The paper applies a non-parametric propensity score matching approach to evaluate the effects of two types of farm programs (agri-environment (AE) programs and the less favoured area (LFA) scheme) on input use and farm output of individual farms in Germany. The analysis reveals a positive and significant treatment effect of the LFA scheme for farm sales and the area under cultivation. Participants in AE schemes are found to significantly increase the area under cultivation (in particular grassland), resulting in a decrease of livestock densities. Furthermore, participation in AE programs significantly reduced the purchase of farm chemicals (fertilizer, pesticide). We also find substantial differences in the treatment effect between individual farms (heterogeneous treatment effects). Farms which can generate the largest benefit from the program are most likely to participate. (authors' abstract) / Series: Department of Economics Working Paper Series

Uppsatsen syftar till att utvärdera effekten av Förberedande och orienterande utbildning, vilket är en insats som ges till arbetssökande för att gynna framtida aktivitet inom arbetskraften. Insatsen är en av Arbetsförmedlingens största, och bedöms öka ytterligare framöver. En kombination av coarsened exact matchning samt propensity score matchning används för att försöka fastställa kausala effekter på individnivå med mikrodata från Arbetsförmedlingen och SCB. Deltagare som skrevs in hos Arbetsförmedlingen 2008 studeras och utfallen som tillämpas är inkomster och socialbidragstagande, vilka mäts 2010, 2011, 2012 respektive 2013. Resultaten visar att deltagare i insatsen generellt har sämre arbetsmarknadsutfall än jämförelsegruppen som består av öppet arbetslösa. Känslighetsanalysen visar att det kan finnas vissa problem med selektion, men dessa bedöms inte omkullkasta resultaten. Vidare fastställs att de skattade effekterna avtar med uppföljningstiden, vilket tyder på att det kan finnas inlåsningseffekter. En annan förklaring till att negativa effekter av deltagande kan uppstå är enligt den nationalekonomiska litteraturen att deltagande i insatsen signalerar något negativt till potentiella arbetsgivare, snarare än att insatsen bidrar med en produktivitetsökning för den arbetssökande individen.

Thesis advisor: Joseph J. Pedulla / There is considerable evidence that new teachers are ill prepared for classroom practice, including self-reported evidence collected from teachers (e.g., Levine, 2006), and statistical evidence for differences in the achievement of students with new versus more experienced teachers (Rivkin, Hanushek, and Kain, 2005). In light of the challenges encountered by new teachers (e.g., Levine, 2006), this study examined the value of different forms of teacher knowledge for teachers with different levels of experience. In particular, this study investigated the interactive relationship between teaching experience and teacher content knowledge, and student achievement in mathematics and science. In New York City, Boyd et al. (2009) linked practice-focused teacher preparation to student mathematics achievement in the first year of teaching and teacher content preparation to achievement in the second. However, other studies demonstrated interactions between teaching experience and content knowledge with different interpretations (e.g., Kukla-Acevedo, 2009; Monk, 1994). At the same time, this study examined the interactive relationship between teaching experience and teachers' pedagogical content knowledge, and student achievement. Extant models of teacher career development (Huberman, 1989; National Research Council, 2010) and how teacher education affects student achievement (e.g., Desimone, 2009) offered theoretical grounding for the study. With nationally representative samples of fourth and eighth grade U.S. students--participants in the 2011 Trends in International Mathematics and Science Study--this study employed hierarchical linear modeling to address its research questions among an array of student achievement outcomes in the domains of mathematics and science. This study attempted to account for salient student, teacher, and contextual factors, and the probabilities of teachers' receipt of various teacher education "treatments" (i.e., propensity score analysis) to reduce the plausibility of selection threats to internal validity. The study found no evidence for relationships between teacher content knowledge or pedagogical content knowledge and student mathematics and science achievement in fourth and eighth grade. Furthermore, the results indicated no interactive relationships between forms of teacher knowledge and teaching experience, and student achievement in these grades/subjects. The limitations of cross-sectional, observational studies using large-scale data and directions for further research are discussed. / Thesis (PhD) — Boston College, 2013. / Submitted to: Boston College. Lynch School of Education. / Discipline: Educational Research, Measurement, and Evaluation.

