O Other,Where Art Thou? Support for Multiparty Politics in the United States*

Objective. We test hypotheses about support for multiparty politics in the United States. We expect that individual‐level attitudes and state‐level partisan context determine who supports having more parties. Methods. Survey data are used to model attitudes about having additional parties to challenge Democrats and Republicans. Results. Self‐identified partisans are opposed to additional parties, but independents who “lean” toward a major party are most supportive. Independents who say they are closer to Democrats, as well as independents who say they are closer to Republicans, tend to support having new parties. The latter effect is contingent on a state's partisan context. Independents closer to Republicans are most receptive to additional parties in states where fewer conservative representatives are elected. Conclusions. Support for multiparty‐ism among “leaning” independents challenges our understanding of how they may be similar to partisan identifiers, and illustrates that a base for new parties may exist in the “mainstream” public if electoral rules were changed.     

Comparison of sample size formulae for 2 × 2 cross‐over designs applied to bioequivalence studies

We consider the comparison of two formulations in terms of average bioequivalence using the 2 × 2 cross‐over design. In a bioequivalence study, the primary outcome is a pharmacokinetic measure, such as the area under the plasma concentration by time curve, which is usually assumed to have a lognormal distribution. The criterion typically used for claiming bioequivalence is that the 90% confidence interval for the ratio of the means should lie within the interval (0.80, 1.25), or equivalently the 90% confidence interval for the differences in the means on the natural log scale should be within the interval (?0.2231, 0.2231). We compare the gold standard method for calculation of the sample size based on the non‐central t distribution with those based on the central t and normal distributions. In practice, the differences between the various approaches are likely to be small. Further approximations to the power function are sometimes used to simplify the calculations. These approximations should be used with caution, because the sample size required for a desirable level of power might be under‐ or overestimated compared to the gold standard method. However, in some situations the approximate methods produce very similar sample sizes to the gold standard method. Copyright © 2005 John Wiley & Sons, Ltd.     

Intimate partner homicide methods in heterosexual, gay, and lesbian relationships

Previous research indicates that the killing method used in homicides may reflect the motivation of the offender and qualities of the victim-offender relationship. The effect of gender and sexual orientation of intimate partner homicide offenders (N = 51,007) was examined with respect to the brutality of killing methods. Guided by previous research and theory, it was hypothesized that homicide brutality will vary with the offender's sexual orientation and gender, such that the percentage of killings coded as brutal will be higher for (a) gay and lesbian relative to heterosexual relations, (b) men relative to women, (c) gay relative to heterosexual men, and (d) lesbian relative to heterosexual women. The rates of intimate partner homicide were also hypothesized to vary with the gender of the partners, such that (a) homicide rates will be higher in gay relative to heterosexual and lesbian couples and (b) homicide rates will be lowest in lesbian couples. The results support all but one prediction derived from the two hypotheses. We predicted that men would kill their partners more brutally than would women, but the results indicate that the opposite is true.     

Using our brains to develop better policy

Current governmental practices often use a method called weight of evidence (WoE) to integrate and weigh different sources of information in the process of reaching a decision. Recent advances in cognitive neuroscience have identified WoE-like processes in the brain, and we believe that these advances have the potential to improve current decision-making practices. In this article, we describe five specific areas where knowledge emerging from cognitive neuroscience may be applied to the challenges confronting decisionmakers who manage risks: (1) quantifying evidence, (2) comparing the value of different sources of evidence, (3) reaching a decision, (4) illuminating the role of subjectivity, and (5) adapting to new information. We believe that the brain is an appropriate model for structuring decision-making processes because the brain's network is designed for complex, flexible decision making, and because policy decisions that must ultimately depend on human judgment will be best served by methods that complement human abilities. Future discoveries in cognitive neuroscience will likely bring further applications to decision practice.     

A Method for Simulating Multivariate Non Normal Distributions with Specified Standardized Cumulants and Intraclass Correlation Coefficients

Intraclass correlation coefficients (ICCs) are commonly used indices in subject areas such as biometrics, longitudinal data analysis, measurement theory, quality control, and survey research. The properties of the ICCs most often used are derived under the assumption of normality. However, real-world data often violate the normality assumption. In view of this, a computationally efficient procedure is developed for simulating multivariate non normal continuous distributions with specified (a) standardized cumulants, (b) Pearson intercorrelations, and (c) ICCs. The linear model specified is a two-factor design with either fixed or random effects. A numerical example is worked and the results of a Monte Carlo simulation are provided to demonstrate and confirm the methodology.     

Statistical Tools on the World Wide Web

The World Wide Web (WWW) represents a powerful tool for furthering the development and practice of statistics. The GASP (Globally Accessible Statistical Procedures) WWW site has been set up as a primary listing of statistical procedures which can be used over the WWW. This article highlights several possible approaches for making a procedure WWW accessible. These approaches effectively solve many of the problems typically encountered when using a new statistical procedure. Applying the methods discussed, any statistical technique can be made available to anyone with a forms- or Java-capable WWW browser. Procedures can be delivered in a virtually platform-independent manner with only minimal requirements on a user's hardware or software.     

Evaluating Direct Multistep Forecasts

ABSTRACT

This paper examines the asymptotic and finite-sample properties of tests of equal forecast accuracy and encompassing applied to direct, multistep predictions from nested regression models. We first derive asymptotic distributions; these nonstandard distributions depend on the parameters of the data-generating process. We then use Monte Carlo simulations to examine finite-sample size and power. Our asymptotic approximation yields good size and power properties for some, but not all, of the tests; a bootstrap works reasonably well for all tests. The paper concludes with a reexamination of the predictive content of capacity utilization for inflation.     

A multistage analysis strategy for a clinical trial to assess successively more stringent criteria for a primary endpoint with a low event rate

This paper describes how a multistage analysis strategy for a clinical trial can assess a sequence of hypotheses that pertain to successively more stringent criteria for excess risk exclusion or superiority for a primary endpoint with a low event rate. The criteria for assessment can correspond to excess risk of an adverse event or to a guideline for sufficient efficacy as in the case of vaccine trials. The proposed strategy is implemented through a set of interim analyses, and success for one or more of the less stringent criteria at an interim analysis can be the basis for a regulatory submission, whereas the clinical trial continues to accumulate information to address the more stringent, but not futile, criteria. Simulations show that the proposed strategy is satisfactory for control of type I error, sufficient power, and potential success at interim analyses when the true relative risk is more favorable than assumed for the planned sample size. Copyright © 2013 John Wiley & Sons, Ltd.     

Company structure and market strategy

The importance of the relationships between company strategies and structures is noted, as is the relative dearth of empirical work in this area. The article reports a preliminary study of 24 companies investigating the relationships between strategies of diversification and geographic dispersion of territories covered and structural variables relating to the number of structural differentiations in a company, the decentalization of decision-making and functional specialization. Contrary to the hypotheses proposed, these two strategic variables have different structural implications with geographic dispersion particularly associated with structural differentiation and diversification associated with decentralization.