Thesis advisor: Laura M. O'Dwyer / Education researchers are frequently interested in examining the causal impact of academic services and interventions; however, it is often not feasible to randomly assign study elements to treatment conditions in the field of education (Adelson, 2013). When assignment to treatment conditions is non-random, the omission of any variables relevant to treatment selection creates a correlation between the treatment variable and the error in regression models. This is termed endogeneity (Ebbes, 2004). In the presence of endogeneity, treatment effect estimates from traditionally used regression approaches may be biased. The purpose of this study was to investigate the causal impact of an integrated student support model, namely City Connects, on student academic achievement. Given that students are not randomly assigned to the City Connects intervention, endogeneity bias may be present. To address this issue, two novel and underused statistical approaches were used with school admissions lottery data, namely Gaussian copula regression developed by Park and Gupta (2012), and Latent Instrumental Variable (LIV) regression developed by Peter Ebbes (2004). The use of real-world school admissions lottery data allowed the first-ever comparison of the two proposed methods with Instrumental Variable (IV) regression under a large-scale randomized control (RCT) trial. Additionally, the researcher used simulation data to investigate both the performance and boundaries of the two proposed methods compared with that of OLS and IV regression. Simulation study findings suggest that both Gaussian copula and LIV regression are useful approaches for addressing endogeneity bias across a range of research conditions. Furthermore, simulation findings suggest that the two proposed methods have important differences in their set of identifying assumptions, and that some assumptions are more crucial than others. Results from the application of the Gaussian copula and LIV regression in the City Connects school lottery admissions study demonstrated that receiving the City Connects model of integrated student support during elementary school has a positive impact on mathematics achievement. Such findings underscore the importance of addressing out-of-school barriers to learning. / Thesis (PhD) — Boston College, 2019. / Submitted to: Boston College. Lynch School of Education. / Discipline: Educational Research, Measurement and Evaluation.

Submitted by William Justo Figueiro (williamjf) on 2015-07-17T21:31:00Z No. of bitstreams: 1 27c.pdf: 347222 bytes, checksum: 042d80068b716bfa26d3f8b003fb4ada (MD5) / Made available in DSpace on 2015-07-17T21:31:00Z (GMT). No. of bitstreams: 1 27c.pdf: 347222 bytes, checksum: 042d80068b716bfa26d3f8b003fb4ada (MD5) Previous issue date: 2011-03-24 / Nenhuma / A partir de 1990, diversos países, entre eles o Brasil, passaram por processos de liberalização comercial, esperando elevar suas taxas de crescimento econômico. Essa tendência de maior integração comercial influenciou diversos autores a buscar inferir os efeitos do comércio internacional sobre o crescimento econômico. No entanto, os artigos que encontraram uma relação negativa entre as barreiras ao comércio e o crescimento econômico sofrem ou do uso de indicadores de abertura comercial inapropriados ou de métodos econométricos questionáveis, especialmente no que se refere à endogeneidade do comércio. Frankel e Romer (1999) superaram este problema construindo uma variável instrumental, usando as características geográficas dos países que não são correlacionadas com a renda, especialmente a distância entre os parceiros comerciais e o seu tamanho. Esse modelo visava mensurar como o volume de comércio, e não mais a redução das barreiras ao comércio, impactou na renda de diversos países em 1985, já que uma redução de barreiras ao comércio influenciaria positivamente o comércio internacional. Seguindo esse modelo, esta dissertação busca estimar o impacto do aumento do comércio sobre a renda no Brasil, comparando o período do final dos anos 1980 e início dos anos 1990 (1989-1991) com um mais recente (2005-07), através dos atributos geográficos dos estados brasileiros, baseado em um modelo gravitacional. O principal resultado aponta para um forte impacto do aumento do volume de comércio sobre a renda per capita no Brasil no período mais recente, com o aumento de um ponto percentual no grau de abertura do país levando a uma elevação da renda per capita entre 6% e 7%. / Since the 1990s many countries, including Brazil, adopted trade liberalization measures expecting to increase their economic growth. This trend influenced many authors to search for signs of the effects of liberal trade policies on economic growth. However, the papers in the literature that claimed to find a negative association between barriers to trade and economic growth relied either on constructing inappropriate indicators of openness or on a questionable use of econometric methodologies, especially the failure to account for the endogeneity of trade. Frankel e Romer (1999) overcame this problem by using an instrumental variable, based on a country’s geographic attributes not related to income, notably its distance from trading partners and size. They sought to measure the impact of trade volume not trade barriers on growth in many countries in 1985, since the reduction of trade barriers would affect positively international trade. Following this methodology, this dissertation estimates the effect of the increase in trade flows on income of Brazilian states, comparing the period in the late eighties and early nineties (1989-1991) with one more recent (2005-07), using geographic characteristics of Brazilian states, based on a gravity model. The main result shows a significant impact of trade on per capita income in Brazil in the more recent period, with a one percentage increase in trade shares increasing per capita income by 6% or 7%.

