Further influence measures in factor analysis

Various influence measures are discussed for sensitivity analysis in factor analysis. The internal norm is used to characterize the vector-valued influence curves in factor analysis. Influence curves for the chi-square goodness of fit test, and the determinants of the model covariance matrix and unique variance matrix are derived. They are found to have simple formulas which are easy to be interpreted and have nice distributions for calibration. The likelihood displacement is also applied to sensitivity analysis in maximum likelihood factor analysis.     

INCIDENCE AND CONSEQUENCES OF RISK‐TAKING BEHAVIOR IN TOURNAMENTS—EVIDENCE FROM THE NBA

We empirically explore the relevance and efficiency of risk‐taking behavior in tournaments. We use data from the National Basketball Association (NBA) and measure risk‐taking by the fraction of three‐point shots in basketball games. We examine how point differences between teams during games affect their subsequent risk‐taking behavior. It is found that teams that are trailing are more likely to increase their use of three‐point shots. We additionally analyze the consequences of this change in behavior. Enhanced risk‐taking is inefficient in the vast majority of cases and is only beneficial if a team is trailing by a rather large margin. We discuss possible explanations for these decision errors. (JEL M5, D81, J30)     

The Relationship of Practice Beliefs and Practice Behaviors among Social Workers with Lesbian and Gay Clients

This study explores the levels of beliefs about practice and practice behaviors with lesbian and gay clients among social work practitioners in a medical setting. The Gay Affirmative Practice scale was utilized to measure the levels of gay affirmative practice beliefs and practice behaviors among the sample surveyed. The relationship of practice beliefs and practice behaviors is explored and reveals that practice beliefs are significant to practice behaviors with lesbian and gay individuals. The results identify variables that contribute to practice beliefs and practice behaviors.     

Costs,Benefits, and Management Considerations of a Prepaid Student Dental Health Plan

Abstract

The Student Dental Health Program (SDHP) is an open panel, prepaid dental insurance plan at the University of Alabama in Birmingham. The program was developed primarily to provide optimal dental health service; however the SDHP also serves educational and research functions that are consistent with the objectives of the university. The student bodies of the Schools of Dentistry, Medicine, Nursing, and Optometry constitute the enrollee population.

Total cost rose from $18,418 in 1969–70 with 862 enrollees, to $70,930 in 1975–76 with 1,618 enrollees. During the same period the number of procedures completed increased from 2,868 to 6,796, with the average cost per procedure increasing from $6.42 to $10.44. The number of restorative services declined and preventive services increased. In 1976–77 benefits per student treated were estimated at $80 as compared to the annual fee of $48. The value of unmet needs upon entrance to the program averaged $59 per enrollee in 1974–75. By 1976–77 this figure was reduced to $26 due to implementation of an initial screening. The program is self-supporting due to a variety of cost containment measures. On balance, the program can be considered successful and beneficial to all parties concerned.     

Correlation Coefficient Revisited

Elementary approaches to prove basic properties of the correlation coefficient are of pedagogical interest. Besides posing another proof, this article gives variations of the proofs already existing in the statistical literature     

An Improved Boxplot for Univariate Data

The boxplot is an effective data-visualization tool useful in diverse applications and disciplines. Although more sophisticated graphical methods exist, the boxplot remains relevant due to its simplicity, interpretability, and usefulness, even in the age of big data. This article highlights the origins and developments of the boxplot that is now widely viewed as an industry standard as well as its inherent limitations when dealing with data from skewed distributions, particularly when detecting outliers. The proposed Ratio-Skewed boxplot is shown to be practical and suitable for outlier labeling across several parametric distributions.