THE RESULTS OF THIS STUDY CLEARLY INDICATE THE INADVISABILITY OF ATTEMPTING TO PERFORM ITEM ANALYSES BASED UPON SMALL SAMPLES. THE SINGLE ITEM DISCRIMINATION INDEX BASED UPON SAMPLES OF SIZE 15, 30, OR 60, HAS LITTLE OR NO VALUE. THOSE BASED UPON SAMPLES OF 120 ARE ONLY INDICATIVE IN NATURE. (LP)

This study was concerned with the identification of the job dimension underlying the job elements of the Position Analysis Questionnaire (PAQ), Form B. The PAQ is a structured job analysis instrument consisting of 187 worker-oriented job elements which are divided into six a priori major divisions. The statistical procedure of principal components…

An index that reflects the accuracy of selection associated with a predictive validity of "r" is presented. Based on Sheppard's thoerem on median dichotomies, it is a measure of improvement over chance assignment to "accept" (or "reject"). Because the index is a measure of the accuracy of this assignment, rather than…

The consistently high suicide rates of older White males provide a strong rationale for increased study of the geriatric suicide problem. Data related to White males aged 60 and older who killed themselves in Arizona during 1970 through 1975 are presented. (Author)

In some linear regression problems samples may be unwittingly drawn from a heterogeneous population. The exploratory procedure described in this paper can be used to detect heterogeneity in a regression problem associated with the intercept. (Author)

The judged accuracy of narrative paragraph types and total reports of a new Minnesota Multiphasic Personality Inventory (MMPI) clinical interpretation simulation program is studied. A linear regression analysis of variance of overall narrative ratings with two narrative and four patient variables suggest that this system has considerable…

In discussion about search keys, concern has been expressed as to how the number of items retrieved by a single value relates to collection size. This paper creates a statistical model that attempts to give some insight into this behavior. (6 references) (Author)

Attempts to ascertain to what degree structural factors might explain variance in the work-value systems of white collar workers. (Author)

A variety of problems associated with the interpretation of traditional canonical correlation are discussed. A response surface approach is developed which allows for investigation of changes in the coefficients while maintaining an optimum canonical correlation value. Also, a discrete or constrained canonical correlation method is presented. (JKS)

Similarity between pairs of scores with a common mean and variance can be expressed in terms of the absolute differences between them. The distribution of such absolute differences between pairs of normally distributed scores with a common mean and variance is discussed, with procedures for calculating moments and areas within this distribution.…

Four-year colleges and universities submit faculty teaching load and instructional cost data annually to the Delaware Study of Instructional Costs and Productivity. While the Delaware Study currently adjusts the calculation of annual FTE students to account for the difference in annual student credit hours (SCH) earned by students at semester and…