Although understanding causation is an essential part of nearly every problem domain, it has resisted formal treatment in the languages of logic, probability, and even statistics. Autonomous artificially intelligent agents need to be able to reason about cause and effect. One approach is to provide the agent with formal, computational notions of causality that enable the agent to deduce cause and effect relationships from observations. During the 1990s, formal notions of causality were pursued within the AI community by many researchers, notably by Judea Pearl. Pearl developed the formal language of structural models for reasoning about causation. Among the problems he addressed in this formalism was a problem common to both AI and law, the attribution of causal responsibility or actual causation. Pearl and then Halpern and Pearl developed formal definitions of actual causation in the language of structural models. <p>Within the law, the traditional test for attributing causal responsibility is the counterfactual "but-for" test, which asks whether, but for the defendant's wrongful act, the injury complained of would have occurred. This definition conforms to common intuitions regarding causation in most cases, but gives non-intuitive results in more complex situations where two or more potential causes are present. To handle such situations, Richard Wright defined the NESS Test. Pearl claims that the structural language is an appropriate language to capture the intuitions that motivate the NESS test. While Pearl's structural language is adequate to formalize the NESS test, a recent result of Hopkins and Pearl shows that the Halpern and Pearl definition fails to do so, and this thesis develops an alternative structural definition to formalize the NESS test.

Although understanding causation is an essential part of nearly every problem domain, it has resisted formal treatment in the languages of logic, probability, and even statistics. Autonomous artificially intelligent agents need to be able to reason about cause and effect. One approach is to provide the agent with formal, computational notions of causality that enable the agent to deduce cause and effect relationships from observations. During the 1990s, formal notions of causality were pursued within the AI community by many researchers, notably by Judea Pearl. Pearl developed the formal language of structural models for reasoning about causation. Among the problems he addressed in this formalism was a problem common to both AI and law, the attribution of causal responsibility or actual causation. Pearl and then Halpern and Pearl developed formal definitions of actual causation in the language of structural models. <p>Within the law, the traditional test for attributing causal responsibility is the counterfactual "but-for" test, which asks whether, but for the defendant's wrongful act, the injury complained of would have occurred. This definition conforms to common intuitions regarding causation in most cases, but gives non-intuitive results in more complex situations where two or more potential causes are present. To handle such situations, Richard Wright defined the NESS Test. Pearl claims that the structural language is an appropriate language to capture the intuitions that motivate the NESS test. While Pearl's structural language is adequate to formalize the NESS test, a recent result of Hopkins and Pearl shows that the Halpern and Pearl definition fails to do so, and this thesis develops an alternative structural definition to formalize the NESS test.

Population aging has been an increasing in many societies during the last century, andespecially in China this issue has become one of the most urgent social phenomenonin the recent twenty years. Meanwhile, being healthy matters to the senior populationthe most. The main purpose of this paper is to investigate how to measure Chineseelderly health condition, and what the main factors are influencing their health. Thedata of this paper is from the China Health and Retirement Longitudinal Study(CHARLS). A structural equation model(SEM) was established to verify therelationship between different influencing factors and the elderly health. The latentvariables in this model were pre-studied by both exploratory factor analysis andconfirmatory factor analysis. The conclusion based on this data is elderly health canbe measured in four aspects physical condition, emotional condition, body functionand pain. The significant influencing effects of each aspects of health are time sharing,exercise, family environment and lifestyle.

