To be or not to be: Control and balancing of type I and type II errors

The various costs of Type I and Type II errors of inference from data are discussed. Unfortunately, other things equal, Types I and II errors are inversely related. Six methods of minimizing Type I error in studies are presented, each of which may be employed even after data are collected. Type II errors may also be minimized by a combination of study design and analytic means. Although some of these techniques must be built into the study others can be undertaken when beginning work on a data set. Only one of these techniques will increase Type I error, and those involving data reduction may decrease both types of error.     

Socio-economic status and high-school friendship choice: Elmtown's youth revisited

Whereas Hollingshead's 1942 study of Elmtown High found socio-economic status homophily to be a prime criterion in students' friendship and dating choices, the present study, a secondary analysis of James Coleman's 1958 Elmtown data, finds socio-economic status homophily unimportant for friendship and dating choices. This decline in status homophily occurred at a time when class distinctions were becoming blurred in Elmtown and the “rating-and-dating” system was losing in national importance. Status homophily is not a constant, but a variable: clear class distinctions facilitate status homophily, while shifting class and occupational systems make for class-heterogeneous social networks.     

Distribution of power and agrarian reform: An explanatory model applied to India and Chile

This article appraises within an interdisciplinary framework key social, political, and economic aspects of the distribution of power in agrarian societies, and their interaction with various agrarian reform measures. Reform measures investigated include land reform proper, tenancy and wage regulations, land taxes, institutional credit, and productivity promotion measures.Use is made of alternate models that simulate economic development and agrarian reform under alternative sociopolitical regimes. The models distinguish four actor-groups: landlords, peasants, nonagriculturalists, and the state. Regimes differ, on the one hand, according to which actor-group is in power, and, on the other hand, according to which instruments of economic policy the state employs. The models are empirically applied and used to analyze the impact of agrarian reform measures in the 1960s and 1970s in two different regimes, India and Chile (under Allende).     

Consumer fraud and the elderly: a review of Canadian challenges and initiatives

Financial abuse is the most common type of elder abuse. Consumer fraud, a form of financial abuse perpetrated by criminals who do not know the victim, is not well studied. Seniors represent a disproportionate percentage of the victims of consumer fraud. This article reviews the data on the prevalence of consumer fraud (primarily telemarketing scams) in Canada. It examines the reasons why Canadian seniors are targets of fraud. It also describes many unique initiatives developed at the local, provincial and national level in Canada to educate seniors and those who care for them about the types of scams and the risks of fraud.     

Estimated Data Collection Organization Effect in the National Medical Care Expenditure Survey

The National Medical Care Expenditure Survey, which has a complex survey design, is further complicated by combining two independently drawn national samples of households from the Research Triangle Institute (RTI) and the National Opinion Research Center (NORC). It is assumed that since the structures of these national-area samples are similar, they are thereby compatible and allow for the derivation of statistically equivalent unbiased national estimates of relevant health parameters. In this paper, national parameter estimates of relevant demographic and health measures were separately produced for each of the two survey organizations and a determination was made regarding statistical equivalence across samples. In addition, the data-collection organization effect was estimated and tested within the framework of a multivariate analysis appropriate for complex survey data. The findings consistently revealed a statistically nonsignificant data-collection organization effect when testing for differences in domain estimates of the relevant health parameters under consideration.     

Double Sampling Designs in Multivariate Linear Models with Missing Variables

We provide a method for finding the optimal double sampling plan for estimating the mean value of a continuous outcome. It is assumed that the fallible and true outcome data are related by a multivariate linear regression model where only some of the explanatory variables are sampled. Conditions under which double sampling is preferred over standard sampling plans are determined. An application of the method to a well-known data set on air pollution is presented.