In two-level fractional factorial designs, homogeneous variance is a commonly made assumption in analysis of variance. When the variance of the response variable changes when a factor changes from one level to another, we called the factor dispersion factor. Formerly, many researches have discussed about how to define the dispersion effects, but the problem of finding optimal designs when dispersion effects present is relatively unexplored. However, a good design not only save the experiment cost but also let the estimation more efficiency. In this research, we focus on finding optimal designs for the estimation of location main effects when there are one or two dispersion factors, in the class of regular unreplicated two-level fractional factorial designs of resolution Ⅲ and higher. We show that by an appropriate choice of the defining contrasts, A-optimal and D-optimal designs can be identified. Efficiencies of an arbitrary design are also investigated.

<p>The defining equations of Rees algebras provide a natural pathway to study these rings. However, information regarding these equations is often elusive and enigmatic. In this dissertation we study Rees algebras of particular classes of ideals and modules over hypersurface rings. We extend known results regarding Rees algebras of ideals and modules to this setting and explore the properties of these rings.</p> <p><br></p> <p>The majority of this thesis is spent studying Rees algebras of ideals in hypersurface rings, beginning with perfect ideals of grade two. After introducing certain constructions, we arrive in a setting similar to the one encountered by Boswell and Mukundan in [3]. We establish a similarity between Rees algebras of ideals with linear presentation in hypersurface rings and Rees algebras of ideals with <em>almost</em> linear presentation in polynomial rings. Hence we adapt the methods developed by Boswell and Mukundan in [3] to our setting and follow a path parallel to theirs. We introduce a recursive algorithm of <em>modified Jacobian dual iterations</em> which produces a minimal generating set for the defining ideal of the Rees algebra.</p> <p><br></p> <p>Once success has been achieved for perfect ideals of grade two, we consider perfect Gorenstein ideals of grade three in hypersurface rings and their Rees algebras. We follow a path similar to the one taken for the previous class of ideals. A recursive algorithm of <em>gcd-iterations</em> is introduced and it is shown that this method produces a minimal generating set of the defining ideal of the Rees algebra. </p> <p><br></p> <p>Lastly, we extend our techniques regarding Rees algebras of ideals to Rees algebras of modules. Using <em>generic Bourbaki ideals</em> we study Rees algebras of modules with projective dimension one over hypersurface rings. For such a module $E$, we show that there exists a generic Bourbaki ideal $I$, with respect to $E$, which is perfect of grade two in a hypersurface ring. We then adapt the techniques used by Costantini in [9] to our setting in order to relate the defining ideal of $\mathcal{R}(E)$ to the defining ideal of $\mathcal{R}(I)$, which is known from the earlier work mentioned above.</p> <p><br></p> <p>In all three situations above, once the defining equations have been determined, we investigate certain properties of the Rees algebra. The depth, Cohen-Macaulayness, relation type, and Castelnuovo-Mumford regularity of these rings are explored.</p>

Existing commonly used maternal-fetal attachment instruments have not been thoroughly tested for reliability and validity; criticism can be levelled for a variety of problems ranging from lack of reliability due to an inadequate underpinning framework to facilitate objective interpretation to limited generalizability due to the sample. The aim of this study is to acknowledge the centrality of the mother, to use the experiences of pregnant women to generate a definition of maternalfetal attachment and ultimately create a tool that will act as a reliable, valid and simple measurement. A mixed method framework utilising a sequential exploratory strategy has allowed qualitative exploration of the phenomenon under investigation followed by quantitative testing of the emerging theory on a much larger and different sample. Phase 1 involved face to face open structured interviews on an opportunity sample of 10 (5 primigravid; 5 multiparous) women in the final trimester of pregnancy followed by 3 focus groups targeting specific groups – primigravid women (6 participants); multiparous women (7 participants) and teenagers (4 participants). Data analysis was through constant comparative methodology. A multidimensional, psycho-biological definition of attachment was generated from the women’s own perception of their attachment to their fetus. This was used as a framework to design a questionnaire for the measurement of maternal-fetal attachment. Phase 2 involved the validation of the questionnaire and further testing of the definition. Cohort 1 tested for reliability with 200 participants within their second or third trimester of pregnancy. Following modification of the questionnaire, Cohort 2 a sample of 150 women within the final trimester of pregnancy tested the tool for internal reliability and validity. The generated Maternal-Fetal Attachment Tool (MFAT) following rigorous testing proved both reliable and valid. Maternal fetal attachment is founded in psycho-biological theory and is a complex multi-dimensional construct. Central to the definition is the woman’s need to protect her fetus, attachment develops as the fetus becomes more tangible, it is facilitated through the woman’s intergenerational experience of attachment and through appropriate social support. Maternal-fetal attachment facilitates behavioural change to ensure a favourable intra-uterine environment.

