Thesis (Ed.D.)--Boston University.

Thesis (Ed.D.)--Boston University.

Thesis (MTech (Electrical Engineering))--Peninsula Technikon, Cape Town, 2002. / Tne purpose of this research or study was to perform an investigation in the use of optical fibers as sensor elements in order to develop a millimetre wave instrument capable of measuring microwave power within microwave cavities. Included in the scope of the project was an investigation of microwaves and microwave power measurements techniques. The emphasis of the research work was to develop expertise in photonics, by modeling and developing a measuring technique using optical techniques. This was deemed to be highly beneficial in laboratory experimentation and for possible use by microwave technicians. The implementation was amongst others, done by means of computer simulation and associated hardware, together with fiber-optic accessories. In order to conduct this research a literature and technology survey of current non-optical microwave power measurement technique was done. With this a review different power measurements systems and their relationship towards microwave power measurements was conducted. Within the scope of the project, a study of fiber optics sensors and its components was also conducted, which enabled models for a Mach-Zehnder microwave sensor to be developed. This resulted in the development of inexpensive electronic signal conditioning and detection techniques to enable measurements that employed a Mach-Zehnder Interferometer for this sensor technique. Finally, as microwaves are difficult to measure with pure electronic equipment, different approaches were made to investigate the temperature changes and other parameters on optical fiber to avoid damage to it. The specifications of hardware and circuitry suitable to measure these effects were determined.

De nos jours, les dirigeants consultent les résultats de programmes nationaux et internationaux d'évaluation des apprentissages afin de se renseigner au sujet de la qualité de l'enseignement dispensé dans les salles de classe. Cependant, des études ont remis en question la qualité des données issues de ces questionnaires contextuels au sujet des pratiques et croyances pédagogiques (Clare, 2000; Bali & Rowan, 2004, Hill, 2002; Mayer 1999, Mullens & Kasprzvk, 1999; Mullis, 2002; NCES, 1999; NCES, 2002; Porter & Smithson, 2001; Rogers et al., 2004; Rowan, Correnti & Miller, 2002; Stecher et al. 2002; Stigler & Hiebert, 1996). La présente recherche procède à l'évaluation de la qualité psychométrique des items issus des questionnaires contextuels de la Troisième enquête sur les mathématiques et les sciences (TEIMS) portant sur les pratiques pédagogiques recommandées par les réformes scolaires. L'étude cible la population des enseignants francophones de mathématiques oeuvrant en Ontario. Une méthodologie mixte est employée pour mesurer divers concepts de validité et de fidélité des items de la TEIMS. Les lacunes relevées au niveau de la validité proviennent principalement de la conception des items (formulation de la question, vocabulaire utilisé, période de référence de l'item, pertinence du sujet et des choix de réponse). La concordance entre élève et enseignant démontre que les items simples produisent des résultats davantage comparables que les items complexes faisant appel à des concepts vagues ou trop nombreux. La validité théorique n'a pu être établie en fonction des regroupements conceptuels inspirés par le curriculum de l'Ontario (1997). Les propriétés métriques des items s'avèrent également problématiques compte tenu du type de données générées et des lacunes distributives des items. Toutefois, les analyses n'ont pas permis de déceler la présence d'un biais de réponse systématique en fonction du phénomène de désirabilité sociale. Les analyses de cohérence interne démontrent une faible cohésion entre les items des questionnaires qui traitent des nouvelles pratiques et croyances pédagogiques. Ces résultats indiquent en grande partie que les données provenant de questionnaires contextuels des programmes à grande échelle n'arrivent pas à dresser un portrait valide ni fidèle des nouvelles pratiques pédagogiques chez les enseignants. Les résultats de cette étude doivent cependant, être interprétés avec discernement à cause de certaines faiblesses méthodologiques, dont les données non-representatives de la population ciblée et le choix limite des méthodologies.

Personalization, the idea that teaching can be tailored to each students’ needs, has been a goal for the educational enterprise for at least 2,500 years (Regian, Shute, & Shute, 2013, p.2). Recently personalization has picked up speed with the advent of mobile computing, the Internet and increases in computer processing power. These changes have begun to generate more and more information about individual students that could theoretically be used to power personalized education. The following dissertation discusses a novel algorithm for processing this data to generate estimates of individual level attributes, the Inverse Bayes Filter (IBFi). A brief introduction to the use of Bayes Theorem is followed by a theoretical chapter and then two empirical chapters that describe alternately how the model is constructed, and how it performs on real student data. The theoretical chapter presents both the theory behind Inverse Bayes, including subjective probability, and then relates this theory to student performance. The first empirical chapter describes the prediction accuracy of IBFi on two proxies for students’ subjective probability, partial credit and cumulative average. This prediction performance is compared to the prediction accuracy of a modified Bayesian Knowledge Tracing model (KTPC) with IBFi performing reasonably, out-predicting the KTPC model on a per-student basis but not across all predictions. In the second empirical chapter I validate the theory behind the Inverse Bayes Filter through testing whether student certainty (or confidence) improves prediction performance. The inclusion of student certainty is shown to improve the predictive performance of the model relative to models that do not use certainty. This evidence supports the IBFi model and its underlying theory, indicating that students’ judgments about their levels of certainty in their answers contains information that can be successfully identified by the model. A final summary chapter describes the consequences of using this model for education broadly. / Human Development and Education

