Design and analysis of computer experiments when the output is highly correlated over the input space

To build a predictor, the output of a deterministic computer model or “code” is often treated as a realization of a stochastic process indexed by the code's input variables. The authors consider an asymptotic form of the Gaussian correlation function for the stochastic process where the correlation tends to unity. They show that the limiting best linear unbiased predictor involves Lagrange interpolating polynomials; linear model terms are implicitly included. The authors then develop optimal designs based on minimizing the limiting integrated mean squared error of prediction. They show through several examples that these designs lead to good prediction accuracy.     

Examining the Effects of Parental Absence on the Academic Achievement of Adolescents: The Challenge of Controlling for Family Income

Increasingly, researchers are concerned about how to best control for family income when examining the effects of parental divorce and the death of a parent on the children's academic achievement. Some researchers have argued that a predissolution control is preferable over a postdissolution control for family income, because parental divorce or the death of a parent nearly always causes family income reduction. Using the National Educational Longitudinal Study 1988–92 data set, this study has examined whether using a predissolution control for family income yields a different pattern of effects from when a postdissolution control is used. The results indicate that using a predissolution control rather than a postdissolution control for family income does yield a different pattern of effects.     

The Implication of Wife's Occupational Attainment for Husband's Class Identification*

Drawing upon research indicating that the subjective class identifications of working wives are to a considerable extent the result of their own occupational attainments, this study seeks to determine if husbands of working wives also derive perceptions of status from their wives' occupational attainments. Regression analysis is conducted on a subset of husbands of working wives in three General Social Surveys gathered from 1973 to 1975. A husband's level of subjective class identification is unaffected by the occupational achievements of his working wife. If husbands and wives derive perceptions of status from different sources, then they may not perceive themselves as sharing the same level of status in society.     

Participation in Athletics and Academic Achievement: A Replication and Extension*

Accumulating research based upon cross-sectional designs has shown relationships between involvement in athletics and academic achievement and aspiration. These results have been interpreted as contrary to Coleman's early conclusion that athletics diverted talent from academic programs. However, more recent study of this relationship in five metropolitan high schools seriously questions the causal interpretation of the effects of high school athletics upon academic achievement. Athletes have higher GPAs at the end of their high school career than at the start, but they do not gain as much as nonathletes over the years of high school, thus experiencing a relative decline in achievement. These results replicate an earlier study in twenty Wisconsin high schools. These similar results obtained in schools separated in time and space support Coleman's original view of the effect of athletics.     

Optimal design for the proportional odds model

The authors construct locally optimal designs for the proportional odds model for ordinal data. While they investigate the standard D‐optimal design, they also investigate optimality criteria for the simultaneous estimation of multiple quantiles, namely D_A ‐optimality and the omnibus criterion. The design of experiments for the simultaneous estimation of multiple quantiles is important in both toxic and effective dose studies in medicine. As with c‐optimality in the binary response problem, the authors find that there are distinct phase changes when exploring extreme quantiles that require additional design points. The authors also investigate relative efficiencies of the criteria.