Thesis advisor: Henry I. Braun / City Connects is a school-based model that identifies the strengths and needs of every student and links each child to a tailored set of intervention, prevention, and enrichment services in the school or community. The purpose of this study was to conduct a comprehensive evaluation of the City Connects treatment effects on academic performance (both MCAS scores and grade point average (GPA) grades) in middle school using student longitudinal records. Parallel analyses were conducted: one evaluated the City Connects elementary intervention (serving kindergarten to fifth grades) and the other one evaluated the City Connects middle school intervention (serving sixth to eighth grades). A series of two-level hierarchical linear models with middle school achievement scores adjusted and/or propensity score weights applied were used to answer the research questions of interest. In addition, to make a causal inference, a sensitivity analysis was conducted to examine whether or not the estimated treatment effects resulted from the first two analyses were robust to the presence of unobserved selection bias. The results showed that students who were exposed to the City Connects elementary intervention significantly outperformed their counterparts, who graduated from the comparison elementary schools, on academic achievement in all middle school grades. However, in the case of the City Connects intervention schools that served middle school grades, since all students only received a maximum of one year of City Connects middle school intervention, it was still too soon to expect any significant changes. Moreover, the estimated treatment effects of the City Connects elementary intervention were only mildly sensitive to the presence of some forms of hidden bias, which made the causal inference of City Connects on middle school academic achievement quite plausible. / Thesis (PhD) — Boston College, 2015. / Submitted to: Boston College. Lynch School of Education. / Discipline: Educational Research, Measurement and Evaluation.

O objetivo dessa dissertação é avaliar o impacto de um programa de treinamento voltado para trabalhadores, utilizando o propensity score matching, mas com dois tipos de abordagem, uma não-paramétrica e a outra semi-paramétrica. Para estimação não paramétrica foi utilizado um método proposto por Li, Racine e Wooldridge (2009) e para estimação semi-paramétrica, o modelo utilizado foi o Generalized Additive Model proposto por Hastie e Tibshirani (1990). Os resultados obtidos indicam que os dois métodos utilizados apresentam estimativas tão boas ou melhores do que quando estimadas paramétricamente. / The goal of this thesis is to evaluate the impact of a job training program using propensity score matching methods with two types of approaches: a nonparametric e another semiparametric. For non-parametric estimation was used a method proposed by Li, Racine and Wooldridge (2009) and for the semiparametric model the Generalized Additive Model proposed by Hastie and Tibshirani (1990). The results indicate that both methods provide estimates as good or better than when parametrically estimated.

En recherche médicale, les cohortes permettent de mieux comprendre l'évolution d'une pathologie et d'améliorer la prise en charge des patients. La mise en évidence de liens de causalité entre certains facteurs de risque et l'évolution de l'état de santé des patients est possible grâce à des études étiologiques. L'analyse de cohortes permet aussi d'identifier des marqueurs pronostiques de l'évolution d'un état de santé. Cependant, les facteurs de confusion constituent souvent une source de biais importante dans l'interprétation des résultats des études étiologiques ou pronostiques. Dans ce manuscrit, nous présentons deux travaux de recherche en Biostatistique dans la thématique des scores de propension. Dans le premier travail, nous comparons les performances de différents modèles permettant d'évaluer la causalité d'une exposition sur l'incidence d'un événement en présence de données censurées à droite. Dans le second travail, nous proposons un estimateur de courbes ROC dépendantes du temps standardisées et pondérées permettant d'estimer la capacité prédictive d'un marqueur en prenant en compte les facteurs de confusion potentiels.En cohérence avec l'objectif de fournir des outils statistiques adaptés, nous présentons également dans ce manuscrit une application nommée Plug-Stat®. En lien direct avec la base de données, elle permet de réaliser des analyses statistiques adaptées à la pathologie afin de faciliter la recherche épidémiologique et de mieux valoriser les données de cohortes observationnelles. / In medical research, cohorts help to better understandthe evolution of a pathology and improve the care ofpatients. Causal associations between risk factors andoutcomes are regularly studied through etiological studies. Cohorts analysis also allow the identification of new markers for the prediction of the patient evolution.However, confounding factors are often source of bias in the interpretation of the results of etiologic or prognostic studies.In this manuscript, we presented two research works in Biostatistics, the common topic being propensity scores.In the first work, we compared the performances of different models allowing to evaluate the causality of an exposure on an outcome in the presence of rightc ensored data. In the second work, we proposed anestimator of standardized and weighted time-dependentROC curves. This estimator provides a measure of theprognostic capacities of a marker by taking into accountthe possible confounding factors. Consistent with our objective to provide adapted statistical tools, we also present in this manuscript an application, so-calledPlug-Stat®. Directly linked with the database, it allows toperform statistical analyses adapted to the pathology in order to facilitate epidemiological studies and improve the valorization of data from observational cohorts.

