<p>Statistical models often rely on several assumptions including distributional assumptions on outcome variables and relational assumptions where we model the relationship between outcomes and independent variables. Further assumptions are also made depending on the complexity of the data and the model being used. Model diagnostics is, therefore, a crucial component of any model fitting problem. Residuals play important roles in model diagnostics. Residuals are not only used to check adequacy of model fit, but they also are excellent tools to validate model assumptions as well as identify outliers and influential observations. Residuals in univariate models are studied extensively and are routinely used for model diagnostics. In multivariate models residuals are not commonly used to assess model fit, although a few approaches have been proposed to check multivariate normality. However, in the analysis of longitudinal data, the resulting residuals are correlated and are not normally distributed. It is, therefore, not clear as to how ordinary residuals can be used for model diagnostics. Under sufficiently large sample size, a transformation of ordinary residuals are proposed to check the normality assumption. The transformation is based solely on removing correlation among the residuals. However, we show that these transformed residuals fail in the presence of model mis-specification. In this thesis, we investigate residuals in the analysis of longitudinal data. We consider ordinary residuals, Fitzmaurice’s transformed (uncorrelated) residuals as well as von Rosen’s decomposed residuals. Using simulation studies, we show how the residuals behave under multivariate normality and when this assumption is violated. We also investigate their properties under correct fitting as well as wrongly fitted models. Finally, we propose new residuals by transforming von Rosen’s decomposed residuals. We show that these residuals perform better than Fitzmourice’s transformed residuals in the presence of model mis-specification. We illustrate our approach using two real data sets.</p> / Master of Science (MSc)

5 pp. / Eight different landscape mulches were tested for their flammability using a propane torch, charcoal briquette, and a cigarette at two different times of the year. Three randomized compete blocks with eight one square meter plots were tested at three locations; Tucson, Prescott, and Flagstaff, Arizona. Each of the mulches was subjected to the heat of a handheld propane torch (15 seconds), a glowing charcoal briquette (five minutes), and a lit cigarette (until burned out). We found that the least dense mulches (pine needles and straw) burned rapidly when subjected to the torch and ignited after the briquette was removed. The medium density mulches (pine bark nuggets and wood chips) had low flame lengths and smoldered. Heavy density mulches (garden compost and shredded bark) only smoldered. The decomposed granite and sod did not ignite or smolder.

Amplitude-versus-offset, AVO, approximations allow the estimation of various properties from pre-stack seismic gathers. Recently it has been suggested that fluid mobility is a controlling factor in pore pressure equalisation and can result in anomalous velocity dispersion in the seismic bandwidth. However, current approximations all assume an elastic subsurface and are unable to account for velocity dispersion. I have applied existing methodologies to a real dataset to qualitatively detect and interpret spectral amplitude anomalies. Three areas had AVO and spectral signature consistent with frequency-dependent AVO theory. The results suggest that it is feasible to measure such effects on real data in the presence of random noise. It would imply that the relaxation parameter, tau, is larger in the field than has been measured in water-saturated real and synthetic sandstones in the laboratory. I extended a two-term AVO approximation by accounting for velocity dispersion and showed how the resultant reflection coefficient becomes frequency-dependent. I then used this to measure P- and S-wave reflectivity dispersion using spectrally-balanced amplitudes. The inversion was able to quantify the affect of the P-wave velocity dispersion as an instantaneous effect on the reflection. NMO stretch was an issue at the far offsets and I limited myself to the near offsets and effectively measured only the P-wave reflectivity dispersion. I showed how the P-wave reflectivity dispersion signs depend on the AVO classification of the reflection whilst the magnitude depends on the crack density of my model. I showed how the effect of noise and thin-bed tuning can enter uncertainties into the interpretation of spectral anomalies. Whilst it is possible to detect frequency-dependent AVO signatures on pre-stack gathers, the interpretation remains non-unique. I have quantitatively measured a new physical property, reflectivity dispersion, from pre-stack seismic data. I have presented a method of detecting and measuring velocity dispersion in pre-stack gathers but there remain ambiguities in the interpretation of such results. The approach incorporates spectrally decomposed data in an extended AVO inversion scheme. Future work should investigate the application of the methodology to a real seismic dataset.

