一方面也有只為了找出新議題並沒有顧及這個議題是否具有未來的潛力。更重要的是單一的使用出現頻率的曲線只能在議題已經成熟之後才能確定這是一個重要的議題，使得這種方法成為落後的指標。 本研究試圖提出解決這些困境的指標進而發展成衡量新興議題潛力的方法。這些指標包含了新穎度指標、發表量指標和偵測點指標，藉由這些指標和曲線可以在新興議題的偵測中提供更多前導性的資訊幫助研究者去建構各自領域中新興議題的偵測標準。偵測點所代表的意義並非這個議題開始新興的正確日期，它代表了這個議題在自己發展的生命週期上最具有研究的潛力和價值的時間點，因此偵測點會根據後來的蓬勃發展而在時間上產生遞延的結果，這表示我們的指標可以偵測出議題生命力的延續。相對於傳統的次數分配曲線可以看出議題的崛起和衰退，本研究的發表量指標更能以生命週期的概念去看出議題在各個時間點的發展潛力。本研究希望從這些過程中所發現的學術智慧可以幫助研究者建構各自領域的議題偵測標準，節省大量人力與時間於探究新興議題。本研究所提出的新方法不僅可以解決影響因子這個指標的缺點，此外還可以使用作者和刊物的影響力去針對一個尚未累積任何索引次數的論文進行潛力偵測，解決Google 學術搜尋目前總是在論文已經被很多檢索之後才能確定論文重要性的缺點，學者總是希望能夠領先發現重要的議題或論文。然而，我們以議題為導向的檢索方法相信可以更確實的滿足研究者在搜尋議題或論文上的需求。 / This research presents endeavors that seek to identify the emerging topics for researchers and pinpoint research intelligence via academic papers. It is intended to reveal the connection between topics investigated by conference papers and journal papers which can help the research decrease the plenty of time and effort to detect all the academic papers. In order to detect the emerging research topics the study uses the Bayesian estimation approach to estimate the impact of the authors and publications may have on a topic and to discover candidate emerging topics by the combination of the impact authors and publications. Finally the research also develops the measurement tools which could assess the research potential of these topics to find the emerging topics. This research selected huge of papers in data mining and information retrieval from well-known databases and showed that the topics covered by conference papers in a year often leads to similar topics covered by journal papers in the subsequent year and vice versa. This study also uses some existing algorithms and combination of these algorithms to propose a new detective procedure for the researchers to detect the new trend and get the academic intelligence from conferences and journals. The research uses the Bayesian estimation approach and citation analysis methods to construct the prior distribution and likelihood function of the authors and publications in a topic. Because the topics published by these authors and publications will get more attention and valuable than others. Researchers can assess the potential of these candidate emerging topics. Although the topics we recommend decrease the range of the searching space, these topics may so popular that even all of the impact authors and publications discuss it. The measurement tools or indices are need. But the current methods only focus on the frequency of subjects, and ignore the novelty of subjects which is critical and beyond the frequency study or only focus one of them and without considering the potential of the topics. Some of them only use the curve of published frequency will make the index as a backward one. This research tackles the inadequacy to propose a set of new indices of novelty for emerging topic detection. They are the novelty index (NI) and the published volume index (PVI). These indices are then utilized to determine the detection point (DP) of emerging topics. The detection point (DP) is not the real time which the topic starts to be emerging, but it represents the topic have the highest potential no matter in novelty or hotness for research in its life cycle. Different from the absolute frequent method which can really find the exact emerging period of the topic, the PVI uses the accumulative relative frequency and tries to detect the research potential timing of its life cycle. Following the detection points, the intersection decides the worthiness of a new topic. Readers following the algorithms presented this thesis will be able to decide the novelty and life span of an emerging topic in their field. The novel methods we proposed can improve the limitations of impact factor proposed by ISI. Besides, it uses the impact power of the authors and the publication in a topic to measure the impact power of a paper before it really has been an impact paper can solve the limitations of Google scholar’s approach. We suggest that the topic oriented thinking of our methods can really help the researchers to solve their problems of searching the valuable topics.