Submitted by PPG Economia do desenvolvimento (economia-pg@pucrs.br) on 2018-10-19T21:27:53Z No. of bitstreams: 1 BLANCA_LILA_GAMARRA _MOREL_TES.pdf: 4056778 bytes, checksum: c115258b7c6a31660496964e5ea4548e (MD5) / Approved for entry into archive by Sheila Dias (sheila.dias@pucrs.br) on 2018-10-22T13:12:26Z (GMT) No. of bitstreams: 1 BLANCA_LILA_GAMARRA _MOREL_TES.pdf: 4056778 bytes, checksum: c115258b7c6a31660496964e5ea4548e (MD5) / Made available in DSpace on 2018-10-22T13:18:58Z (GMT). No. of bitstreams: 1 BLANCA_LILA_GAMARRA _MOREL_TES.pdf: 4056778 bytes, checksum: c115258b7c6a31660496964e5ea4548e (MD5) Previous issue date: 2018-08-24 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior - CAPES / The main objective of the thesis is to analyze the level of quality of life between the coastal and non-coastal municipalities of Brazil associated to the environmental, economic and social dimensions. The specific objectives are: to characterize the socioeconomic structure of coastal and non-coastal municipalities and to identify whether there are differences in the quality of life of the population. For that, the approach of human development and economic geography was used as a theoretical reference in order to define quality of life and well-being, as well as the economic and social characteristics that are related to these concepts. The use of this theoretical framework allowed to reach a better definition of the dimensions of analysis and to proceed to the empirical analysis using the Propensity Score Matching (PSM) method. The aim was to identify differences in the quality of life between the coastal and non-coastal areas of Brazil. The empirical analysis suggests evidence contrary to the common sense that in regions with greater potential for concentration of economic activities there is a higher quality of life. Regarding the economic characteristics, the results found corroborate the theory, showing that, in fact, there is a flagrant economic concentration in the coastal municipalities. However, the analysis of environmental and social characteristics showed that although coastal municipalities are more developed from a purely economic point of view, when considered other aspects of development these municipalities perform worse. / O objetivo principal da tese ? analisar o n?vel de qualidade de vida entre os munic?pios costeiros e n?o costeiros do Brasil associado ?s dimens?es ambiental, econ?mica e social. Os objetivos espec?ficos constituem-se em: caracterizar a estrutura s?cio econ?mica dos munic?pios costeiros e n?o costeiros e identificar se existem diferenciais de qualidade de vida da popula??o. Para tanto, foram utilizados como referencial te?rico a abordagem do desenvolvimento humano e Geografia Econ?mica de modo a definir qualidade de vida e bem-estar, bem como as caracter?sticas econ?micas e sociais que est?o relacionadas com esses conceitos. A utiliza??o desse referencial te?rico permitiu alcan?ar tamb?m uma melhor defini??o das dimens?es de an?lise para, a partir de ent?o, proceder ? an?lise emp?rica utilizando-se o m?todo Propensity Score Matching (PSM). Buscou-se, assim, identificar as diferen?as de qualidade de vida entre a zona costeira e n?o costeira brasileira. A an?lise emp?rica realizada sugere evid?ncias contr?rias ao senso comum de que em regi?es com maior potencial de concentra??o de atividades econ?micas h? maior qualidade de vida. No que diz respeito ?s caracter?sticas econ?micas, os resultados encontrados corroboram a teoria, mostrando que existe, de fato, flagrante concentra??o econ?mica nos munic?pios costeiros. No entanto, a an?lise das caracter?sticas ambientais e sociais mostraram que embora os munic?pios costeiros sejam mais desenvolvidos do ponto de vista puramente econ?mico, quando considerados outros aspectos do desenvolvimento esses munic?pios apresentam desempenho pior.

