Search

As online learning platforms and computerized testing become more common, an increasing amount of data are collected about users. These data include, but are not limited to, response time, keystroke logs, and raw text. The desire to observe these features of the response process reflect an underlying interest in the cognitive processes and behaviors respondents engage in while taking tests and navigating learning platforms. Alongside this interest in response processes is an increased desire to use computational methods, artificial intelligence, and machine learning to combine this process data with traditional item response data. This dissertation attempts to advance a notion of "computational validity," defined as arguments for the use and interpretation of test scores from evidence collected or analyzed using computational methods. Chapter one looks at item position effects as a threat to comparability across test forms. The presence of position effects is especially threatening to computer adaptive tests. Here a mixture item response theory model that decomposes item and person components of position effects is developed and applied to a large Brazilian college admissions exam to demonstrate its ability to both recover information about position effects in items and automatically produce comparable ability distributions across forms. Chapter two looks at data from a browser-based reading assessment and proposes drift diffusion models as a theoretically grounded way to combine response time and response accuracy data. The individual parameters estimated from the drift diffusion models are combined with scores on the browser-based assessment and used to predict scores on more expensive and time consuming measures of reading ability. This provides a line of evidence justifying the clinical and classroom use of the tool for diagnostic purposes. Chapter three looks at the issue of model selection. To make decisions from a model requires justification that the model is appropriate for the data. Current methods of model selection are well suited to make decisions on which model fits better, but not to communicate about how much better or if the degree of fit is appropriate. This paper uses a tool called the InterModel Vigorish that aims to provide measures of increases in predictive performance on an interpretable scale and extends it to polytomous item response theory models. Simulations are performed to explore what can be learned about data generating processes from this tool and then it is applied to a selection of empirical datasets. Combined, this work aims to highlight ways that modern computational methodologies can be used to improve not only assessments, but the decisions we make from them and our ability to communicate about them. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]

This dissertation explores the significance of the bargaining power of the players involved in the wage decisions for teachers in Brazil, namely the teacher unions and the municipal governments. The dissertation is divided in two parts, which together form a broad analysis of the labor market for public school teachers in Brazil and the underlying institutional factors that have determined its functioning in the past twenty years. This research indicates that the bargaining power of municipal governments and teacher unions plays an essential role in setting wages. Moreover, the results suggest that the bargaining power is more important to the wage levels than the available budget of the municipalities, and that market power is driven by the lack of alternative job opportunities to the teachers. The first paper analyzes the degree of monopsony power held by municipal public employers in teacher hiring. It shows evidence of very low elasticities for the teachers' labor supply curve (of 0.11 - 0.36, depending on the model specifications), suggesting that the market power held by municipal governments allows them to enjoy rents that are, in absence of other players to counterbalance their action, as large as 275% - 901% of the teachers wage value. It also indicates that the proportion of female teachers increase the power that municipalities have to set wages, given the relatively lower elasticity of women's supply curve. The second paper examines the actual behavior of governments in teacher hiring and uses two models -- the Monopoly-Monopsony and the Insider-Outsider -- as possible interpretations of the results. I use the FUNDEF funding reform in Brazil that has redistributed funds across municipal and state systems within each Brazilian state to identify the wage-setting decisions. It shows evidence that municipalities that benefitted from the redistribution were more likely to increase teacher salaries (by an average of 3.9%), while municipalities relatively harmed by the redistribution tended to increase the number of teachers (by 8.8%) and to decrease salaries (by -4.4%). Surprisingly, the analyses of the value transferred to the municipalities reveal that receiving more funds for education does not necessarily translate to higher teacher salaries, suggesting that the bargaining power of the municipal governments and teacher unions matter more to wage determination than the amount of funds the school systems have. This dissertation provides evidence, in the context of Brazil, that policies focused on raising across-the-board wages can be desirable as a counterforce to the market power held by municipal governments. Moreover, it suggests that federal policies focused on changing the bargaining power of the municipalities would be more effective in influencing wage levels than the amount of grants given to local governments. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]

German-born composer Ernst Mahle (b. 1929) holds a respected position in the music field in Brazil, his adopted country since 1951. Mahle is the artistic director and co-founder of the "Escola de Musica de Piracicaba," one of the most important musical institutions in Brazil. His work has focused on conducting, composition, and elementary music education. The main focus of this research will be to examine Mahle's four concertinos for piano and orchestra intended for young performers. This paper will also provide information regarding Mahle's sixty years of musical legacy in Brazil and his pedagogical idiom for piano solo. For a better understanding of Mahle in this context, the paper is divided in four chapters. Chapter I consists of a biography of the composer, the contextualization of him on the musical scenario from Brazil and his close association with the Escola de Musica de Piracicaba. Chapter II discusses the factors that have influenced Mahle's eclecticism and his compositional development. Chapter III includes a pedagogical overview of Mahle's works for piano offering some characteristics of each piece regarding historical backgrounds, musical forms, harmony, and materials used, as well as levels of difficulty, timing, available editions, and a list of pedagogical features. Finally, Chapter IV provides a study of the concertinos for piano and orchestra and the discussion of my pedagogical edition of the concertinos. Chapter IV also includes a historical background and analysis of the concertinos form, harmony, phrase structure, and materials. In addition, there is an analysis of the works' pedagogical features--their challenges and the skills that each piece helps the young pianist develop, and suggestions for how to address these issues. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]