L’industrie aéronautique nécessite l’utilisation de métaux durs et légers, comme le Titane et l’Inconel mais différentes problématiques d’usinage en fraisage ou par usinage chimique compliquent leur mise en oeuvre. L’objectif du travail de thèse est de développer une méthode d’usinage non débouchant de ces métaux durs par jet d’eau abrasif (JEA) pour la réalisation de poches. Actuellement, ce procédé est limité à la découpe mais en contrôlant les paramètres d’usinage il permet la réalisation de poches à profondeurs constante ou variable en maitrisant les caractéristiques géométriques imposées. Ce travail présente l’étude de passages élémentaires et leur superposition afin de générer un parcours nécessaire à la réalisation d’une poche. Il introduit également diverses stratégies pour l’usinage de poches rectangulaires et propose une correction des paramètres opératoires afin d’obtenir un fond à profondeur contrôlée. / The aviation industry requires the use of hard and light metals such as Titanium and Inconel, but different machining problems in milling process or by chemical machining complicate their implementation. The aim of this thesis is to develop a controlled depth milling on these hard metals with abrasive water jet (AWJ) technology for pockets machining. Currently, this process is limited to cutting operation but by controlling the input parameters and the geometric characteristics it allows the realization of pockets with constant or variable depth. This work shows the study of elementary passages and their superposition to generate a necessary path for the realization of a pocket in AWJ. It also introduces a variety of strategies for machining rectangular pockets and provides a correction of the operating parameters to obtain a controlled depth.

Recently, it has been admitted by many researchers that students today are "digital natives" who already utilize several different technologies everyday with different purposes. Furthermore, while there is a plethora of research about learners' perceptions in language pedagogy, there is a paucity of information and research that could move beyond generic perception studies especially regarding new technologies. Accordingly, educators are still concerned not only with how to encourage EFL learners to adopt emerging technologies that could be invaluable in their language learning processes, but also with how to keep students interested in what they are learning. Thus, an understanding beyond students' perceptions with a purposive focus on their approach to technology by also exploring factors that have an influence on their adoption of emerging technologies is the key to knowing how to motivate students to integrate new technologies, and how to keep students interested in the learning process. Therefore, the purpose of this study was three-fold: to identify language learners' approaches to technologies; to examine their attitude toward emerging technologies with a focus on their familiarity, actual use, intentions and perceptions; and finally to analyze the factors and relationships among these factors that best predict language learners' intentions and decisions to use emerging technologies. The study specifically aimed to explore the following emerging technologies: (a) social networking, (b) mobile learning, and (c) digital games as major emerging technologies of today with also a focus on other emerging technologies: (a) augmented reality, (b) wearable technologies, (c) virtual assistants, (d) massive online open courses, (e) 3D printing and (f) online language learning platforms. This mixed methods study benefitted from multiple disciplines, and presented several different perspectives to achieve its aim. The data were obtained through a survey, open-ended questions, and semi-structured interviews. The quantitative data were analyzed through descriptive and inferential statistics. In addition, structural equation modeling was utilized and path analysis was employed to draw on two complementary frameworks: innovation diffusion theory (Rogers, 2003), especially its application in technology adopter categories (TACI) (Dugas, 2005), and the decomposed theory of planned behavior (DTPB) (Taylor & Todd, 1995), which was adapted for this research. The qualitative data was analyzed through thematic content analysis, and used to triangulate and affirm what the quantitative data was showing. The findings indicated that technology adopter categories were normally distributed among EFL learners at a public and a private university in Turkey. Although most EFL learners were quite familiar and confident with major emerging technologies, they were not very familiar with minor emerging technologies. As for an awareness of the benefits of all listed emerging technologies, the study revealed that EFL learners' awareness is quite high; however, it was also found that participants' intention to use these emerging technologies and their actual use were very low. Finally, the results showed that the adapted DTPB was useful in explaining much of the variance in the intention to integrate technology into language learning processes by EFL learners, and attitude was the most important predictor and factor of behavioral intention. Given these findings, this research aims to contribute to the literature in innovation diffusion, user adoption and language pedagogy by offering several theoretical, methodological, and pedagogical implications and directions for future research and applications.

