ON SELECTING VARIABLES AND ASSESSING THEIR PERFORMANCE IN LINEAR DISCRIMINANT ANALYSIS

Linear discriminant analysis between two populations is considered in this paper. Error rate is reviewed as a criterion for selection of variables, and a stepwise procedure is outlined that selects variables on the basis of empirical estimates of error. Problems with assessment of the selected variables are highlighted. A leave-one-out method is proposed for estimating the true error rate of the selected variables, or alternatively of the selection procedure itself. Monte Carlo simulations, of multivariate binary as well as multivariate normal data, demonstrate the feasibility of the proposed method and indicate its much greater accuracy relative to that of other available methods.     

U.S. Aviation Regulations Increase Probability of Midair Collisions

A purely stochastic Monte Carlo model is used to compare the relative midair collision course probabilities and mean closing velocities of four systems of rules for aircraft cruising altitudes as a function of altitude error: (1) current U.S. federal rules, (2) random altitudes, and (3) two proposed alternatives to the current rules. This model increments error while: (1) counting collisions among cruising pairs of aircraft following the four rules being tested on random headings between randomly placed airports, and (2) calculating mean closing velocities for each rule. The calculations verify that: (1) federal rules increase collision course probabilities by about four times more than for a chaotic system of aircraft cruising at randomly selected altitudes, (2) risk is directly proportional to the level of compliance, and (3) mean closing velocities resulting from the current rule are slightly less than for random altitudes, while being almost twice as high as for the proposed rules. High closing velocities are shown to increase the collision probability.     

Public Opposition to a Proposed Nuclear Waste Repository in Canada: An Investigation of Cultural and Economic Effects

Residents of four northern communities were surveyed about Atomic Energy of Canada Limited's proposal to site an underground repository for high-level nuclear waste somewhere in the Canadian Shield. Opposition to the repository was relatively strong in all communities, but was strongest among aboriginal respondents. Path analysis revealed that trust in nuclear regulators, faith in science and technology, and anticipated net costs were important mediators of this effect. Aboriginals were less trusting, exhibited less faith in science and technology, and perceived the costs associated with the repository to be higher than their nonaboriginal counterparts. No support was found for the hypothesis that, after controlling for aboriginal status, financially insecure individuals would display greater support for the nuclear waste repository than financially secure individuals. Policy implications for balancing perceived risks and siting needs are discussed.     

Acting out in children: A review of the literature

This paper presents a review of the literature on the meaning, etiology and categorization of acting out symtomatology in children. It is apparent that there is no one cause nor is there one well defined disorder.DSM-III categories as they relate to behavior disorders are discussed and the problems of diagnosis are addressed. Appropriate treatment goals and interventions cannot be developed unless more global aspects of a child's functioning and the environment in which he lives are assessed.     

The development and construct validation of a consumer satisfaction questionnaire for psychiatric inpatients.

A consumer satisfaction scale was developed for psychiatric inpatients. The scale was administered to a state-wide sample of (n = 366) patients discharged from acute psychiatric units throughout Missouri. The scale was factor analyzed and given to another sample of (n = 390) discharges. Three out of five factors that were found in the first sample were replicated in a factor analysis of data from the second sample. Acceptable coefficient alphas were obtained on all three factors, demonstrating internal reliability. The three factors were labeled Treatment Effectiveness, Trust of Staff, and Hospital Environment. Comparable levels of satisfaction with prior studies were found on the first general satisfaction factor of Treatment Effectiveness. Subjects discriminated areas of satisfaction by expressing higher levels of dissatisfaction on the factors of Trust of Staff and Hospital Environment.