There has been a growing interest in using propensity scores in the analysis of observational studies. The propensity score is a balancing mechanism that works to create groups of subjects which have a similar distribution on background covariates. Matching, stratification, inverse propensity treatment weighting and regression adjustment are all strategies that can be used with the propensity score to create balance between groups of subjects. The aim of this paper is to introduce propensity scores and the different techniques which make use of them. We use data obtained from the Women's Health Initiative to demonstrate each of the different methods for propensity score analysis. In the example analysis we examined the association between dog ownership and CVD. The results of our analysis were quite consistent, and demonstrate the propensity score analysis can be used effectively to balance treated and untreated groups within an observational study.

In Sweden and many other countries of the European Union throughout the entire twentieth century the self-employment has been important and essential for sustainable growth and welfare. The dynamics of self-employment have had an variance between areas that are characterized and distinguished by different labour market conditions, entrepreneurial traditions and other structural factors. In general, the agricultural importance in Swedish rural regions has declined over time, hence other small and medium industries have grown and gained more significance in these type of regions. A good example of the decline in industrialised importance for Swedish rural region is Bergslagen. Where the majority of the jobs in the traditional sectors of iron-ore mining were lost without any renewal. This study highlights the self-employment option out of the two-folded perspective. The first and foremost is to investigate the spatial i.e. rural-urban differences with the relation to individual motives as their choice for self-employment. Also, a second and as a side goal of the research, the propensity for self-employment is considered in order to get wider insight of the regional start-up activity in urban and rural regions. By using the rich survey dataset on ex-post motives and the register-based longitudinal data from Swedish Statistics (SCB). The regional differences in determinants for self-employment are examined by applying the binary probit and multinomial logit regressions. Additionally, in order to get a coherent unity within the multidimensional motives a factor analysis is used for grouping the motive variables into the four groups. Simultaneously for easier association to labour market the motives are also divided into typology of pull-push categories. The results regarding propensity for self-employment show small or no differences in the tendency for starting the own venture in rural side or urban regions. On the other hand, results concerning spatial aspects and motive behind the choice of occupation shows that a mutually pull and push reasons are more linked to the urban region than to the rural. Comparable results are observed on the subject of single ex-post motive “non-monetary” in respective area i.e. rural and urban.

Entrepreneurial activity is a vital part of any economy whether developed or developing. In South Africa the primary focus of the government has been on the development of previously disadvantaged communities and designing programs to encourage the participation of women in entrepreneurship. This study sought to profile South African female entrepreneurs as well as investigate their risk taking propensity in relation to other constructs. T-tests as well as ANOVAs were conducted on data obtained from 122 female entrepreneurs across South Africa. The results indicated that the female entrepreneurs in the current study were typically white, English speaking, married with children, were well educated and possessed previous working experience mainly in managerial positions. The results also indicated a significant difference in risk taking propensity according to age with entrepreneurs who are 35 years and younger having a higher risk taking propensity than those who are 36 years and older. However, no significant differences were found in risk taking according to entrepreneurial motivations, gender role orientation, level of education and previous experience. The practical implications of the study are discussed together with the limitations.

The annual incidence of knee replacement (KR) procedures in the United States is predicted to reach over 3.5 million by the year 2030. KR is the current definitive treatment for debilitating knee osteoarthritis (KOA). There has yet to be substantial research regarding the impact of KR on participation in community activities and quality of life. The hypotheses evaluated in this dissertation were that persons following KR will have 1) faster gait speed and 2) lower risk of participation restrictions than persons without KR; and 3) a decreased risk of all-cause mortality compared to persons without KR. To address the first two hypotheses, we collected data from subjects with KOA from the Multicenter Osteoarthritis Study and the Osteoarthritis Initiative, large cohorts of older adults with or at risk of KOA at the time of enrollment. In the first study, KR did not have an effect on gait speed overall and among most subgroups, however subjects with a slow gait speed prior to KR did have an 80% increased risk (RR 1.8, 95% CI 1.1, 3.0) of having a healthy gait speed compared with non-KR subjects. In the second study, KR was associated with a small decreased risk of having participation restriction (RR 0.82, 95% CI 0.67, 0.99). The third study used data on patients with KOA from the Clinical Practice Research Datalink, a database of clinical information on > 8 million people throughout the United Kingdom. There was a decrease in the death rate among KOA subjects who had a KR compared to those who did not, and the hazard of death was reduced by over one half in the first five years after the procedure (HR 0.46 (95% CI 0.43, 0.51). For most subjects, this benefit did not extend longer than five years, and patients least likely to have KR (due to clinical and medical presentation) showed an increased hazard of death compared to the non-KR subjects. In conclusion, the results of this dissertation support the hypotheses that KR confers a positive benefit to activity and participation related pursuits which may extend to survival in the short term for some people.

No description available.

The research presented in this thesis considers the question of whether individual-level risk preferences are consistent or inconsistent across decision domains. For example, do people make the same decisions with respect to work, health and finance? Some previous authors have suggested that risk preferences are inconsistent, e. g. Kahneman and Tversky (1979), while others have put forward the idea that people have generalised tendencies to take or avoid risks, e. g. Sitkin and Pablo (1992). The work of Sitkin and Pablo was drawn upon to develop hypotheses concerning the conceptualisation and construction of risk propensity. Risk propensity was operationalised as the degree of consistency of cross-domain risk preferences. It was proposed that a propensity to take or avoid risks is associated with whether individuals have consistent tendencies across different decision domains, that personality will be a key predictor of risk propensity, and that inconsistent cross-domain risk preferences will be associated with risk domain-specific cognitive and emotional aspects of decision making. A survey measure was developed to assess risk and decision preferences both across and within the domains of work, health and finance. Biographical and personality factors were also measured. The sample comprised 360 participants drawn from five sample groups chosen to capture a range of risk preferences. The results showed that risk propensity can be conceptualised and measured in terms of the consistency of cross-domain risk preferences. People who were consistent in their risk preferences were characterised by the personality traits of emotional stability, low extroversion, low openness and high agreeableness. Additionally, consistent risk preferences were associated with relative consistency of attention to situational information and perceived risk. The majority of participants, however, had different risk preferences in different domains, and showed variability in their decision preferences. The implications of the research for understanding risk propensity and risk management are discussed.