Across three papers, I examine the influence of school-related factors on student outcomes in Brazil. In the first and second papers, I investigate the effects of teachers' contractual ties with the school on different outcomes. Two aspects of contractual ties were considered: 1) whether teachers hold a permanent or temporary contract, and 2) whether they work at a single or multiple schools during the week. In Brazil, a large proportion of teachers work on temporary contracts and at multiple schools. However, little is known about the educational implications of these contractual arrangements. My dissertation work aims to address this gap. The first paper makes causal inferences about the impact of permanent and single-school teachers on teacher-student relationships and student performance. Analyses use unique administrative and test data from Sao Paulo, Brazil. To account for nonrandom sorting of students and teachers across schools and classrooms, I use a cross-subject analysis with student-fixed effects. Results show that single-school and permanent teachers have a positive and significant impact on teachers' provision of support and academic feedback. Permanent teachers also have a positive effect on student achievement. However, findings indicate that the effects of teachers' contractual ties are smaller for students at higher risk of school failure. The second paper builds on these findings and explores whether teachers' contractual ties affect school dropout. This study uses extensive data from Brazil's Census of Basic Education over four academic years (2011-2014) and employs two classes of fixed effects models based on school-grade-year variation. Results are mixed. School dropout rates are 0.5 to 1 percent lower when students are taught exclusively by single-school teachers; however, dropout rates are 0.4 to 0.8 percent higher when students are taught by permanent teachers. The third paper extends the research on socioeconomic achievement gaps by looking at the extent to which inequality in test scores explains attainment gaps in Brazil. This study draws on Boudon's (1974) theoretical model of inequality in educational opportunity to estimate the primary effects of social background on high school graduation: that is, the proportion of the overall inequality in high school completion associated with the socioeconomic test score gap between students from low and high social backgrounds. Analyses are based on administrative and test data from Sao Paulo and use a decomposition technique for non-linear probably models. Results indicate that the test score gap in middle school is associated with half of the overall socioeconomic inequality in high school graduation. These primary effects are similar to those estimated for the U.S. This is a contribution to previous studies of primary effects, which are based primarily on data from European countries and the United States. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]

This dissertation encompasses three chapters that study the Brazilian higher educational system and determines the barriers to acquiring higher education in the country. Below are the individual abstracts for each chapter. Chapter 1: "Rural and urban schools' performance in Brazil and its impact on access to higher education". Recent research on academic achievement in Brazil shows that public school students face strong barriers in gaining access to higher education compared to private school students. However, little is known about differences between public schools located in urban and rural areas, another source of opportunity inequality. I estimate the effectiveness of rural and urban schools in Brazil using data from a major university entrance exam. To account for bias in Ordinary Least Square (OLS), I use a technique recently developed by Altonji, Elder, and Taber (2005b) that estimates the ratio of the influence of omitted variables relative to observed variables that would be required to completely explain the estimates one obtains via OLS. The results indicate that students from urban areas outperform their rural counterparts and provide suggestive evidence that the poor quality of rural public schools is the source of such score differences. Chapter 2: "Climbing the Educational Ladder: The Relative Performance of Rural and Urban Students in Brazilian Universities". Recent research on academic achievement in Brazil shows that rural school students face strong barriers in gaining access to higher education compared to urban school students. However, little is known about differences between rural and urban students when they are granted access to the university. We estimate the relative performance of rural and urban students in Brazil using data from a major university. Furthermore, to observe how performance varies along the different GPA quantiles, we use Koenker and Bassett (1978) quantile regression approach. The results indicate that students from rural areas outperform their urban counterparts providing suggestive evidence that using policed implementations that increase rural acceptance rates would decrease regional educational inequalities and help fight the high income inequality rates observed in the Brazilian society. Chapter 3: "College Dropouts and Entrance Test Scores: A Censored Quantile Regression Approach". The problem of university dropouts has generated increased interest among researchers, policymakers, and educators in recent years. This paper examines the many issues involved in trying to understand and solve this complex social and educational problem. Using a dataset from students enrolled in a Brazilian major university, we estimate the effect of student entrance exam performance on dropout rates using a censored quantile regression approach proposed by Portnoy (2003). We find that students with higher entrance exam scores are more likely to dropout of college in the beginning of their college journey while being less likely to dropout after some time. Such results are likely due to the perception that a high ability student has over his probabilities of entering a more difficult or better major the following year. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]

This paper focuses on the transition to military rule in Brazil (1964), Uruguay (1973), Chile (1973), and Argentina (1976) and examines the argument that the disciplines which lend themselves to analysis of the social order, such as the social sciences and humanities, are most often the target of political repression. Argentina, Uruguay, and Chile all had stable democracies for most of the 20th century, and each suffered an abrupt transition to brutal military rule. Brazil had a shorter tradition of democratic government when it fell to military rule in the 1960s. In each case the transition to military rule was followed by repression against academic freedom, including purging of university faculty and students, closing entire departments and universities, and at times the murder of students and faculty by the military. The social sciences suffer more than other disciplines under repressive regimes because they sensitize students to problems inherent in the existing social order. In order to determine if student enrollment in social sciences decreased under authoritarian regimes, data were analyzed for each of the countries. The results tended to support the contention that the social sciences suffer under dictatorships. (SM)