股市發展是否影響長期經濟成長？本研究以Levine(1991)的內生成長模型作為實證分析之理論依據，股票市場的出現可以消弭流動性風險與生產性風險對個人投資決策的負面影響，進而促進經濟體的成長。我們採取一般成長文獻所最常使用之跨國橫斷面迴歸方式，分析52國1985-1997的平均股市與社會經濟資料，以法系與法律環境做為工具變數，來捕捉股市發展之外生部分對經濟成長的影響。實證結果如下：1.整體而言股市發展的確顯著促進了一國的經濟成長，而影響成長的主要管道則為資本成長。2.對於先進工業化國家而言，我們無法證明其股票市場對於經濟成長具有影響力。3.相反的，對開發中國家而言，股票市場則扮演著重要的角色，不僅促進經濟的成長，對於資本成長與要素生產力成長也有顯著的提升。4.股市與私人儲蓄之間的關係則較為模糊，雖然實證結果隱約透露，對開發中國家而言，發展股市有助於提升私人儲蓄，但我們無法找到統計上顯著強固的證據來加以支持。5.法律因子對於各國股市發展差異所具有的顯著解釋能力給予我們一個啟示：倘若一國的法律對於股東權益的保障越好，執行法律的品質與效率越高，那麼一國將會擁有一個功能完善與發展良好的股票市場，進而對其經濟成長有所助益。 / Do well-developed stock markets promote long-run economic growth? This study evaluates the empirical relationship between the level of stock market development and (i) economic growth, (ii) physical capital accumulation, (iii) total factor productivity, and (iv) private saving rate. By using a pure cross-country instrumental variable estimator to extract the exogenous component of stock market development, we find that (1) stock market development (measured by liquidity) is positively associated with economic growth and physical capital growth; (2) the links between stock market development and both productivity growth and private saving rates are tenuous; (3) for the developing sub-sample, the growth-promoting effect is as strong as with the full sample; for the developed sub-sample, however, this effect is at best very weak; and (4) legal reforms that strengthen shareholder rights and contract enforcement can boost stock market development and therefore accelerate economic growth.

We first studied factors associated with the rate of colonoscopy by primary care physicians (PCPs) in Ontario between the years 1996 and 2005. Next, we conducted an Instrumental Variable Analysis (IVA) to estimate the effect of colonoscopy on colorectal cancer (CRC) incidence and mortality on average-risk subjects aged 50-74 years. Finally, we explored two study cohorts, one by including subjects who had the outcomes during the exposure period (unselected cohort) and the other cohort by excluding those subjects (restricted cohort). We estimated the absolute risk reduction associated with colonoscopy in preventing CRC incidence and mortality using traditional regression analysis, propensity score analysis and IVA. PCPs who were Canadian medical graduates and with more years of experience were more likely to use colonoscopy. PCPs were more likely to use colonoscopy if their patient populations were predominantly women, older, had more illnesses, and if their patients resided in less marginalized neighborhoods (lower unemployment, fewer immigrants, higher income, higher education, and higher English/French fluency). Using PCP rate of discretionary colonoscopy as an instrumental variable, receipt of colonoscopy was associated with a 0.60% absolute reduction in 7-year CRC incidence and a 0.17% absolute reduction in 5-year risk of death due to CRC. The unselected cohort showed an increase in CRC incidence and mortality associated with colonoscopy, whereas the restricted cohort showed a reduction in CRC incidence and mortality associated with colonoscopy. In the restricted cohort, using different statistical models, the absolute risk reduction varied from 0.52-0.60% for CRC incidence and 0.08-0.17% for CRC mortality. There were social disparities in the use of colonoscopy by PCPs and this disparity increased as the overall use of colonoscopy increased over time. Colonoscopy is effective in reducing incidence and mortality due to CRC. Different methods of subject selection and statistical analysis provided different estimates of colonoscopy effectiveness.