Amid a growing debate over the use of standardized test scores, states and districts across the country have begun using alternative measures of school quality, including surveys of students and teachers. As a result, many schools now have access to troves of diagnostic data on student and teacher perception. However, with few established practices for analyzing or planning with survey data, there has been wide variation in how schools actually use their results. I completed my residency at UChicago Impact, a non-profit connected to the University of Chicago’s Urban Education Institute. UChicago Impact administers the 5Essentials Survey, based on the research in Organizing Schools for Improvement (Bryk et al, 2010), in nearly 5,000 schools nationwide. My strategic project was to design and pilot a series of workshops to help teams of teachers and administrators use their 5Essentials data to improve the organization of their schools and, ultimately, student outcomes. My research identified two core problems that often prevent practitioners from using survey data effectively: (1) the complex social problems rooted in survey data require brave conversations, unintuitive planning, and collective action, making the data hard to influence; (2) the infrequent administration and release of survey data make it difficult for schools to collect new data and adjust their actions accordingly. I found that as a result of attending the workshops, most teams were able to have productive conversations about their schools and coalesce around a plan for improvement. However, when it came time to implement their plans, the teams faced obstacles around accountability, coherence, and assessment of impact. Moreover, it is unclear that any of the actions that teams did implement will lead to improvements in student outcomes. Given the relatively limited nature of the this intervention – eight total hours of workshops – these findings suggest that the analysis of student and teacher survey data may be an effective way to help schools begin to build trust between stakeholders. However, in order for survey data to drive sustained, measurable improvement, the reporting infrastructure needs to become more nimble, and leaders have to balance support and accountability, while integrating survey data with other data sources and initiatives.

10 million students walk into rural schools every school day, representing about 20% of the United States’ public school population. More than a third of all public schools and almost three-fifths of local education agencies serve rural students. Creating coherent, scalable strategies to impact rural students can be difficult given the diffuse and often isolated context of the nation’s 7,000+ rural districts. I completed my residency at the College Board, a New York-based nonprofit organization best known for its SAT, PSAT/NMSQT, and Advanced Placement assessments. In 2015, these exams reached 1.7 million, 3.8 million, and 2.5 million students, respectively. Participation in College Board assessments continues to grow over time, but rural students participate at rates lower than their peers. Working in Washington, DC, I completed a strategic project in the College Board’s Global Policy, Advocacy, and Communications division. The goals of the project were to determine the potential value of a strategy focused on rural schools and to develop a proposal outlining how to enact such a strategy. Through my strategic project, I created a College Board rural database, conducted interviews and case studies, and developed strategy proposals for potential pilot work in rural schools and districts. In this capstone, I describe the actions and results of the strategic project in three phases and analyze the results through an analytic framework called the “strategic triangle.” This tool allows a decision maker to understand the three issues in nonprofit strategy: the potential public value, the sources of legitimacy and support, and the operational capabilities of the implicated organization. The work of my residency resulted in the authorization of time, resources, and human capital needed to launch a College Board Rural Strategy pilot in the states of Colorado and Idaho. This pilot intends to create new partnerships and opportunity pipelines in rural schools while increasing the value of College Board-provided assessments and supports for rural students.

It is well established that children’s early life environments can have significant consequences on their long-term outcomes. Yet, there is still limited empirical evidence on the effects that being exposed during the prenatal and early postnatal periods to positive shocks, such as conditional cash transfers, has on long-term cognitive function. By exploiting the initial village-level randomization, I estimate the causal effect on long-term cognitive achievement of prenatal and early postnatal exposure to Oportunidades, Mexico’s conditional cash transfer program. I find that those eligible children born into the program who received its benefits early on have higher standardized assessment scores in mathematics and Spanish in third through sixth grade—up to 15 years after the program began. In line with previous research on CCTs, the effects are largely driven by children living in the poorest villages as measured by a marginalization index. Notably, the effect on these children is large enough to put their cognitive achievement on a par with children from more moderately poor villages.

Abstract not available.

Abstract not available.