The binomial error model includes form-to-form difficulty differences as error variance and leads to Ruder-Richardson formula 21 as an estimate of reliability. If the form-to-form component is removed from the estimate of error variance, the binomial model leads to KR 20 as the reliability estimate. (Author/BW)

This paper explains how to combine probabilities from discrete distributions, such as probability distributions for nonparametric tests. (Author/BW)

In response to criticism of the rise in indirect cost rates and their effect on federally funded research, the methods for calculating and applying indirect costs rates according to the new cost principles applicable to sponsored agreements are examined, and specific criticisms are addressed. (MSE)

A linear programing approach is applied to college cost information to suggest optimal tuition increases under some broad policy constraints concerning consistency of the rates. Factors such as balance of credit hour and per-student charges, and differentiation by program and other student groupings, are considered. (MSE)

Practical formulas for several analysis of variance (ANOVA) designs and models are presented which make it possible for readers to compute strength of association measures without the use of complete ANOVA tables. (Author/PN)

Aside from essential managerial information realized from faculty load studies, there exist two underlying fundamental principles: that equity is important (among individual faculty members, departments, and institutions); and there is a relationship between workload and the quality of education. (MLW)

A study of the Supplemental Educational Opportunity Grants program in the 1980-81 award period revealed that the state allotment formula produces a distribution of funds that undermines, rather than supports, the SEOG program's statutory purpose of delivering funds to needy students. (MLW)

The determination of acceptable performance in Thorndike's constant ratio standard of fair selection is considered. It is shown that suitable choice of acceptable performance can make any minority-majority selection disparity consistent with Thorndike's standard. A rule for determining acceptable performance which avoids the Petersen-Novick…

The statistically correct conditional maximum likelihood (CML) estimation method has not been used because of numerical problems. A solution is presented which allows a rapid computation of the CML esitmates also for long tests. CML has decisive advantages in the construction of statistical tests of goodness of fit. (Author/CP)

Discussed is the classical partition function for the ideal gas and how it differs from the exact value for bosons or fermions in the classical regime. The differences in the two values are negligible hence the classical treatment leads in the end to correct answers for all observables. (Author/DS)

Starting from the energy and degeneracy of the Landau levels of a free-electron gas in a magnetic field, the nonoscillatory term of the Landau diamagnetism is derived for T=O, using elementary algebra only. (Author/JN)

In this paper, a proposal is made by which a content validity coefficient can be calculated. An example of the use of the coefficient is given, demonstrating that different question combinations in a CSE biology examination in which a choice of questions was given gave different levels of content validity. (Author)

Mathematical formulas are proposed for estimating the market compatibility of academic programs, taking into account employment opportunities in the area and graduate production rate of competing area schools. Implications for program marketing are discussed. (MSE)

Methods are described for obtaining upper and lower bounds to both false-positive and false-negative decisions with a mastery test. These methods make no assumptions about the form of the true score distribution. (CTM)

A Monte Carlo simulation was employed to determine the accuracy with which the shrinkage in R squared can be estimated by five different shrinkage formulas. The study dealt with the use of shrinkage formulas for various sample sizes, different R squared values, and different degrees of multicollinearity. (Author/JKS)