We first studied factors associated with the rate of colonoscopy by primary care physicians (PCPs) in Ontario between the years 1996 and 2005. Next, we conducted an Instrumental Variable Analysis (IVA) to estimate the effect of colonoscopy on colorectal cancer (CRC) incidence and mortality on average-risk subjects aged 50-74 years. Finally, we explored two study cohorts, one by including subjects who had the outcomes during the exposure period (unselected cohort) and the other cohort by excluding those subjects (restricted cohort). We estimated the absolute risk reduction associated with colonoscopy in preventing CRC incidence and mortality using traditional regression analysis, propensity score analysis and IVA. PCPs who were Canadian medical graduates and with more years of experience were more likely to use colonoscopy. PCPs were more likely to use colonoscopy if their patient populations were predominantly women, older, had more illnesses, and if their patients resided in less marginalized neighborhoods (lower unemployment, fewer immigrants, higher income, higher education, and higher English/French fluency). Using PCP rate of discretionary colonoscopy as an instrumental variable, receipt of colonoscopy was associated with a 0.60% absolute reduction in 7-year CRC incidence and a 0.17% absolute reduction in 5-year risk of death due to CRC. The unselected cohort showed an increase in CRC incidence and mortality associated with colonoscopy, whereas the restricted cohort showed a reduction in CRC incidence and mortality associated with colonoscopy. In the restricted cohort, using different statistical models, the absolute risk reduction varied from 0.52-0.60% for CRC incidence and 0.08-0.17% for CRC mortality. There were social disparities in the use of colonoscopy by PCPs and this disparity increased as the overall use of colonoscopy increased over time. Colonoscopy is effective in reducing incidence and mortality due to CRC. Different methods of subject selection and statistical analysis provided different estimates of colonoscopy effectiveness.

This thesis has three self-contained articles. Economic growth and trade in human capital: A salient empirical pattern in the East Asian “miracle” is a large increase in output and factor accumulation despite  only a modest increase in TFP. I develop and calibrate a model of growth and catch-up to provide a possible explanation. A novel element of the model is a globalized education market allowing human capital transfer from frontier to developing economies – an assumption motivated by the experience of countries like Korea and Taiwan where domestic universities employed graduates of Western universities to provide advanced training. The political economy of urban bias in dictatorial regimes. In many developing countries, public resource allocation is often biased against the rural population – a policy that hurts the vast majority of the poor living in rural areas. This paper develops a dynamic political economy model of urban bias in a dictatorial regime. A novel result of the model is that urban bias can emerge in predominantly agrarian economies even if there is no bias  in political power toward urban residents. The empirical evidence from a recently compiled country-level panel dataset on agricultural taxes/subsidies is consistent with the prediction of the model. Does agricultural growth cause manufacturing growth? Empirically assessing the impact of agricultural growth on manufacturing growth is challenging because of endogeneity concerns. This paper attempts to circumvent the identification challenge by using weather variations to instrument for agricultural growth. The IV estimations show that agricultural growth has a significant positive impact on manufacturing growth, and it is larger than the OLS estimates. I discuss the empirical implications for agricultural policies, efficiency of the manufacturing sector, and for the role of agricultural growth in Africa's industrialization.

abstract: Understanding how adherence affects outcomes is crucial when developing and assigning interventions. However, interventions are often evaluated by conducting randomized experiments and estimating intent-to-treat effects, which ignore actual treatment received. Dose-response effects can supplement intent-to-treat effects when participants are offered the full dose but many only receive a partial dose due to nonadherence. Using these data, we can estimate the magnitude of the treatment effect at different levels of adherence, which serve as a proxy for different levels of treatment. In this dissertation, I conducted Monte Carlo simulations to evaluate when linear dose-response effects can be accurately and precisely estimated in randomized experiments comparing a no-treatment control condition to a treatment condition with partial adherence. Specifically, I evaluated the performance of confounder adjustment and instrumental variable methods when their assumptions were met (Study 1) and when their assumptions were violated (Study 2). In Study 1, the confounder adjustment and instrumental variable methods provided unbiased estimates of the dose-response effect across sample sizes (200, 500, 2,000) and adherence distributions (uniform, right skewed, left skewed). The adherence distribution affected power for the instrumental variable method. In Study 2, the confounder adjustment method provided unbiased or minimally biased estimates of the dose-response effect under no or weak (but not moderate or strong) unobserved confounding. The instrumental variable method provided extremely biased estimates of the dose-response effect under violations of the exclusion restriction (no direct effect of treatment assignment on the outcome), though less severe violations of the exclusion restriction should be investigated. / Dissertation/Thesis / Doctoral Dissertation Psychology 2018

This paper aims to quantify the municipal tax revenue effects of built-up area increases. The assumed existence of these effects is one of the key reasons for ongoing land consumption on the side of the municipalities. Some previous case studies however suggested that these effects might be not large enough especially in rural municipalities and would thus make land development not profitable. We estimate the effect of built-up industrial and commercial (BIC) area change on the business tax revenues in cross-sectional instrumental variable (IV) estimations. Based on detailed data for Bavaria, we find a significant and positive tax revenue effect of an increase in municipal BIC area. There exist strong differences in the size of this effect between urban and rural municipalities. The largest effects are generated by the BIC area in the large cities and become substantially smaller when these are dropped from the sample. Based on these findings, we reflect on the tradable planning permits (TPP) scheme recently discussed in the land use literature in the context of policies aiming to limit land consumption. Furthermore, we relate our estimates to the average municipal costs for land development and execute a number of robustness checks.

