Yeow Meng Thum

Improving Relationships: Mechanisms of Change in Couple Therapy.

In a sample of 134 married couples randomly assigned to traditional or integrative behavioral couple therapy (TBCT vs. IBCT), a multivariate hierarchical growth curve analysis using latent variable regression revealed that measures of communication, behavior frequency, and emotional acceptance acted as mechanisms of change. TBCT led to greater changes in frequency of targeted behavior early in therapy, whereas IBCT led to greater changes in acceptance of targeted behavior both early and late in therapy. In addition, change in behavioral frequency was strongly related to improvements in satisfaction early in therapy; however, in the 2nd half of therapy, emotional acceptance was more strongly related to changes in satisfaction. Research and clinical implications are discussed.

Examining Relationships Between Where Students Start and how Rapidly they Progress: Using New Developments in Growth Modeling to Gain Insight into the Distribution of Achievement Within Schools

Studying change in student achievement is of central importance in numerous areas of educational research, including efforts to monitor school performance, investigations of the effects of educational interventions over time, and school effects studies focusing on how differences in school policies and practices relate to differences in student progress. In this article, we argue that in studying patterns of change, it is often important to consider the relationship between where students start (i.e., their initial status) and how rapidly they progress (i.e., their rates of change). Drawing on recent advances in growth modeling methodology, we illustrate the potential value of such an approach in the context of monitoring school performance. In particular, we highlight the ways in which attending to initial status in analyses of student progress can help draw attention to possible concerns regarding the distribution of achievement within schools. To convey the logic of our approach and illustrate various analysis possibilities, we fit a series of growth models to the time series data for students in several schools in the Longitudinal Study of American Youth (LSAY) sample. In a final section, we discuss some of the possibilities that arise in employing a modeling approach of this kind in evaluating educational programs and in conducting school effects research.

Measuring Progress Toward a Goal: Estimating Teacher Productivity Using a Multivariate Multilevel Model for Value-Added Analysis

This article develops a procedure for measuring how much is gained by students in a pretest and posttest situation against a target score on the posttest. The author defines a productivity index, Mj, for teacher j as the ratio of estimated gains to an estimated standard that is the distance between an estimate of the pretest and target score. Using language, mathematics, and reading scores on the SAT 9 for 1999 and 2000 from 75 public elementary classrooms (Grades 3-6 in 2000), the author employs a Bayesian implementation of a multivariate mixed model for repeated test scores from individual students. The analysis points to statistically significant gains on the whole for Grades 3, 4, and 6. The strength of the approach lies in a straightforward estimation of the productivity index and a procedure for representing its uncertainty in the form of a productivity profile. This approach also facilitates a Bayesian effect size analysis free from frequentist appeals to noncentral t or F distributions.

Assessing School Academic Productivity: The Case of Chicago School Reform

This paper considers the issues raised in using standardized achievement test scores for purposes of examining the academic productivity of schools. We critique some commonly used practices by urban school districts and suggest an alternative approach – the school productivity profile. This profile is based on an assessment of each schools contribution to student learning, or value-added, rather than just the overall level of student attainment. We illustrate, using 10-year achievement trend data from the Chicago Public Schools (CPS), both the problems with some commonly reported indicators of school effectiveness and the idea of a school productivity profile. While our analyses suggest broad-based improvements in student learning in many Chicago schools over the past 10 years, we also found that the current testing system is not well designed to make such judgments accurately. We conclude that extant standardized testing systems, like the ITBS used in Chicago, do not afford an accurate basis for assessing school productivity and how this might be changing over time. These results have important policy implications. As school districts seek to become more “outcome oriented,” they will need to invest in better testing and reporting systems in order to know whether they are making genuine progress in this regard.

Detecting a Change in School Performance: A Bayesian Analysis for a Multilevel Join Point Problem:

A substantial literature on switches in linear regression functions considers situations in which the regression function is discontinuous at an unknown value of the regressor, Xk , where k is the so-called unknown “change point.” The regression model is thus a two-phase composite of yi ∼ N(β01 + β11xi, σ1 2), i=1, 2,..., k and yi ∼ N(β02 + β12xi, σ2 2), i= k + 1, k + 2,..., n. Solutions to this single series problem are considerably more complex when we consider a wrinkle frequently encountered in evaluation studies of system interventions, in that a system typically comprises multiple members (j = 1, 2, . . . , m) and that members of the system cannot all be expected to change synchronously. For example, schools differ not only in whether a program, implemented system-wide, improves their students’ test scores, but depending on the resources already in place, schools may also differ in when they start to show effects of the program. If ignored, heterogeneity among schools in when the program takes initial effect undermines any program evaluation that assumes that change points are known and that they are the same for all schools. To describe individual behavior within a system better, and using a sample of longitudinal test scores from a large urban school system, we consider hierarchical Bayes estimation of a multilevel linear regression model in which each individual regression slope of test score on time switches at some unknown point in time, kj. We further explore additional results employing models that accommodate case weights and shorter time series.