The process of research is fraught with rote terminology that, when used blindly, can bend our methodological actions away from our theoretical intentions. This investigation is aimed at developing two methods for bringing meaning and interpretability to research when we work with confounds. I argue, with the first method, that granting confounds substantive influence in a network of related variables (rather than viewing confounds as nuisance variables) enhances the conceptual dimension with which phenomena can be explained. I evaluated models differing in how confounds were specified using data from the Survey of Health, Ageing and Retirement in Europe (SHARE). Generally, minor alterations to model specifications, such as direction of causal pathways, did not change model parameter estimates; however, the conceptual meaning of how the confounds interacted with other variables in the model changed drastically. Another frequent misconceptualization of confounds, detailed by the second method, occurs when confounds are used as proxy variables to control for variance that is not directly measureable, and no explicit attempt is made to ensure that the proxy variable adequately represents the underlying, intended construct. For this second demonstration, I used SHARE data to estimate models varying in the degree to which proxy variables represent intended variables. Results showed that parameter estimates can differ substantially across different levels of proxy variable representation. When imperfect proxy variables are used, an insufficient amount of variance is removed from the observed spurious relationship between design variables. The findings from this methodological demonstration underscore the importance of precisely imbuing confounds with conceptual meaning and selecting proxy variables that accurately represent the underlying construct for which control is intended.

No description available.

In the present study, parameter estimates, standard errors and chi-square statistics were compared using normal and elliptical estimation methods given three research conditions: population data contamination (10%, 20%, and 30%), sample size (100, 400, and 1000), and kurtosis (kappa =1,10, 20).

Structural equation models combine factor analysis models and multivariate regression models to estimate associations between observed variables and unobserved variables. The main achievement of this Capstone Project is the understanding of structural equation models and application of the models to real-world data. In this report, we reviewed structural equation models and several prerequisite topics. We performed a simulation study to compare maximum likelihood structural equation model estimation versus two-stage sequential estimation using multiple linear regression and maximum likelihood factor analysis. The simulation study confirmed that confidence intervals produced by structural equation models are valid and those obtained by two-stage sequential estimation are largely inaccurate. We applied structural equation models to an educational data comparing the efficacy of teaching conditions on learning scientific inquiry skills among 177 middle school students in Fitchburg, Massachusetts using a computer simulated science microworld. Application of structural equation models to the educational data showed that there were no significant differences in test score gains between three learning conditions, while controlling for latent factors measured by survey responses.

Understanding the causal architecture underlying complex systems biology has a great value in agriculture production for the development of optimal management strategies and selective breeding. So far, most studies in this area use only prior knowledge to propose causal networks and/or do not consider the possible genetic confounding factors on the structure search, which may hide important relationships among phenotypes and also bias the resulting inferred causal network. In this dissertation, we explore many structural learning algorithms and present a new one, called PolyMaGNet (Polygenic traits with Major Genes Network analysis), to search for recursive causal structures involving complex phenotypic traits with polygenic inheritance and also allowing the possibility of major genes affecting the traits. Briefly, a multiple-trait animal mixed model is fitted using a Bayesian approach considering major genes as covariates. Next, posterior samples of the residual covariance matrix are used as input for the Inductive Causation algorithm to search for putative causal structures, which are compared to each other using the Akaike information criterion. The performance of PolyMaGNet was evaluated and compared with another widely used approach in a simulated study considering a QTL mapping population. Results showed that, in the presence of major genes, our method recovered the true skeleton structure as well as the causal directions with a higher rate of true positives. The PolyMaGNet approach was also applied to a real dataset of an F2 Duroc × Pietrain pig resource population to recover the causal structure underlying on carcass, meat quality and chemical composition traits. Results corroborated with the literature regarding the cause-effect relationships between these traits and also provided new insights about phenotypic causal networks and its genetic architectures in complex systems biology. / Compreender a arquitetura causal subjacente à sistemas biológicos complexos é de grande valia na produção agrícola para o desenvolvimento de estratégias de manejo e seleção genética. Até o momento, a maior parte dos estudos neste contexto utiliza apenas conhecimento prévio para propor redes causais e/ou não considera fatores de confundimento genético na busca de estruturas, fato que pode ocultar relações importantes entre os fenótipos e viesar inferências sobre a rede causal. Nesta tese, exploramos alguns algoritmos de aprendizagem de estruturas e apresentamos um novo, chamado PolyMaGNet (do inglês, Polygenic traits with Major Genes Network analysis), para buscar estruturas causais recursivas entre características fenotípicas poligênicas complexas e permitindo, também, a possibilidade de efeitos de genes maiores que as afetam. Resumidamente, um modelo misto de múltiplas características é ajustado usando abordagem Bayesiana considerando os genes maiores como covariáveis no modelo. Em seguida, amostras posteriores da matriz de covariância residual são usadas como entrada para o algoritmo de causação indutiva para pesquisar estruturas causais putativas, as quais são comparadas usando o critério de informação de Akaike. O desempenho do PolyMaGNet foi avaliado e comparado com outra abordagem bastante utilizada por meio de um estudo simulado considerando uma população de mapeamento de QTL. Os resultados mostraram que, na presença de genes maiores, o método PolyMaGNet recuperou a verdadeira estrutura do esqueleto, bem como as direções causais, com uma taxa de efetividade maior. O método é ilustrado também utilizando-se um conjunto de dados reais de uma população de suínos F2 Duroc × Pietrain para recuperar a estrutura causal subjacente à características fenotípicas relacionadas a qualidade da carcaça, carne e composição química. Os resultados corroboraram com a literatura sobre as relações de causa-efeito entre os fenótipos e também forneceram novos conhecimentos sobre a rede fenotípica e sua arquitetura genética.