Submitted by Fabio Sobreira Campos da Costa (fabio.sobreira@ufpe.br) on 2016-07-14T15:34:21Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) TESE.pdf: 2470605 bytes, checksum: 6f8822a2b24b723e4595a90b6172168f (MD5) / Made available in DSpace on 2016-07-14T15:34:21Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) TESE.pdf: 2470605 bytes, checksum: 6f8822a2b24b723e4595a90b6172168f (MD5) Previous issue date: 2015-12-22 / FACEPE / A presente tese analisa os principais aspectos teóricos e a relação causal entre corrupção e desigualdade de renda nos municípios brasileiros. Para isso, utilizamos a metodologia de variáveis instrumentais. Essa estratégia requer a utilização de uma variável exógena que afeta o nível de interesse, em nosso caso a variável corrupção, e requer, também, que não seja correlacionada com nenhum fator não observável, relacionado com a variável dependente em um segundo estágio, conhecida como variável instrumental. Para analisar as possíveis heterogeneidades do problema, procedemos com uma análise a partir de regressão quantílica com variável instrumental. Como indicador de corrupção, utilizamos as informações disponíveis nos relatórios da Controladoria Geral da União (CGU) entre 2003 e 2006 e, para medir desigualdade, usamos os índices de Gini e de Theil. Os resultados indicam haver uma relação causal e positiva entre corrupção e desigualdade de renda. Nos quantis, a corrupção afeta os níveis de desigualdade até próximo a quantil 0,50. / The present Thesis aims to analyze the main theoretical features and the causal relationship between corruption and income inequality in the Brazilian municipalities. For this purpose, we applied the Instrumental Variable method. This strategy requires the use of an exogenous variable that affects the level of interest, in our case the variable corruption, and which is not correlated with any other not observable factor, related to a dependent variable in a second stage, known as Instrumental Variable. In order to analyze the possible heterogeneity of the problem, we investigated the data based on the quantile regression method. As a corruption indicator, we deployed the information contained in the Controladoria Geral da União (CGU) reports, released between 2003 and 2006, and as a device for measuring income inequality we applied the index of Gini and Theils. The results point to the presence of a causal and positive relationship between corruption and income inequality. Regarding to quantiles, the corruption affects inequality levels thereabout the quantile 0,50.

本文利用台灣2010年降低營利事業所得稅稅率之改革，檢驗公司稅率對受雇員薪資之影響。與過去文獻不同之處為，本文考量稅率與公司決策互相影響而產生內生性，因此以工具變數進行兩階段最小平方法以避免產生偏誤。並以綜合所得稅與營利事業所得稅資料對全體受雇員之個體進行分析，而非僅針對上市櫃之公司。最後發現有效稅率下降1% 能提升台灣整體受雇員平均 0.16% 的薪資。而考量個體異質性下，進行次組分析，發現在技術密集產業、大型公司、公司頂層員工所受到的影響較大；而其他產業，中小型公司以及公司基層員工的薪資受到稅率的影響較小，使得政府稅改政策並不如預期提升全民薪資。 / This paper estimates the effect of corporate income tax rate on wage by using the Taiwan Corporate Income Tax Reform of the year 2010 as a natural experiment. Different from the previous literature, we consider the endogenous relationship between the tax rate and the policy of the company. Thus, the two stage least squares method is used to overcome the endogeneity bias. Additionally, our participants include all the individual employees from the data of Individual Income Tax and Corporate Income Tax rather than only the employees of listed company. Our results indicate that a drop of 1% in corporate income tax rate would increase the wage by 0.16%. Furthermore, we test for the heterogeneous firm and worker effects by using subgroup analysis and find the technology-intensive industries, the large companies and the high-classed employees within firm have a greater impact. On the other hand, the non-technology-intensive industries, the small and medium company and the low-classed employees within firms have a smaller impact. Therefore, the Tax Reform didn’t increase the overall wage as expected.