Poisson regression with misclassified counts: application to cervical cancer

"Errors in disease classification can give misleading inferences for covariate effects when the probability of error is itself related to the covariates. More accurate inferences are possible using supplemental data on both true and fallible disease counts at various covariate levels. We present a method for incorporating such supplemental data into disease rate regression and use it to show that, although observed intercountry differences in European cervical cancer mortality rates are exaggerated by errors in death certification, they are not completely explained by such errors." The data concern Belgium, England and Wales, France, and Italy.     

Population changes and the cost of health care for the elderly in Poland

"The author...has attempted to measure the effects of population changes upon the costs of health care [in Poland] by applying a simulation model. In this model the total cost of health care is a function of the per capita cost of health care by age, sex, and place of residence (urban, rural) and population structure.... The paper includes...the results concerning population 60 years of age and over." Data are from several official health-related surveys carried out in 1989.     

Strategic planning for the board

Although the planning operation is regarded by some observers as unrealistic in conditions of rapid change and increasing competition, the discipline of strategic thinking and the need for strategic leadership continue to be of vital importance. The author examines the purpose of the Board of Directors and its role in the management of strategy.     

THE SHORT-TERM EFFECT OF ELECTION POLLS ON FOREIGN EXCHANGE RATES: THE 1988 CANCDIAN FEDERAL ELECTION

Polls do not simply measure public opinion; they also providevital information that the public can use to form opinions andto make decisions. Using multiple regression analysis, the authorsfound that published polls had a powerful impact on the valueof the Canadian dollar during the 1988 Canadian federal election.This appears to have been due to the unprecedented importanceof economic issues in the federal campaign, the distinctivepositions taken by the major political parties with respectto the U.S.-Canadian Free Trade Agreement, and to the proliferationof "horse-race" polls in the media. These results have implicationsfor understanding the behavior of other elite groups, such asfinancial contributors and volunteer campaign workers, who mayalso use published horse-race polls to handicap election outcomes.     

Recent trends in the process of stratification

Using the 14 annual cross-sections from the General Social Survey, we specify a "basic model" of attainment and describe the year-by-year fluctuations in its parameters. The results are partially consistent with theories describing the gradual growth of universalistic patterns of stratification and mobility. Under a linear model of educational achievement, we find that the direct effects of race are weakening and the returns to class-based advantages are declining in tandem. The contours of the socioeconomic "gender gap" are also changing in important ways, with the male intercept declining at a rapid pace and the female term registering small and insignificant year-by-year gains. At the same time, the returns to experience and schooling are increasing for men, whereas the corresponding returns for women have remained stable over the 15-year period. This pattern of interaction effects implies that the size of the gender gap varies over time and across different population groups.     

Refugee research in Africa: problems and possibilities

The author discusses the magnitude of the refugee problem in selected countries of Africa, with a focus on the "conceptual, methodological and practical problems faced by refugee researchers in Africa." (SUMMARY IN FRE AND SPA)     

Measurement of Hispanic ethnicity in the U.S. census: an evaluation based on latent-class analysis

"Two models, the U.S. census model and the latent-class model, are compared in their application to evaluating measurements of ethnicity. Although the census approach assumes that the response categories of a questionnaire item correspond to groups in the population, the latent-class approach seeks to assess whether any set of response categories can represent observed ethnic heterogeneity. Data collected using the 1990 census Hispanic-origin question and other instruments for measuring ethnicity suggest that the latent-class approach is superior whenever the response categories are not known to be valid. In particular, using the latent-class model, this article rejects the census model's assumption of a single dimension of meaning underlying responses to the Hispanic-origin question."     

THE EFFECTS OF PRESTIGE NAMES IN QUESTION WORDING

Although survey researchers are often warned against using prestigenames in questions (e.g., identifying Contra aid as "PresidentReagan's policy"), prestige names are still commonly used. Tolearn more about the effects of using prestige names, we analyzetwo sets of experiments—on judicial confirmation electionsand on an initiative on tax indexing. The results indicate thatprestige names do more than shift responses in one direction—they eliminate the effect of education on DKs and provide moreof a political basis for the responses.