Several studies in hydrology have reported differences in outcomes between models in which spatial autocorrelation (SAC) is accounted for and those in which SAC is not. However, the capacity to predict the magnitude of such differences is still ambiguous. In this thesis, I hypothesized that SAC, inherently possessed by a response variable, influences spatial modeling outcomes. I selected ten watersheds in the USA and analyzed them to determine whether water quality variables with higher Moran’s I values undergo greater increases in the coefficient of determination (R²) and greater decreases in residual SAC (rSAC) after spatial modeling. I compared non-spatial ordinary least squares to two spatial regression approaches, namely, spatial lag and error models. The predictors were the principal components of topographic, land cover, and soil group variables. The results revealed that water quality variables with higher inherent SAC showed more substantial increases in R² and decreases in rSAC after performing spatial regressions. In this study, I found a generally linear relationship between the spatial model outcomes (R² and rSAC) and the degree of SAC in each water quality variable. I suggest that the inherent level of SAC in response variables can predict improvements in models before spatial regression is performed. The benefits of this study go beyond modeling selection and performance, it has the potential to uncover hydrologic connectivity patterns that can serve as insights to water quality managers and policy makers.

In this thesis a regression analysis of ten independent data sets is analysed in order toestimate losses and Key Risk Indicators (KRI). Each data set contains a list of objects,impacts that each object contains and revenue stream values (RSV) to each impact.The project investigates the data and simulate yearly losses as response variables in theregression modelling. The three regressors that influence the yearly losses are numberof objects, sum of revenue streams and expected aggregated losses. Given the responsevariable from each data set a percentage scale of KRI’s is determined indicating howlarge losses each set possess. / I denna avhandling analyseras en regressionsmodellering av tio oberoende mängderdata för att uppskatta förluster och Key Risk Indicators. Den givna dataupsättningeninnehåller en lista med objekt, påverkan varje objekt erhåller och vad respektiveobjekt omsätter. Projektet undersöker den givma datan och simulerar årliga förlustersom svarsvariabler i regressionsmodelleringen. De tre regressorerna som påverkarde årliga förlusterna är antalet objekt, summan av intäckterna och förvämtadesammanlagda förlusterna. Från den givna svarsvariabeln från varje datamängdbestäms en procentuell skala av KRIer som indikerar hur stora förluster varjeuppsättning har.

本研究旨在探討國中教師的經師角色（投入教學）、人師角色（投入輔導）、導師角色（注重班級成績、注重班級整潔秩序）等變項對於國中生偏差行為的貫時性影響。研究資料取自中央研究院調查研究專題中心學術調查研究資料庫中的「台灣教育長期追蹤資料庫」第一波國中學生問卷【限制版原始數據】、第二波國中學生問卷【限制版原始數據】、第一波國中家長問卷【限制版原始數據】、第二波國中教師問卷【限制版原始數據】。 本研究又分為兩個子研究。研究一使用兩波問卷中的20,055筆國中生問卷及家長問卷，研究二使用兩波問卷中的8,794筆國中生問卷、家長問卷、導師問卷。本研究採用Rasch模式垂直等化、結構方程模式、階層卜瓦松模式分析資料，研究結果如下： 一、「人師角色」「經師角色」對國中生偏差行為影響的貫時性分析 （一）國一教師投入教學對於追蹤樣本國一學生偏差行為有顯著影響，教師越認真教學，學生偏差行為分數越低。 （二）國一教師投入教學對於追蹤樣本國三學生偏差行為有貫時性顯著影響，國一教師越認真教學，國三時追蹤樣本學生的偏差行為分數越低。 （三）國一二教師投入輔導對於追蹤樣本國三學生偏差行為有顯著影響，國一國二教師越投入輔導，追蹤樣本學生國三時的偏差行為分數越低。 二、「導師角色」對國中生偏差行為影響的貫時性分析 （一）追蹤樣本國三偏差行為顯著較該生國一時嚴重。 （二）偏差行為與性別有關，男生偏差行為分數顯著比女生高。 （三）追蹤樣本國一至國三偏差行為惡化程度與性別有關，男生偏差行為惡化情形顯著較女生嚴重。 （四）追蹤樣本國一至國三偏差行為惡化程度與該生國一時的學業成就有關。國一成績較差，國三偏差行為惡化程度顯著較嚴重。 （五）追蹤樣本國一至國三偏差行為惡化程度與該生國一時的讀書習慣有關。國一讀書習慣較差者，國三偏差行為惡化程度顯著較嚴重。 （六）導師要求班級學業成績，對於國中生偏差行為有顯著影響，導師越注重學業成績，該班學生偏差行為越低，且該影響具貫時性。 （七）導師要求班級整潔秩序等生活常規，對於國中生偏差行為有顯著影響，導師越要求班級整潔、秩序，該班學生偏差行為越低，且該影響具貫時性。 （八）國中生偏差行為與就讀學校公私立與否有關。私立學校學生偏差行為分數顯著較公立學校學生為低，且該影響具貫時性。 研究者並提出建議供教育實務工作者及教育主管單位參考。 / The purposes of the study were to explore teacher roles including instruction role, guidance role and headroom teacher’s role, and to find out the effects of teacher roles on junior high school students' deviant behavior. The data used in the analysis are from the nationally representative 2001–2003 Taiwan Education Panel Survey, including students’ questionnaire, students’ academic tests, parents’ questionnaire, and teachers’ questionnaire. In study 1, data from a 2-wave panel (N = 20,055), tested in the 7th and 9th grades, were used to estimate a structural equation model. The model was used to compose a previously observed teacher’s instruction role (teaching engagement) and guidance role (guidance engagement) in wave 1. In study 2, data also from a 2-wave panel (N=8,794), were to estimate a hierarchical Poisson regression model, in which independent variables were achievement-oriented headroom teacher role and discipline-oriented headroom teacher role, and dependent variable was students' deviant behavior. The results indicate that teachers who devoted themselves in teaching and guiding students could reduce junior high students’ deviant behavior, and the effects were longitudinal for at least 2 years. Also, both achievement-oriented and discipline-oriented headroom teachers could reduce junior high students’ deviant behavior, and the effects were longitudinal. Implications for current practice and future research are also discussed.

No description available.