本論文提出追蹤特定目標線的二階段混合整數非線性隨機規劃模型，以建立追蹤目標線的投資組合。藉由引進情境樹(scenario tree)，我們將此類二階段隨機規劃問題，轉換成為等價的非隨機規劃模型。在金融商品的價格波動及交互作用下，所建立的投資組合在經過一段時間後，其追蹤目標線的能力可能會日趨降低，所以本論文亦提出調整投資組合的規劃模型。為符合實務考量，本論文同時考慮交易成本、股票放空的限制，並且加入期貨進行避險。為了反應投資者的預期心理，也引進了選擇權及情境樹。最後，我們使用台灣股票市場、期貨交易市場及台指選擇權市場的資料進行實證研究，亦探討不同成長率設定之目標線與投資比例對於投資組合的影響。 / To construct a portfolio tracking specific target line, this thesis studies how to do it via two-stage stochastic mixed-integer nonlinear model. We introduce scenario tree to convert this stochastic model into an deterministic equivalent model. Under the volatility of price and the interaction of each financial derivatives, the performance of the tracking portfolio may get worse when time elapses, this thesis proposes another mathematical model to rebalance the tracking portfolio. These models consider the transactions cost and the limitation of shorting a stock, and the tracking portfolio will include a futures as a hedge position. To reflect the expectation of investors, we introduce scenario tree and also include a options as a hedge position. Finally, an empirical study will be performed by the data from Taiwan stock market, the futures market and the options market to explore the performance of the proposed models. We will analyze how the different benchmarks settings and invest ratio will affect the value of the tracking portfolio.

本研究使用「臺灣教育長期追蹤資料庫」第二波（2003）國三與第三波（2005）高二的CP追蹤數據，採用反事實模型框架下的傾向分數配對法估計單一性別環境對臺灣高中女生選組行爲（自然組／社會組）的因果效應。使用學生進入高中前的國三資料，本研究在學生因素，家庭因素和學校、班級因素三個方面充分平衡處置組（進入女校）和控制組（進入男女合校）的樣本，使配對樣本之間除了進入女校或進入男女合校之外，在以上三個方面儘可能相等，嘗試克服觀察性數據因選擇性偏誤問題對因果效應的估計帶來的阻礙。結果顯示，在進行傾向分數配對後，處置組和控制組之間達到了很好的平衡，進入女校的女生與進入男女合班的男女合校中的女生相比，女校顯著地促進了高中女生選自然組的機率，然而該效應在數學成績水平不同的女生中存在異質性，女校顯著地促進了數學成績處於高水平的女生的選擇自然組的行爲。作爲教育分流的重要組成部分，臺灣高中生選組行爲的性別隔離長期存在，選組行爲與學生未來大學科系選擇和職業選擇關係密切，自然組中低比例的女生組成顯現出隱藏的教育機會不平等。該現象長期存在的背後因素，除了學生個人因素和學生家庭背景因素帶來的影響，本研究關注學校這一角色對高中女生選組行爲的影響。 / Based on the Taiwan Educational Panel Survey’s core panel data from wave 2 (junior high school/9th grade) and wave 3 (high school/11th grade), researcher applies the propensity score analysis in counterfactual framework to study the causal effect of single-sex schooling on high school girls’ curriculum tracking (science track or humanity track) selection. Observational data always be the obstacles of making the causal analysis because it's lacking random assignment and being under threat of selection bias and unobserved variables. Using the pretreatment variables from students in 9th grade, students in the treatment group were matched with those in control groups by the individual factors, family factors and school, classroom environment factors. After matching, students come from different groups are supposed to be relatively equivalent on all the matching variables except for attending single-sex school or not. The researcher can make suitable comparison and prediction between well-matched samples. The result shows that more girls in single-sex high schools tend to choose science track. Effect of single-sex schooling is not the same among girls with different level of math scores. It benefits girls with top math performance significantly. As an important part of the educational tracking system, alternative curriculum tracking selection in Taiwan high school is worth noteworthy. The long lasting gender segregation is closely associated with students college major as well as occupation selection and it reveals the hidden educational inequality between gender. Other than focusing on the individual and family background effect on high school girls’ curriculum tracking selection, this study highlights the school effect on it.