This study evaluates the adoption of Open Source Software (OSS) by IT Small to Medium-sized Enterprises (SMEs) in the UK. The growing popularity and acceptance of OSS continues to draw much attention in research and practice. However, researchers and IT practitioners within the UK SME sector still face challenges in understanding the issues that influence the acceptance, adoption, and diffusion of OSS. While previous research studies have focused mainly on the software development model and the unique characteristics of OSS, the area of OSS adoption by UK SMEs has largely been ignored. Furthermore, there is a lack of widely-acceptable theories that explain the adoption of OSS, implying that there is limited understanding of OSS adoption by UK SMEs. This gap in research has led this thesis to evaluate existing adoption theories and then apply the 'Decomposed Theory of Planned Behaviour' to model the adoption of OSS by SMEs. Based on the emerged conceptual model, an innovative and structured qualitative research design that uses a case study strategy was developed to evaluate the adoption of OSS across 10 UK SMEs in the IT industry. The analysis of the standardised data from the case study interviews led to the definition of the 16 factors of an emergent theory of OSS adoption by IT SMEs. The analysis of that empirical model has led to important conclusions including the following five issues, summarily. (1) The participant IT SMEs were drawn to different benefits, and experienced different challenges, in using OSS, suggesting that there is subjectivity and complexity in the factors influencing OSS adoption. (2) As in most Information and Communication Technology (ICT) adoption, ITcapability was identified to be essential for successful adoption of OSS, and therefore, it presents potential for important cooperative and collaborative support with OSS communities. (3) The emergent theory from this research study provide researchers and practitioners with variables for surveying critical-success-factors and a reference model for understanding the adoption of OSS. (4) The emergent theory and other general findings from this study are likely to have relevance in other areas of Information Systems research and practice, owing to the factors and theoretical framework that are common to OSS and general ICT acceptance, adoption, and diffusion. (5) This study appears to be the first that has focused on developing a widely-acceptable theory of OSS adoption by IT SMEs in the UK, suggesting that this innovative research study is a novel contribution that has important implications for theory and practice in OSS and general ICT acceptance, adoption, and diffusion.

Today¡¦s motorcycle business has come to the saturation point in the market of Tawian; consequently, the major motorcycle companies recently competed with each other in building the DMS (Dealer Management System) by using the e-solutions. Through the deployment of an e-DMS (e-solutions of Dealer Management System) for shops of motorcycle, the manufacturers hope that all the channels could be more competitive. The purpose of this research is to explore the influence factors concerning the adoption of e-DMS of motorcycle¡¦s shops. After the studies of literature and empiric, the research is based on ¡§Decomposed Theory of Planned Behavior¡¨ (Taylor and Todd, 1995b) to establish the research model.This resrerch suveryed 250 samples of motorcycle¡¦s shops for study cases The result of the research indicated that factors influenced the adoption of e-DMS for motorcycle shops as follows: (1).¡§Behavioral Intention¡¨ was principally influenced by ¡§Attitude¡¨ and ¡§Perceived Behavioral Control¡¨. The later was less important than the former. ¡§Subject Norms¡¨ showed no obvious influence. (2).¡§Attitude¡¨ was mainly influenced by ¡§Perceived Usefulness¡¨, ¡§Perceived Ease of Use¡¨ and ¡§Compatibility¡¨. The first two factors were more important than the last one. (3).¡§Perceived Behavioral Control¡¨ was chiefly influenced by ¡§Self-efficacy¡¨ and ¡§Technology Facilitating Conditions¡¨. The later was less essential than the former. ¡§Resource Facilitating Conditions¡¨ showed no apparent influence. In the end, this research checks explanation by using three acceptance models, TAM (Davis, 1989), TPB (Ajzen, 1985) and D-TPB (Taylor and Todd, 1995b) to evaluate. All the explanations were nearly close. Because D-TPB considered the contruct of society psychology, it shows better explanation than the others.

Abstract Purpose - The purpose of this master thesis is to research the Generation Y’s online purchase intention for fashion items while creating and using a customised avatar. Overall, the objective is to create a better understanding of this technology’s potential, formulate managerial implications for fashion businesses and strengthen business viability. Design/Methodology/Approach - The research approach of this study is deductive, whereby hypotheses derive from the Decomposed Theory of Planned Behaviour. After secondary data is reviewed, a single quantitative data collection is applied, thus following a mono-method. This primary data is gathered virtually through a self-administered online questionnaire. A total number of 205 qualified responses from the Generation Y are statistically analysed using a structural equation modelling. This descriptive research design is chosen to conduct the relationships between the latent variables and the behavioural intention. Findings - The empirical findings reveal, that the attitude, subjective norm and perceived behavioural control significantly and positively influence the Generation Y’s online purchase intention to create and use a customised avatar. While the attitude, with the behavioural belief of perceived usefulness specifically, shows the strongest influence on the behavioural intention, the research sample also sees a fit to all technology facilitating conditions, affecting the perceived behavioural control. In comparison to this, the subjective norm influences the behavioural intention in a weaker manner, whereby the research sample is influenced more by external than interpersonal factors. Implications - To enlarge the Generation Y’s online fashion purchase intention while creating and using a customised avatar, fashion marketers are advised to highlight and improve the usefulness of the technology. Fashion businesses are recommended to implement interactive digital platforms, by employing influencer marketing, in order to endorse and promote the brand awareness in regard to the technology. Originality/Value - This master thesis addresses the online purchase intention for fashion items while creating and using a customised avatar from a commercial perspective. Where prior literature findings lack the link to managerial implications, this study examines the Generation Y’s behavioural intention towards this technology. The Generation Y has an immense and increasing purchasing power, which is accompanied with technical skills, thus making them crucial for the market success of online fashion businesses. Therefore, the authors examine the technology's commercial potential and encompass the whole fashion industry.