Master of Science / Department of Industrial and Manufacturing Systems Engineering / Todd Easton / Graph theory is a widely researched topic. A graph contains a set of nodes and a set of edges. The nodes often represent resources such as machines, employees, or plant locations. Each edge represents the relationship between a pair of nodes such as time, distance, or cost. Integer programs are frequently used to solve graphical problems. Unfortunately, IPs are NP-hard unless P = NP, which implies that it requires exponential effort to solve them. Much research has been focused on reducing the amount of time required to solve IPs through the use of valid inequalities or cutting planes. The theoretically strongest cutting planes are facet defining cutting planes. This research focuses on the node packing problem or independent set problem, which is a combinatorial optimization problem. The node packing problem involves coloring the maximum number of nodes such that no two nodes are adjacent. Node packings have been applied to airline traffic and radio frequencies. This thesis introduces a new class of graphical structures called suns. Suns produce previously undiscovered valid inequalities for the node packing polyhedron. Conditions are provided for when these valid inequalities are proven to be facet defining. Sun valid inequalities have the potential to more quickly solve node packing problems and could even be extended to general integer programs through conflict graphs.

<div>In the first part of this thesis, we study Rees algebras of modules. We investigate their Cohen-Macaulay property and their defining ideal, using <i>generic Bourbaki ideals</i>. These were introduced by Simis, Ulrich and Vasconcelos in [65], in order to characterize the Cohen-Macaulayness of Rees algebras of modules. Thanks to this technique, the problem is reduced to the case of Rees algebras of ideals. Our main results are the following.</div><div><br></div><div><div>In Chapters 3 and 4 we consider a finite module <i>E</i> over a Gorenstein local ring <i>R</i>. In Theorem 3.2.4 and Theorem 4.3.2, we give sufficient conditions for <i>E</i> to be of linear type, while Theorem 4.2.4 provides a sufficient condition for the Rees algebra <i>R(E)</i> of <i>E</i> to be Cohen-Macaulay. These results rely on properties of the residual intersections of a generic Bourbaki ideal <i>I</i> of<i> E</i>, and generalize previous work of Lin (see [46, 3.1 and 3.4]). In the case when <i>E</i> is an ideal, Theorem 4.2.4 had been previously proved independently by Johnson and Ulrich (see [39, 3.1]) and Goto, Nakamura and Nishida (see [20, 1.1 and 6.3]).</div></div><div><br></div><div><div>In Chapter 5, we consider a finite module <i>E</i> of projective dimension one over <i>k</i>[X₁, . . . , X_n]. Our main result, Theorem 5.2.6, describes the defining ideal of <i>R(E)</i>, under the assumption that the presentation matrix φ of <i>E</i> is <i>almost linear</i>, i.e. the entries of all but one column of φ are linear. This theorem extends to modules a known result of Boswell and Mukundan on the Rees algebra of almost linearly presented perfect ideals of height 2 (see [5, 5.3 and 5.7]).</div></div><div><br></div><div><div>The second part of this thesis studies the Cohen-Macaulay property of the special fiber ring<i> F(E)</i> of a module <i>E</i>. In Theorem 6.2.14, we prove that the generic Bourbaki ideals of Simis, Ulrich and Vasconcelos allow to reduce the problem to the case of fiber cones of ideals, similarly as for Rees algebras. We then provide sufficient conditions for <i>F(E)</i> to be Cohen-Macaulay. Our Theorems 6.2.15, 6.1.3 and 6.2.18 are module versions of results proved for the fiber cone of an ideal by Corso, Ghezzi, Polini and Ulrich (see [10, 3.1] and [10, 3.4]) and by Monta˜no (see [47, 4.8]), respectively.</div></div><div><br></div>

Thesis advisor: Karen Arnold / The purpose of this study was to determine the impact of a business ethics course on the cognitive moral reasoning of freshmen business students. The sample consisted of 268 college students enrolled in a required business ethics course. The students took Rest's Defining Issues Test - Version 2 (DIT2) as a pre-test and then post-test (upon completing the course). Descriptive analyses, t-tests, ANOVA, and multiple regression were employed to compare the pre-test and post-test scores of the students and to determine the relationship, if any, between the variables of gender, class size, instructor, class time of day, SAT scores, and students' GPAs; and the dependent variable of moral reasoning (N2) scores. Descriptive analyses showed that taking this kind of a course made a difference for virtually everyone. While women had higher pre-test scores, post-test scores, and overall gains in moral reasoning than men, men also had gains across the board. Gains were also found across all instructors, in both small and medium class sizes and regardless of what time the class occurred. Further, compared to the national sample of college students, the study sample was at the level of college seniors. Results of ANOVA testing showed that the gains in moral reasoning scores were statistically significant ones. However, the moral reasoning gains of students enrolled in the smaller sections (n = 19) were not statistically significantly different than students enrolled in medium sized (n = 27) sections of the same course. Finally, the independent variable of class time had the most statistically significant relationship with gains in moral reasoning scores. The findings of this study suggest several practices for institutions of higher education. First, certain curricular conditions should be considered by institutions with ethics courses. The content should be explicit and pedagogical strategies should include role-taking, the discussion of moral dilemmas, reflection, active learning, and cognitive disequilibrium. Second, increasing class sizes by eight students can help to maintain moral growth while still being financially efficient. Third, these findings may inform administrators when planning class times (the earlier, the better). / Thesis (PhD) — Boston College, 2011. / Submitted to: Boston College. Lynch School of Education. / Discipline: Educational Administration and Higher Education.

