Observed-Score Equating With a Heterogeneous Target Population

The Use of Quality Control and Data Mining Techniques for Monitoring Scaled Scores: An Overview

“…This perspective follows the recommendations of Dorans (2002), who described the scale of the reported scores as an infrastructure of the test, which as with any infrastructure, needs careful and regular maintenance. Consequently, the need to monitor and maintain the stability of the scale score leads to new approaches for control and adjustment, such as tightening the control of the linking parameters, adjusting the samples to match a target distribution, and investigating relevant equating subgroups in order to obtain a better equating procedure (for details, see the papers of Duong & von Davier, 2012;Haberman & Yang, 2011;Qian et al, 2012).…”

“…Defining a target population for complex educational assessments and creating appropriate weights for its subgroups is an area that needs further research. The studies conducted by Qian et al (2012) and Duong and von Davier (2012) are preliminary attempts to address these questions.…”

“…If the test scores and the relevant demographic variables are continuously monitored, then a continuous and minimal adjustment can be implemented as part of the measurement process (see Duong & von Davier, 2012;Haberman & Yang, 2011;Qian, von Davier, & Jiang, 2012). In addition to the shifts due to demographics, operational mistakes, such as the computing or reporting of an incorrect score (one that is lower or higher than the correct score), may have serious implications on the test validity in the context of educational measurement.…”

Weighting Test Samples in Irt Linking and Equating: Toward an Improved Sampling Design for Complex Equating

“…As pointed out by Kolen (2004), most of these studies are sample relevant because linkings and equatings are data dependent. In some studies, the equating sample was matched to a target population (Duong & von Davier, 2012;9 T Qian, 209 T 129 T ; 9 T von Davier, Holland, & Thayer, 2004). Our approach of weighting has similarities to the methods described in the equating literature.…”

Contributions to Score Linking Theory and Practice

“…The ±2SEED band is narrower for larger sample sizes and wider for smaller sample sizes. Duong and von Davier (2012) illustrated the flexibility of the observed-score equating framework and the availability of the SEED in allowing practitioners to compare statistically the equating results from different weighting schemes for distinctive subgroups of the target population.…”