Causal Effects of Single-Sex Schools on College Entrance Exams and College Attendance: Random Assignment in Seoul High Schools

Despite the voluminous literature on the potentials of single-sex schools, there is no consensus on the effects of single-sex schools because of student selection of school types. We exploit a unique feature of schooling in Seoul—the random assignment of students into single-sex versus coeducational high schools—to assess causal effects of single-sex schools on college entrance exam scores and college attendance. Our validation of the random assignment shows comparable socioeconomic backgrounds and prior academic achievement of students attending single-sex schools and coeducational schools, which increases the credibility of our causal estimates of single-sex school effects. The three-level hierarchical model shows that attending all-boys schools or all-girls schools, rather than coeducational schools, is significantly associated with higher average scores on Korean and English test scores. Applying the school district fixed-effects models, we find that single-sex schools produce a higher percentage of graduates who attended four-year colleges and a lower percentage of graduates who attended two-year junior colleges than do coeducational schools. The positive effects of single-sex schools remain substantial, even after we take into account various school-level variables, such as teacher quality, the student-teacher ratio, the proportion of students receiving lunch support, and whether the schools are public or private.     

Self-Regulatory Driving Behaviors: Gender and Transportation Support Effects

This study examined the relationship between transportation support and self-regulatory driving behaviors of 566 community-dwelling older adults living in retirement communities, with a focus on gender differences. The results of logistic regression analysis showed that older women were more likely to avoid driving at night or on the highway than their male counterparts. Transportation support from peer friends was found to increase the likelihood of self-regulatory driving behaviors. The findings of this study imply that transportation policy and driving safety programs for older adults need to be developed, considering available transportation alternatives and gender differences in driving behaviors.     

Assessing the trait theory of leadership using self and observer ratings of personality: The mediating role of contributions to group success

The trait theory of leadership suggests that personality traits influence leader emergence and effectiveness. While initial empirical evidence supports this perspective, the majority of studies have examined the relationship between personality and leadership using self ratings of personality. We believe that this research may underestimate the relationship between personality and leadership. We propose that personality assessed using both self and observer ratings explains more variance in leadership than self ratings of personality alone. Results from 155 participants in leaderless group discussions supported this hypothesis. Further, relative weight analysis revealed that observer ratings of extraversion explained the largest percentage of variance in leadership, followed by self ratings of openness to experience and observer ratings of openness to experience. Results of two-stage least squares regression analysis showed that the relationship between personality and leadership was mediated by contributions to group success. The implications of these results and directions for future research are discussed.     

On the Performance of Popular Unit-Root Tests Against Various Nonlinear Dynamic Models: A Simulation Study

ABSTRACT

There is a widespread perception that standard unit-root tests have poor discriminatory power when they are applied to time series with nonlinear dynamics. Via Monte Carlo simulations this study re-examines the finite sample properties of selected univariate tests for unit-root and stationarity under a broad class of nonlinear dynamic models. Our simulation experiments produce a couple of interesting findings. First, performance of tests is driven by the degree of underlying persistence rather than the nonlinear dynamics per se. Tests under study exhibit reasonable performance for nonlinear models with mild persistence, while the accuracy of inference deteriorates substantially when the models are highly persistent regardless of the linearity. Second, when it comes to deciding which one to identify first between linearity and stationarity, our results suggest to conduct linearity test first to enhance the reliability of test inference.     

Estimating LAD Regression Coefficients with Best Subset Points

We propose a simple algorithm for estimating LAD (least-sum of absolute deviations) regression coefficients. The algorithm finds the best subset points to use to obtain the global optimizer in the series of a backtracking process by narrowing down the search space. Hence, the computational complexity reduces to a practical level. To evaluate the performance of the algorithm for LAD regression, we conducted simulation studies. The results showed that the algorithm is very competitive in terms of computation times required to get the same or similar results in comparison with other methods even when heavy censoring is given. Moreover, because the algorithm is quite simple and easy to understand, it can be implemented easily for various objective functions that are the variations of LAD.     

Dynamic Bayesian analysis for irregularly and incompletely observed contingency tables

Pre-election surveys are usually conducted several times to forecast election results before the actual voting. It is common that each survey includes a substantial number of non-responses and that the successive survey results are seen as a stochastic multinomial time series evolving over time. We propose a dynamic Bayesian model to examine how multinomial time series evolve over time for the irregularly observed contingency tables and to determine how sensitively the dynamic structure reacts to an unexpected event, such as a candidate scandal. Further, we test whether non-responses are non-ignorable to determine if non-responses need to be imputed for better forecast. We also suggest a Bayesian method that overcomes the boundary solution problem and show that the proposed method outperforms the previous Bayesian methods. Our dynamic Bayesian model is applied to the two pre-election surveys for the 2007 Korea presidential candidate election and for the 1998 Ohio general election.     

On asymptotic properties of Bayesian partially linear models

In this paper, we present large sample properties of a partially linear model from the Bayesian perspective, in which responses are explained by the semiparametric regression model with the additive form of the linear component and the nonparametric component. For this purpose, we investigate asymptotic behaviors of posterior distributions in terms of consistency. Specifically, we deal with a specific Bayesian partially linear regression model with additive noises in which the nonparametric component is modeled using Gaussian process priors. Under the Bayesian partially linear model using Gaussian process priors, we focus on consistency of posterior distribution and consistency of the Bayes factor, and extend these results to generalized additive regression models and study their asymptotic properties. In addition we illustrate the asymptotic properties based on empirical analysis through simulation studies.     

A robust support vector machine for labeling errors

Support vector machine (SVM) is sparse in that its classifier is expressed as a linear combination of only a few support vectors (SVs). Whenever an outlier is included as an SV in the classifier, the outlier may have serious impact on the estimated decision function. In this article, we propose a robust loss function that is convex. Our learning algorithm is more robust to outliers than SVM. Also the convexity of our loss function permits an efficient solution path algorithm. Through simulated and real data analysis, we illustrate that our method can be useful in the presence of labeling errors.     

The Rural/Urban Volunteering Divide

VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations - Are rural residents more likely to volunteer than those living in urban places? Although early sociological theory posited...