No description available.

新科技的變遷總是為傳播產業帶來新的變革，智慧型手機掀起全球通訊產業的熱潮，造就了行動應用程式新經濟，進而帶動行動應用程式廣告的發展。有鑑於過去研究缺乏以解構計畫行為理論的角度，來探討近年才趨於熱門的行動意用程式，更缺少應用在某特定行動應用程式類型之行為意圖。本研究將以Taylor & Todd（1995)解構計畫行為理論，聚焦於社群媒體廣告研究，並以在社交類型App的龍頭Facebook為研究對象，試圖釐清Facebook行動用程式廣告之行為意圖。 本研究採線上問卷調查法，以非隨機抽樣中的立意抽樣（purposive sample）為蒐集資料的方法，以結構方程模式（Structural Equation Modeling, SEM）為資料的統計分析方法。問卷發放於Facebook以及台大批踢踢實業坊（ptt.cc）符合研究主題的討論區進行資料蒐集，問卷連結日期自二0一五年一月一日至一月十六日，為期十六天，有效問卷共計五百一十一份。研究結果發現：(1) Facebook App的知覺有用性、相容性和知覺娛樂性，與廣告整體態度無顯著關係。(2)廣告內容的「娛樂性」與「個人化」有助於廣告態度的影響；「資訊性」與廣告態度無顯著影響，「煩躁性」與廣告態度則呈現負相關。(3)人際影響、自我效能和廣告態度，與行為意圖皆呈現正相關。 本研究證實，對於使用者來說，相較於電腦版使用經驗與習慣，Facebook App仍有改善空間。此外，Facebook App的廣告內容是否具有完整的廣告資訊已不再首要因素，重要的是，該廣告是否具有娛樂性質和符合個人需求，在不讓使用者感到困擾的情況前提之下，吸引使用者的注意、滿足使用者的需求。同時，本研究證實人際影響、自我效能和廣告態度對行為意圖的正向影響性。 / Technology advancement has brought many new changes to the mass media communication industry. Smartphones revolutionized the communication industry by creating the mobile app economy, which lead to the development of mobile app advertisement. Past studies have not taken the decomposed theory of planned behavior into consideration when examining mobile applications that have gained popularity in recent years, let along behavioral intentions, which apply to certain types of mobile applications. Keeping that in mind, this study will focus on advertising within the current social networking industry leader, Facebook, using the above theory by Taylor & Todd (1995), in order to determine the behavioral intention behind Facebook’s mobile application advertisement. Data used in this study was collected through online questionnaires using the purposive sampling method, and analyzed using structural equation modeling. The questionnaires were distributed on Facebook and sections within ptt.cc that fit the research purpose of this study. Online questionnaire was open from January 1st to January 16th of 2015, spanning a total of 16 days. A total of 511 valid questionnaires were collected. The results of this studies revealed that (1) the perceived usefulness, compatibility, and enjoyment of Facebook app are not related to attitude towards advertisement within the app. (2) Entertainment value and personalization of the content of the advertisement positively affects attitude toward advertisement. Informativeness has no significant impact, while irritation negatively affects attitude towards advertisement. (3) Social influence, self-efficacy and attitude towards advertisement all show positive relation with behavioral intention. This research has proven that when examining user experience, Facebook app still has room for improvement in comparison with the computer version of Facebook. Also, whether Facebook app advertisement contains complete information is no longer the most important factor. Entertainment value and personalization, whether the advertisement can satisfy the above needs without troubling users, are now the most important points of consideration. Finally, this study has proven the positive effect social influence, self-efficacy and attitude towards advertisement have on behavioral intention.