Detection of multiple change-points in multivariate data

The statistical analysis of change-point detection and estimation has received much attention recently. A time point such that observations follow a certain statistical distribution up to that point and a different distribution – commonly of the same functional form but different parameters after that point – is called a change-point. Multiple change-point problems arise when we have more than one change-point. This paper develops a method for multivariate normally distributed data to detect change-points and estimate within-segment parameters using maximum likelihood estimation.     

CONSER Updates: The Serials Cataloging Cooperative Training Program at Ten Years and Revision of Integrating Resources: A Cataloging Manual

Abstract

In this column, CONSER specialist Hien Nguyen recognizes the ten-year anniversary of the Serials Cataloging Cooperative Training Program. She discusses program successes, including the development of new material for Web delivery. CONSER coordinator Les Hawkins gives an update on the revision of Integrating Resources: A Cataloging Manual, the basic documentation used by CONSER and BIBCO members to create and maintain records for integrating resources.     

A simple least squares method for estimating a change in mean

A simple least squares method for estimating a change in mean of a sequence of independent random variables is studied. The method first tests for a change in mean based on the regression principle of constrained and unconstrained sums of squares. Conditionally on a decision by this test that a change has occurred, least squares estimates are used to estimate the change point, the initial mean level (prior to the change point) and the change itself. The estimates of the initial level and change are functions of the change point estimate. All estimates are shown to be consistent, and those for the initial level and change are shown to be asymptotically jointly normal. The method performs well for moderately large shifts (one standard deviation or more), but the estimates of the initial level and change are biased in a predictable way for small shifts. The large sample theory is helpful in understanding this problem. The asymptotic distribution of the change point estimator is obtained for local shifts in mean, but the case of non-local shifts appears analytically intractable.     

Influence of a Community Health Worker Intervention on Mental Health Outcomes among Low-Income Latino and African American Adults with Type 2 Diabetes

This study investigated the influence of a community health worker (CHW) diabetes lifestyle intervention on mental health outcomes. Our study was guided by the principles of community-based participatory research. Data were collected from 164 African American (N = 94) and Hispanic adults (N = 70) participating in a randomized, 6-month delayed intervention group design for improving glycemic control. The intervention time periods were baseline to 6 months for the treatment group and 6–12 months for the delayed group. Linear mixed models were used to conduct longitudinal analyses of the Problem Areas in Diabetes (PAID) and Patient Health Questionnaire (PHQ) scales. In the model adjusted for demographics, the PAID dropped significantly from pre-intervention to post-intervention within both the treatment and delayed groups (p < 0.05) with an average intervention effect of ?6.4 (p < 0.01). The PAID dropped even further within the immediate group from 6 to 12 months. Although the PHQ did not change significantly, the PHQ-2 had an average intervention effect of ?0.3 (p < 0.05) in the model adjusted for demographics. This study contributes to an understanding of how a CHW-led diabetes intervention can result in positive mental health outcomes for Latinos and African Americans with Type 2 diabetes. It also highlights the importance of further exploring what factors may contribute to racial/ethnic variation in mental health outcomes for African Americans and Latinos with diabetes and the role CHWs can play.     

A cross‐cultural test of the work‐family interface in 48 countries

This study tests a cross‐cultural model of the work‐family interface. Using multigroup structural equation modeling with IBM survey responses from 48 countries (N= 25,380), results show that the same work‐family interface model that fits the data globally also fits the data in a four‐group model composed of culturally related groups of countries, as well as a two‐group gender model. This supports a transportable rather than a culturally specific or gender‐specific work‐family interface model: notably, job flexibility related to reduced work‐family conflict, reduced family‐work conflict, and enhanced work‐family fit. Work‐family fit related to increased job satisfaction. Findings suggest that investment by multinational companies in job flexibility initiatives may represent a dual‐agenda way to benefit men, women, and businesses in diverse cultures.     

Electronic Journal Forum: Defining an “Access Level” Catalog Record Using MARC 21 and AACR2

Abstract

The Library of Congress (LC) is in the process of developing a new level of MARC 21 and AACR2 cataloging for non-serial Internet resources called “access” level. This article briefly describes the impetus behind the creation of this new standard, information about the proposed standard itself, and the results of a test conducted at LC using the core data set and cataloging guidelines. The future plans of the Library for implementing and possibly expanding the use of access level are identified.     

A power comparison of three tests for design effects in a random effects covariance model

A preliminary testing procedure for design ettecta in a ran-dom effects covariance model is Compared with the usual procedure to see if the power of the latter can be improved. A procedure which ignores the random covariate effects is included for comparison and for study of misspecification effects. Methodology is based on Roebruck's (1982) results for regular linear models.