This is a multi-article format dissertation that explores emotional and cognitive engagement in higher education classrooms. Student engagement in higher education classrooms has been associated with desired outcomes such as academic achievement, retention, and graduation. Student engagement is a multi-faceted concept, consisting of behavioral, emotional, and cognitive components. A deeper understanding of how these components interact would allow instructors and course designers to facilitate more engaging learning experiences for students. The first article is an extended literature review that investigates the extant empirical research on the relationship between emotional and cognitive engagement, and between emotional engagement and academic outcomes in post-secondary classrooms. I find that this topic has been scantily researched in the past 16 years and conclude that the relationship between emotional and cognitive engagement is cyclical, rather than linear, and is influenced by student control appraisals, value appraisals, achievement goals, and the classroom environment. The second article investigates the longitudinal relationship between emotional and cognitive engagement in university blended learning courses across 2 institutions, with 68 students. Using intensive longitudinal data collection and structural equation modeling, I find that course design and student perception variables have a greater influence on engagement than individual student characteristics and that student multitasking has a strong negative influence on engagement. Students' perceptions of the importance of the activity has a strong positive influence on both cognitive and emotional engagement. An important outcome of engagement is the students' perceptions that they were learning and improving. While emotional and cognitive engagement are highly correlated, the results do not indicate that emotional engagement leads to higher levels of cognitive engagement.

Background: High nursing turnover and early nursing career exit rates evidenced by the current global nursing shortage is the impetus for effective strategies aimed at retaining nurses in their current positions. Nurses’ behavioral intentions to leave or stay are not well understood. Aim: This thesis aims to increase understanding of why clinical nurses choose to remain in their current positions and to assess the influence that nursing leaders have on staff nurses’ intent to stay. Methods: Two systematic literature reviews were conducted; one to synthesize current research on clinical nurses’ intentions to stay and the influence of leadership practices on those intentions; the other to determine the appropriateness of conceptualizing intentions to stay and leave as opposite ends of a continuum. Building on two published conceptual models (Boyle et al. 1999; Tourangeau & Cranley (2006), a new theoretical model of nurses’ intent to stay was developed and tested as a structural equation model using LISREL 8.8 and a subset of the QWEST study data provided by 415 nurses working in nine hospitals in one Canadian province. Results: The systematic reviews identified positive relationships between relational leadership practices and nurses’ intentions to stay, supporting the assertion that managers influence the behavioral intentions of nurses and their intentions to stay and leave. Intentions to stay and leave were found to be separate but correlated concepts. Model testing results, χ2=169.9, df=148 and p=0.105, indicated a fitting model that explained 63% of the variance in intentions to stay. Concepts with the strongest direct effects on intent to stay were empowerment, organizational commitment, and desire to stay. Leadership had strong total effects and indirectly influenced intent to stay through empowerment. Conclusions: Findings suggested that intent to stay or leave should be investigated as separate but correlated concepts. Relational leadership that focuses on individual nurses and supports empowering work environments will likely affect nurses choosing to remain in their current positions.