Testing for lack of fit in linear multiresponse models based on exact or near replicates

Three procedures for testing the adequacy of a proposed linear multiresponse regression model against unspecified general alternatives are considered. The model has an error structure with a matrix normal distribution which allows the vector of responses for a particular run to have an unknown covariance matrix while the responses for different runs are uncorrelated. Furthermore, each response variable may be modeled by a separate design matrix. Multivariate statistics corresponding to the classical univariate lack of fit and pure error sums of squares are defined and used to determine the multivariate lack of fit tests. A simulation study was performed to compare the power functions of the test procedures in the case of replication. Generalizations of the tests for the case in which there are no independent replicates on all responses are also presented.     

Heterosexual Masculinities,Anti‐homophobias,and Shifts in Hegemonic Masculinity: The Identity Practices of Black and White Heterosexual Men

In this study, I use in‐depth interview data with black and white heterosexual men to explore shifts in the role of homophobia in the social construction of heterosexual masculinities. A continuum is introduced to map a range of interactional practices through which these men enact heterosexual masculinities. Heterosexual men who, on one end of the continuum, construct their heterosexual masculinities through homophobic practices establish strong boundaries of social distance from gays. The other end documents heterosexual men's anti‐homophobias, moving from men who establish weak boundaries to those who blur them. These heterosexual men's anti‐homophobic stances trade on the prestige of being tolerant of gays, with black men's anti‐homophobias drawing on their experiences with racism.     

Promotability of Host‐Country Nationals: A Cross‐Cultural Study

Drawing on person–environment fit and national identity theory, the article proposes that person–national culture fit is likely to influence the promotability of host‐country nationals in multinational firms. Focusing on fit with upward influence tactics, it suggests that the parent company's national culture influences managerial expectations of host‐country nationals in foreign subsidiaries. It argues that host‐country managers who demonstrate upward influence tactics that are culturally appropriate to the parent company's national culture will be more promotable than those who do not. Higher‐level supervisors were asked to assess the promotability of two direct subordinates, who were independently surveyed about the upward influence tactics they used. The study contrasted ingratiation, exchange of benefits and coalition, and directness influence tactics of host‐country nationals in domestic Ecuadorian firms with American and German multinationals in Ecuador. Compatible with our hypotheses, data from a sample of 79 firms suggest that exchange of benefits and coalition are more likely to be associated with promotability in German than in domestic Ecuadorian firms. In addition, upward‐appeal assertiveness is more likely to be associated with promotability in American than in domestic Ecuadorian firms.     

Combining forecast quantiles using quantile regression: Investigating the derived weights, estimator bias and imposing constraints

SUMMARY A novel proposal for combining forecast distributions is to use quantile regression to combine quantile estimates. We consider the usefulness of the resultant linear combining weights. If the quantile estimates are unbiased, then there is strong intuitive appeal for omitting the constant and constraining the weights to sum to unity in the quantile regression. However, we show that suppressing the constant renders one of the main attractive features of quantile regression invalid. We establish necessary and sufficient conditions for unbiasedness of a quantile estimate, and show that a combination with zero constant and weights that sum to unity is not necessarily unbiased.     

Measuring the complexity of generalized linear hierarchical models

Measuring a statistical model's complexity is important for model criticism and comparison. However, it is unclear how to do this for hierarchical models due to uncertainty about how to count the random effects. The authors develop a complexity measure for generalized linear hierarchical models based on linear model theory. They demonstrate the new measure for binomial and Poisson observables modeled using various hierarchical structures, including a longitudinal model and an areal‐data model having both spatial clustering and pure heterogeneity random effects. They compare their new measure to a Bayesian index of model complexity, the effective number p_D of parameters (Spiegelhalter, Best, Carlin & van der Linde 2002); the comparisons are made in the binomial and Poisson cases via simulation and two real data examples. The two measures are usually close, but differ markedly in some instances where p_D is arguably inappropriate. Finally, the authors show how the new measure can be used to approach the difficult task of specifying prior distributions for variance components, and in the process cast further doubt on the commonly‐used vague inverse gamma prior.     

Studying categorical data over time

A review of recently developed simple techniques for analyzing data in the form of proportions from replicated, cross-sectional, sample surveys. (A) One variable over time—tests for homogeneity, pooling homogeneous proportions, tests for linear trends with as few as three time points, tests for departures from linearity; (B) Two variables with a constant percentage difference—four fold tables as rudimentary causal models, decomposing change with linear flow graphs; (C) Extension of the flow graph techniques to—three or more variables and changing coefficients; (D) Comments implying panel designs are somewhat over-rated and successive cross-sections somewhat under-rated.     

The abortion issue and Title VII employment discrimination: The microcosm of the law

This paper presents a discussion of a 1989 United States federal appeals court ruling in the case of Doe v. First National Bank. The case is the first ruling on the federal appellate level involving a claim of Title VII employment discrimination based upon alleged employer knowledge of and animus towards an employee who had undergone an abortion. Doe raises the ethical questions of how a highly charged, national social issue can be reduced to interpersonal dealings between a supervisor and her subordinate in the workplace. The facts of the case are presented in a chronological manner, followed by a discussion of both the court rulings in the case and the utility of employing this case as an excellent tool in the teaching of business ethics and society.