A procedure for the analysis of judgmental estimates is derived. It uses a t-distribution concept to derive the relative experience of an expert and uses the derived experience as weight to obtain a weighted mean. Using this method, error in estimation was reduced by an average of one-half. (Author/BW)

The regression-discontinuity design (RDD) offers the possibility of making inferences about causal effects from observations on selected groups. Data from such a design are considered to have a truncated bivariate distribution. For the RDD, maximum likelihood parameter estimation procedures and tests of hypotheses are presented. (Author/BW)

Various federal agencies and the courts are facing questions about the legality and potential discrimination in sex-segregated mortality tables and discrepancies in retirement and insurance benefits. By using unisex mortality tables, TIAA-CREF is an unwilling pioneer in the insurance industry, but the tables' ultimate form remains in question.…

To clarify the findings of R. Gillett (1991) about the inequality of the means of test scores of minority and majority examinees, the standard errors of the quota-selected sample means and the sampling distribution of these means were studied through Monte Carlo simulation. Results explain that the quota selection inequality results from…

Distributions of psychiatric classifications among 42,479 persons with mental retardation were reviewed to determine relative rates of disorders for persons with both mental retardation and mental disorder. Analyses indicated that rates of reported schizophrenic and psychotic disorders were generally higher than expected relative to rates of other…

Applies a probabilistic model to a contingency table with cross-citation frequency counts between scientific journals. The differences between model-based expected citation frequencies and observed frequencies (net citation balances) are used to derive an interjournal influence score. This measure of overall influence is compared with two other…

Developed are simple recursion formulas for generating the exact distribution of the longest run of heads, both for a fair coin and for a biased coin. Discusses the applications of runs-related phenomena such as molecular biology, Markov chains, geometric variables, and random variables. (YP)

Presents an activity considering whether a difference exists in the age of Oscar winners. Describes how to draw a stem plot and a box plot as an example of implementing the recommendations of the NCTM Standards. Provides tables showing the name, movie titles, and ages of the Oscar winners since 1928. (YP)

This study investigated the distributional properties of the standardized residual that is commonly used in the Rasch model calibration program to develop various indices of item and person fit. The power of the standardized residual to detect measurement disturbances is also addressed. (TJH)

Describes a study that supported two hypotheses concerning the use of anaphors in information retrieval: first, that anaphors tend to refer to integral concepts rather than peripheral concepts; second, that various term weighting schemes are affected differently by anaphoric resolution. The implications for the use of anaphoric resolution for…

An evolved form of the Edwards and Beckworth (1989) model for probability selection for Scholastic Achievement Test takers using truncated normal distributions is presented. It is shown that the arguments of Holland and Wainer are not sufficient to reject this model. (SLD)

Investigates the relationships between the parameters of the Simon-Yule model and the shapes of three bibliometric distributions: Lotka's Law of Scientific Productivity; Bradford's Law of Bibliometric Scattering; and Zipf's Law of Word Frequency. The results indicate that the probability of a new entry determines the characteristics of all three…

Robustness of the school-level item response theoretic (IRT) model to violations of distributional assumptions was studied in a computer simulation. In situations where school-level precision might be acceptable for real school comparisons, expected a posteriori estimates of school ability were robust over a range of violations and conditions.…

The use of p-values in combining results of studies often involves studies that are potentially aberrant. This paper proposes a combined test that permits trimming some of the extreme p-values. The trimmed statistic is based on an inverse cumulative normal transformation of the ordered p-values. (SLD)

Estimation/smoothing methods that are flexible enough to fit a wide variety of test score distributions are reviewed: kernel method, strong true-score model-based method, and method that uses polynomial log-linear models. Applications of these methods include describing/comparing test score distributions, estimating norms, and estimating…