Cystic fibrosis (CF) is the most common life-restricting, genetically inherited disease among Caucasians affecting approximately 30,000 people in the United States. Lung disease is the major cause of morbidity and mortality in CF. A number of oral, inhaled, and intravenous therapies are available to combat CF lung disease. Of these, this research project focused on inhaled dornase alfa, oral azithromycin, inhaled tobramycin, and inhaled aztreonam. Data to address three research aims were requested and obtained from the Cystic Fibrosis Foundation Patient Registry (CFFPR). The first aim examined the use of inhaled dornase alfa in younger children with CF. With no clinical efficacy data of dornase alfa in children ≤ 6 years of age, the study utilized subsequent forced expiratory volume in 1 second (FEV₁) measured between 6 - 7 years of age, to assess the effectiveness of long-term dornase alfa use ≤ 6 years of age. Propensity score methods were used to reduce the likelihood of treatment indication bias. The results suggested that receiving treatment with dornase alfa before 6 years of age did not improve FEV₁ between 6 - 7 years. Unmeasured covariates leading to treatment indication bias were likely one of the key explanations for these results. Additionally, lack of a more sensitive outcome than FEV₁ to assess lung function in young patients with early lung damage was thought to be another reason for the failure to reject the null hypothesis. The second aim assessed the long-term clinical effectiveness of chronic azithromycin use on the rate of FEV₁ decline in CF patients between 6 - 20 years of age. This study was novel in that the rate of FEV₁ decline, rather than change in FEV₁ from baseline, was the primary outcome, which was characterized using propensity score matching followed by a linear mixed model analysis. The results of the analysis suggested that the rate of FEV₁ decline was slower in patients who did not receive chronic treatment with azithromycin. Treatment indication bias was thought to play an important role in the direction of the association between treatment and outcome. Associations between FEV₁ % predicted and many of the other study variables included in the analysis were consistent with previous studies. The final aim compared the clinical effectiveness of a combination of inhaled tobramycin and aztreonam with inhaled tobramycin alone on the rate of FEV₁ decline in CF patients between 6 - 20 years of age. This aim was novel in that the effect of this combination treatment on rate of decline in FEV₁ has never been assessed. A linear mixed model analysis was used after matching patients in the two treatment groups on their propensity scores. Once again, the results were contrary to the alternative hypothesis with the combination group having a steeper rate of FEV₁ decline than the group that was treated with tobramycin alone. An important reason for this result was thought to be unresolved treatment indication bias that could not be eliminated even with the use of the propensity score methods used to test the associated hypothesis. The use of validated methods of analysis, i.e., propensity scores, to counter treatment indication bias using the largest available observational dataset for CF, was one of the key strengths of this study. Moreover, this study highlighted important weaknesses in the CFFPR with regards to lack of data on patient and physician-level variables - an area of active interest for the Cystic Fibrosis Foundation.

Approximately 35.1% of live births for the state of Georgia were delivered by the cesarean delivery procedure with significant variation among hospitals. The purpose of this research was to develop a population-based hospital profiling methodology for study of the cesarean delivery procedure. This was a retrospective, observational design, using a 2012 linked dataset that included maternity deliveries from all nonfederal hospitals. The research was guided by Robson 10 Group Classification System, propensity score methodologies, and ethical precepts, for the development of hospital profiles and the study of variations in the cesarean delivery procedure. Key research questions aimed to determine whether hospital profiling methodologies differed according to risk adjustment methods and statistical techniques. Propensity score matching with stratification methods aimed to determine whether there were differences in patient treatment effects on the cesarean delivery outcome. Findings suggested there was a significant difference in hospital ranks and model effects according to the statistical technique and the risk adjustment methods applied. Propensity score matching with stratification demonstrated an increased risk of the cesarean delivery procedure across strata, with the majority of high risk patients situated in the 90th percentile ranges and questionable utilization practice among other strata. Applying profiling methodologies at the facility and population level could advance statewide quality improvement programs for the timely reduction in the variation of inappropriate utilization of the cesarean delivery procedure.