Propensity score methods (PSM) that have been widely used to reduce selection bias in observational studies are restricted to a binary treatment. Imai and van Dyk extended PSM to estimate non-binary treatment effect using stratification with P-Function, and generalized inverse treatment probability weighting (GIPTW). However, propensity score (PS) matching methods on multiple treatments received little attention, and existing generalized PSMs merely focused on estimates of main treatment effects but omitted potential interaction effects that are of essential interest in many studies. In this dissertation, I extend Rubin’s PS matching theory to general treatment regimens under the P-Function framework. From theory to practice, I propose an innovative distance measure that can summarize similarities among subjects in multiple treatment groups. Based on this distance measure I propose four generalized propensity score matching methodologies. The first two methods are extensions of nearest neighbor matching. I implemented Monte Carlo simulation studies to compare them with GIPTW and stratification on P-Function methods. The next two methods are extensions of the nearest neighbor caliper width matching and variable matching. I define the caliper width as the product of a weighted standard deviation of all possible pairwise distances between two treatment groups. I conduct a series of simulation studies to determine an optimal caliper width by searching the lowest mean square error of average causal interaction effect. I further compare the ones with optimal caliper width with other methods using simulations. Finally, I apply these methods to the National Medical Expenditure Survey data to examine the average causal main effect of duration and frequency of smoking as well as their interaction effect on annual medical expenditures. Using proposed methods, researchers can apply regression models with specified interaction terms to the matched data and simultaneously obtain both main and interaction effects estimate with improved statistical properties.

This paper uses the Dutch panel data from LISS, Longitudinal Internet Studies for the Social Science in trying to establish if a relationship between individuals’ risk propensity and job search activity is present. When looking at employed and unemployed job seekers jointly, a positive significant relationship is present. Looking at these groups separately shows that the relationship is driven by employed job seekers. No relationship for unemployed job seekers can be identified when being looked at separate. However, when taking into account possible biases from changes in risk propensity over time as well as the classification of actively searching individuals, no relationship is present at all. We hence are cautious towards the significant estimates received that potentially could suffer from biases. Further studies should be made with a bigger sample of individuals and a continuously updated measure of risk propensity to minimizing potential bias.

The first chapter uses the propensity score matching method to measure the average impact of insurance on health service utilization in terms of office-based physician visits, total number of reported visits to hospital outpatient departments, and emergency room visits. Four matching algorithms are employed to match propensity scores. The results show that insurance significantly increases office-based physician visits, and its impacts on reported visits to hospital outpatient departments and emergency room visits are positive, but not significant. This implies that physician offices will receive a substantial increase in demand if universal insurance is imposed. Government will need to allocate more resources to physician offices relative to outpatient or emergency room services in the case of universal insurance in order to accommodate the increased demand. The second chapter studies the sensitivity of propensity score matching methods to different estimation methods. Traditionally, parametric models, such as logit and probit, are used to estimate propensity score. Current technology allows us to use computationally intensive methods, either semiparametric or nonparametric, to estimate it. We use the Monte Carlo experimental method to investigate the sensitivity of the treatment effect to different propensity score estimation models under the unconfoundedness assumption. The results show that the average treatment effect on the treated (ATT) estimates are insensitive to the estimation methods when index function for treatment is linear, but logit and probit model do better jobs when the index function is nonlinear. The third chapter proposes a Cross-Sectionally Varying (CVC) Coefficient method to approximate individual treatment effects with nonexperimental data, the distribution of treatment effects, the average treatment effect on the treated and the average treatment effect. The CVC method reparameterizes the outcome of no treatment and the treatment effect in terms of observable variables, and uses these observables together with a Bayesian estimator of their coefficients to approximate individual treatment effects. Monte Carlo simulations demonstrate the efficacy and applicability of the proposed estimator. This method is applied to two datasets: data from the U.S. Job Training Partnership ACT (JTPA) program and a dataset that contains firms’ seasoned equity offerings and operating performances.

abstract: Random Forests is a statistical learning method which has been proposed for propensity score estimation models that involve complex interactions, nonlinear relationships, or both of the covariates. In this dissertation I conducted a simulation study to examine the effects of three Random Forests model specifications in propensity score analysis. The results suggested that, depending on the nature of data, optimal specification of (1) decision rules to select the covariate and its split value in a Classification Tree, (2) the number of covariates randomly sampled for selection, and (3) methods of estimating Random Forests propensity scores could potentially produce an unbiased average treatment effect estimate after propensity scores weighting by the odds adjustment. Compared to the logistic regression estimation model using the true propensity score model, Random Forests had an additional advantage in producing unbiased estimated standard error and correct statistical inference of the average treatment effect. The relationship between the balance on the covariates' means and the bias of average treatment effect estimate was examined both within and between conditions of the simulation. Within conditions, across repeated samples there was no noticeable correlation between the covariates' mean differences and the magnitude of bias of average treatment effect estimate for the covariates that were imbalanced before adjustment. Between conditions, small mean differences of covariates after propensity score adjustment were not sensitive enough to identify the optimal Random Forests model specification for propensity score analysis. / Dissertation/Thesis / Ph.D. Psychology 2013