Estimating DEA confidence intervals with statistical panel data analysis

This paper describes a statistical method for estimating data envelopment analysis (DEA) score confidence intervals for individual organizations or other entities. This method applies statistical panel data analysis, which provides proven and powerful methodologies for diagnostic testing and for estimation of confidence intervals. DEA scores are tested for violations of the standard statistical assumptions including contemporaneous correlation, serial correlation, heteroskedasticity and the absence of a normal distribution. Generalized least squares statistical models are used to adjust for violations that are present and to estimate valid confidence intervals within which the true efficiency of each individual decision-making unit occurs. This method is illustrated with two sets of panel data, one from large US urban transit systems and the other from a group of US hospital pharmacies.     

Interval estimators for the population mean for skewed distributions with a small sample size

In finite population sampling, it has long been known that, for small sample sizes, when sampling from a skewed population, the usual frequentist intervals for the population mean cover the true value less often than their stated frequency of coverage. Recently, a non-informative Bayesian approach to some problems in finite population sampling has been developed, which is based on the 'Polya posterior'. For large sample sizes, these methods often closely mimic standard frequentist methods. In this paper, a modification of the 'Polya posterior', which employs the weighted Polya distribution, is shown to give interval estimators with improved coverage properties for problems with skewed populations and small sample sizes. This approach also yields improved tests for hypotheses about the mean of a skewed distribution.     

Assessment of motion media on believability and credibility: An exploratory study

The radical shift by news audiences away from newspaper to motion media (video stories on TV, web, cell phones, handhelds) prompted the assessment of media modality (text, text + picture, video) and source (public relations, news, user-generated content [UGC]) effects on credibility and more importantly perceived veridicality, a perception of an object as being accurate and believable regardless of the source, as well as impact of messages. The findings revealed that motion media modality significantly enhances believability judgments and perceived veridicality, which is independent of source cue, in which news source garnered no greater credibility than PR or UGC source.     

A two-stage phase II trial design utilizing both primary and secondary endpoints

Phase II trials in oncology drug development are usually conducted to perform the initial assessment of treatment activity. The common designs in this setting, for example, Simon 2-stage designs, are often developed based on testing whether a parameter of interest, usually a proportion (e.g. response rate), is less than a certain level or not. These designs usually consider only one parameter. However, sometimes we may encounter situations where we need to consider not a single parameter, but multiple parameters. This paper presents a two-stage design in which both primary and secondary endpoints are utilized in the decision rules. The family-wise Type 1 error rate and statistical power of the proposed design are investigated under a variety of situations by means of Monte-Carlo simulations.     

Quality of Parenting as Mediator of the Effect of Childhood Defiance on Adolescent Friendship Choices and Delinquency: A Growth Curve Analysis

Social scientists agree that childhood antisocial behavior portends adolescent delinquency, but there is little agreement regarding the theoretical processes that account for this behavioral continuity. Latent growth curve modeling was used to test latent trait and social influence explanations for this association. The analyses used data collected annually over a 4‐year period from a sample of 149 boys, 157 girls, and their parents. Contrary to latent trait theories, we found no direct association between oppositional/defiant behavior during childhood and a trajectory of increasing involvement with deviant peers and delinquency during adolescence. Rather, early oppositional/defiant behavior undermined effective parenting practices. The latter, in turn, predicted an increasing affiliation with deviant peers and delinquency during adolescence. Improvements in parenting during adolescence decreased delinquency indirectly by reducing affiliation with deviant peers. Overall, the results support a life course development model in which difficult behavior during childhood increases the probability of adolescent deviant behavior because of its disruptive effect on quality of parenting.     

Family Influences on the Social Participation of Youth: The Effects of Parental Social Involvement and Farming

Abstract Parents shape children's social choices through their social and economic actions. Parental social participation connects children to a civic culture and encourages involvement in civic groups. Parents' ties to farming in farm‐dependent communities further enhance children's civic orientations by providing added opportunities and incentives for social participation. Data from the Iowa Youth and Families Project confirm these hypotheses, showing that the children of farmers and of rural leaders are more likely to participate in civic groups. These results establish parental social involvement as a source of social capital and demonstrate the importance of farm influences for understanding the social involvement of youth in rural society.     

Religion and vulnerability among low-risk adolescents

Studies of resilient youth have documented the helpful influence of religion in their lives. In the absence of positive influences among high-risk youth, religion and religious organizations in particular often serve to promote prosocial outcomes. This study, on the other hand, investigates whether religion and religious change are comparably effective influences in reducing the vulnerability of low-risk youth to drinking, drug use, delinquency, problems in school, and poor grades. Using data from the National Longitudinal Study of Adolescent Health, this study identified at least one influential aspect of religion that serves as a direct protective factor across each mode of vulnerability, and also documented indirect pathways of religious influence. We discuss the findings and consider the variable influence of religion on youth.     

Economic Stress in Lives: Developmental Perspectives

A major task for research on the social costs of economic stress is to trace how macrosocial changes affect increasingly smaller social units and ultimately those microsocial phenomena that directly influence children in their families. In this paper, we specify linkages between macroeconomic change and children's development by tracing deprivational effects through family adaptations in the household economy and in personal relationships. Our findings from research on children and families of the Great Depression are discussed in relation to an interactional model of the process by which families adapt to stressful times.     

Life Course Transitions,the Generational Stake,and Grandparent‐Grandchild Relationships

Drawing on past research and prominent theoretical orientations, this research note suggests new approaches to intergenerational dynamics. For 316 grandparent‐grandchild pairs, we found that the transition of grandchildren to higher education, controlling for other transitions, improves the quality of the grandparent‐grandchild relationship. For grandparent mentoring, however, we see evidence of a generational stake, with grandparents overestimating their mentoring role, compared to grandchildren, during this transition. This generational stake reflects the importance of grandparent education, with increased mentoring for the college‐going grandchildren of college‐educated grandparents. These findings indicate that the intergenerational literature can be significantly advanced by taking a long‐term perspective, incorporating multiple points of view, and examining contextual variation. Moreover, greater understanding of these intergenerational ties will benefit research on families and individual development.