Background: Among patients with multiple chronic conditions, care coordination and integration remains one of the major challenges facing the U.S. health care system. A home-based, patient-centered primary care program has been offered through the Veterans Health Administration (VHA) since the 1970s for frail veterans who have difficulty accessing VHA clinics. The VHA Home Based Primary Care (VHA HBPC) aims to integrate primary care, rehabilitation, disease management, palliative care, and coordination of care for frail individuals with complex, chronic diseases within their homes. Early research suggested that VHA HBPC was associated with positive outcomes (e.g., reduced resource use and patient satisfaction). However, evidence regarding the effect of the VHA HBPC program on health services use (especially hospital and nursing home use), expenditures, and other patient outcomes remains limited. The present study is designed to fill this gap as the rise in the number of veterans with complex health care needs will likely increase in the coming decades. Objectives: The current study aimed to examine the impact of VHA HBPC on health services use, expenditures, and mortality among a cohort of new VHA HBPC enrollees identified in the national VHA data system. The specific aims of this study were: 1) to examine the effect of VHA HBPC on major health service use (hospital, nursing home, and outpatient care) paid for by the Veterans Administration; 2) to examine the effect of VHA HBPC on total health services expenditures; and 3) to examine whether VHA HBPC enrollees experienced similar mortality and survival as compared to a matched concurrent cohort. Methods: This study used a retrospective cohort design. A new VHA HBPC enrollee cohort (the treatment group) and a propensity matched comparison cohort (the comparison group) were identified from VHA claims in fiscal years (FY) 2009 and 2010 and were followed through FY 2012. Data on health service use, expenditures, and mortality/survival data were obtained via the VHA administrative datasets (i.e., Decision Support System, Purchased Care, and Vital Status Files). Propensity scores of being enrolled in the VHA HBPC were generated by a logistic regression model controlling for potential confounders. After 41,244 matched pairs were determined adequate through several diagnostic methods, means tests, relative risk analyses, and generalized linear models were used to estimate the effect of VHA HBPC on outcomes. Additionally, a Cox proportional hazards regression model was used to estimate the effect of VHA HBPC on survival. Subgroup analyses were conducted stratifying by age (85 and older), comorbidities (2 or more), and the receipt of palliative care. Based on the results of the original analyses, a series of sensitivity analyses were conducted that modified the described sample selection criteria and matching algorithm. Results: Analyses of the original cohort revealed that VHA HBPC patients had significantly higher risks of being admitted into a hospital (RR 1.53, 95% CI 1.51-1.56) or nursing home (RR 1.65, CI 1.50 - 1.81). The average total expenditures during the study period were significantly higher for the VHA HBPC group as compared to the control group ($85,808 vs. $44,833, respectively; p < .001). In terms of mortality and survival, VHA HBPC enrollees had higher mortality (RR 1.45, CI 1.43 - 1.47), and shorter survival (HR 1.89, CI 1.86 - 1.93) as compared to those in the comparison group. Subgroup analyses found that these relationships generally remained when stratified by age 85 or older or having two or more comorbidities. However, for those who received palliative care, VHA HBPC participants had significantly lower risk of VHA hospitalization overall (RR 0.84, CI 0.81 - 0.87) and immediately prior to death. Finally, exploratory post-hoc analysis suggested that VHA HBPC recipients were at higher risk of VHA hospitalization at 30 (RR 1.11, CI 1.06 - 1.16), 60 (RR 1.16, CI 1.11 - 1.20), and 90 days (RR 1.16, 1.12 - 1.21) prior to death relative to the comparison group. After selecting only those that had a baseline hospitalization and refining the matching algorithm to account for time to death and additional comorbidities, VHA HBPC participants who had been enrolled in the program for at least six months had lower risks for hospital (RR 0.89, CI 0.88 - 0.90) and nursing home admissions (RR 0.74, CI 0.67 - 0.81). However, total expenditures remained significantly higher among those in VHA HBPC relative to the comparison group ($89,761 vs. $85,371, respectively; p < .001). Discussion: This study found that without accounting for important covariates such as initial hospitalization, time to death, and a range of comorbidities, VHA HBPC was associated with higher health service use, higher expenditures, higher mortality, and shorter survival as compared to a similar group of patients not receiving VHA HBPC. After accounting for these factors, VHA HBPC was associated with a lower risk of nursing home use, and after six months, VHA HBPC was associated with lower risk of both nursing home and hospital use. These findings suggest that while VHA HBPC may improve quality of life and patient satisfaction through patient-centered integrated primary care, it may not generate cost savings for the healthcare system. Future research is needed to understand variation in program implementation and how this affects the impact of VHA HBPC on service use and cost.