The distribution of person fit indices is not easy to describe in tests where the item sample is too small to conform to a theoretical asymptotic statistical distribution, particularly the normal N(0,1). In practice, it is always the fact and, consequently, it is difficult to get the critical percentile value indicating person misfit. First, we…

Data contained in this report were collected using the National Science Foundation's (NSF) fiscal year 1996 academic research and development expenditures survey and refer to science and engineering expenditures for separately budgeted research and development (R&D). R&D expenditures were collected from 493 institutions of higher education in the…

Develops stochastic model of library borrowing using Negative Binomial distribution applied to circulation data obtained from Huddersfield Public Library. Evidence concerning process of popularity decay is presented and method is given by which relegation tests can be constructed to maintain optimum turnover. Eight references and statistical…

Argues that the application of Bradford's original formulation to Leimkuhler's choice of bibliographic citations more closely fits the statistical data than does his own. Operational research techniques are assailed as inadequate and wasteful to the needs and purposes of the social sciences. Six references are provided. (RAA)

Examines the distribution of use of items in a large information store and suggests that a simple exponential model describes the process of obsolescence more precisely. Three tables, four figures, and a 17-item reference list are included. (Author/JL)

Reports on an investigation of enrollment of minority groups, primarily Black, in foreign language programs at Indiana University. Provides statistics from 1974 to 1976 comparing enrollment figures for minority and nonminority students and correlates high school scores and college admission measures with participation and success in the study of…

Presents two case studies of learners attempting to understand the concept of normal distribution, specifically why physical phenomena such as height fall into normal distributions. Draws conclusions about a Connected Mathematics learning environment that enables confrontation with epistemological anxiety and the features of modeling languages…

Reviews the use of chi-square statistics for goodness of fit. Examines how to help students relate their earlier work with independence to the use of chi-square statistics. (KHR)

Considers definitions of quantiles. Describes median and quartiles. Compares the usefulness of 3 different definitions of quartile using a computer program to simulate 500 quantiles on a sample of a fixed size. Five references are listed. (YP)

Defines grade level as an age grouping that yields an achievement distribution approximating a normal curve, with the distribution average at grade level in a typical school. With an average teacher, an average child gains a year. Educators must accept this normal range of reading achievement and adjust instruction to it. Includes eight…

Describes classroom activities and shows that statistics is a practical tool for solving real problems. Presents a histogram, a stem plot, and a box plot to compare data involving class enrollments. (YP)

This paper develops a multiplicative model as a means of combining ratings of criticality and frequency of various activities involved in job analyses. The model incorporates adjustments to ensure that effective weights of criticality and frequency are appropriate. An example of the model's use is presented. (TJH)

A comparison is made of the power and actual alpha levels of three tests of homogeneity for independent product-moment correlation coefficients using Monte Carlo methods while selectively studying sample size and varying the number of correlation reliabilities. How robust these are in applied work is discussed. (SLD)

Two experiments examined fifth-grade students' reasoning about data modeling: one involved development of hypermedia documents, the other, use of a simple randomization distribution to test students' hypotheses. Found that data construction and analysis provide an opportunity to involve students in the important enterprise of mathematical…

Type I error rate and power for the t test, Wilcoxon-Mann-Whitney test, van der Waerden Normal Scores, and Welch-Aspin-Satterthwaite (W) test are compared for two simulated independent random samples from nonnormal distributions. Conditions under which the t test and W test are best to use are discussed. (SLD)