Frauds conducted online are crimes that are and has been increasing in numbers over the last years. Government reports shows that this is the case in Sweden as well as internationally. As of today the Swedish legal systems lacks a definition and categorization approach for online frauds. This thesis aims at developing such definition and a classification scheme that can be used to classify online frauds. Further, this thesis suggests and evaluates a defense mechanism for use in arenas were online frauds are executed. The results bring a definition and classification scheme that has been developed in cooperation with the Swedish police and validated with a judge from a Swedish court. The evaluation of the suggested defense mechanism indicates that our suggested defense mechanism has the potential to make internet users less susceptible to online fraud attacks.

This thesis aims to clarify the nature of the relationship between moral reasoning, as per the neo-Kohlbergian DIT approach of Rest and his colleagues, and liberalism-conservatism. Moral reasoning and liberalism-conservatism are consistently found to be related but the resultant interpretation that liberals are more moral-cognitively advanced than conservatives has been challenged by Emler and his colleagues who argue that the DIT is liberally biased. Subsequent research on this issue has produced a methodological quagmire that this thesis aims to proceed beyond. The specific aim of this thesis is to test several different (or competing) hypotheses purporting to explain the relationship between Kohlbergian moral reasoning and liberalism-conservatism. These are (1) that liberals are more morally advanced than conservatives; (2) that "advanced moral reasoning" is merely social presentation; (3) that moral reasoning is separately constrained by moral development and conservatism; (4) that moral development and liberalism represent distinct paths to postconventional reasoning preference; (5) that moral reasoning differences between liberals and conservatives are broader than usually thought; (6) that the political content of moral issues affects moral reasoning differences between liberals and conservatives; and (7) that moral reasoning instruments have exaggerated moral reasoning differences between liberals and conservatives. Study 1 found that a non-ipsative, indirect moral reasoning measure was correlated with liberalism-conservatism thus disconfirming hypotheses 2 and 7. Additionally, hypothesis 5 was not supported by several DIT findings. Opposing hypotheses 2 and 6, Study 2 found that a conservative version of the DIT was correlated with liberalism-conservatism although a potential methodological issue arose. Study 3 developed an objective measure of moral comprehension, broader in scope than previous moral comprehension measures, which demonstrated acceptable reliability and validity. Employing this measure, Study 4 found that moral comprehension and liberalism were weakly correlated and that they independently predicted moral reasoning, although their interaction did not. Together, these findings provide some support to hypotheses 1 and 4 but not hypotheses 3 and 6. Overall these findings reveal that liberals appear more moral-cognitively advanced than conservatives but, compared to conservatives, liberals appear to indicate preference for advanced moral reasoning earlier in their moral-cognitive development. This latter mechanism appears stronger and suggests that, although DIT scores are still somewhat reflective of moral-cognitive development, the DIT�s estimate of liberals� moral-cognitive development is elevated. Future research can continue to explore these hypotheses (e.g., via longitudinal and/or "faking" studies) and, in so doing, further clarify the relationship between Kohlbergian moral reasoning and liberalism-conservatism.

This thesis seeks to explore the points of dissonance and resonance around the understanding and deployment of the term 'religion' between Human Rights Watch, and the government of the People's Republic of China. In doing this, it is highlighted that a fundamental disjunction exists in the meaning of, and the boundaries of, the word 'religion' between these two groups. The space that this difference creates makes discussions on religion and religious freedom between these two groups extremely problematic, primarily because Human Rights Watch seeks to protect the right to religious freedom of groups and individuals that the Chinese government does not consider ‘religion’, but that Human Rights Watch demands they should. This thesis addresses the question of the role of social and cultural relativism in the defining, and the subsequent role in defending, of the term and contents of 'religion'.

This thesis examines the major thoughts on anthropology and selfhood from Plotinus in the third century and the Cappadocians in the fourth, situating the anthropology of the Cappadocians in the much broader context of their culture and their major works. It argues that: i) The inherent unity of all things, intelligible and material, provides the basis for radically intuitive categories such as synchronity, telepathy, and even love. ii) The ontological essence of expressed particularity in the divine or the human is an ekstatic relationship, i.e., it involves the transcending of the boundaries of self, a self identified as hypostasis or person. iii)Truth consists in apprehending that true being alone possesses existence in its own nature, participated in by all without being lessened and knowable only as and in relationship. Human being is participation in existence by an experience of communion. iv) The most essential activity of historical self is to use one's inherent capacity to form one's own identity in relation to the other -- both external and within -- as incarnational and dialogic beings. The findings of this thesis are that the relational notion of authentic human being grounded in open-ended divinity provides both a useful framework and the distinctive characteristics of human beingness for rethinking what it means to be a human being in the twenty-first century.