Reports of Conferences, Institutes, and Seminars

Late winter and early spring of 2009 was a busy conference time for librarians. Some of the conferences held during that time are reported herein: the American Library Association Midwinter Conference, January 23–28; Electronic Resources and Libraries (E&RL), February 9–12; the North Carolina Serials Conference, March 27; the New England Technical Services Librarians' (NETSL) Spring Conference, April 17; and the Technical, Electronic and Digital Services Interest Group (TEDSIG) Spring Meeting, April 24.     

CORE (Cost of Resource Exchange) Standard

This article describes the CORE (Cost of Resource Exchange) standard. CORE defines an XML schema to allow the exchange of financial information relating to library resources among library systems.     

A generalized model of logistic regression for clustered data

This paper proposes a generalized logistic regression model that can account for the correlation among responses on subunits. The subunits may arise as data on multiple observations within an individual. This method generalizes earlier work by Rosner (1984 a,b) and others. Methodological generalizations include: (1) the use of the more general Polya-Eggenberger distribution instead of the beta-binomial distribution to model the correlation structure, so that cases with negative, positive, or zero intraclass correlation can be handled; (2) a stepwise approach; (3) linear and non-linear regression; and, (4) the inclusion of the case of a truncated distribution. The model can accommodate missing data and covariates on the unit and subunit level. The derivative-free simplex algorithm is used to estimate the parameters.

The model is applied to data describing the progression of obstruction in coronary disease where multiple arterial segments are studied for each patient. The correlation in response that may exist for these multiple segments is accounted for in the analyses while attempting to examine associations with individual-specific (e.g., history of diabetes) and segment-specific (e.g., initial percent stenosis) covariates. Analyses were performed on a data set describing 382 patients with unoperated coronary artery disease and two coronary angiograms separated by at least one month and on a data set describing 284 patients undergoing percutaneous transluminal coronary angioplasty and studied by coronary angiograms.     

Multivariate applications of the ash in regression

A simple algorithm for estimating the regression function over the United States is introduced. The approach allows for data obtained from a complicated sampling design, as well as for the inclusion of a few additional covariates. The regression estimates are obtained from an associated probability density estimate, namely the averaged shifted histogram. The algorithm has proven especially successful over a large mesh, say 300 by 200 nodes, in a data rich setting, even on a 486 computer running Splus. We currently run much higher resolution meshes on a Pentium. Commonly available alternative codes including kriging failed to produce useful estimates in this setting.     

New simulation results for the calibration and inverse median estimation problems

This paper gives the results of a new simulation study for the familiar calibration problem and the less familiar inverse median estimation problem. The latter arises when one wishes to estimate from a linear regression analysis the value of the independent variable corresponding to a specified value of the median of the dependent variable. For example, from the results of a regression analysis between stress and time to failure, one might wish to estimate the stress at which the median time to failure is 10,000 hours. In the study, the mean square error, Pitman closeness, and probability of overestimation are compared for both the calibration problem and the inverse median estimation problem for (1) the classical estimator, (2) the inverse estimator, and (3) a modified version of an estimator proposed by Naszodi (1978) for both a small sample and a moderately large sample situation.     

Kronecker Products in ANOVA—A First Step

In the complete balanced model for the analysis of variance, the equivalence of sums of squares and quadratic forms is seen to imply well-fitting patterns involving Kronecker products of identity matrices and scalar multiples of matrices with all elements equal to 1. The questions of symmetry, idempotency, and orthogonality so central to this topic are answered by simple multiplications; ranks are determined from simple traces. The associations between the forms of the two-factor model are presented here in a way that is accessible to first-year students and makes generalizations to higher order models transparent. The lack of patterns in incomplete or unbalanced models is noted. Additional steps in design and analysis are suggested in the references.     

Student Selection and Incentives

The paper discusses the impact of performance based selection in secondary education on student incentives. The theoretical approach combines human capital theory with signaling theory. The consideration of signaling offers an explanation for observed performance of educational systems with a standard peer effect argument. More specifically it can be optimal to select students according to ability even if selective systems do not outperform comprehensive systems in tests. Selection achieves the same output with lower private costs for the students. The paper questions the strong focus on educational tests to measure the efficiency of selective systems as long as these tests provide no information about a student’s incentives and private costs.