Test Equating, Scaling, and Linking

Kolen, Michael J.; Brennan, Robert L.

doi:10.1007/978-1-4939-0317-7

Cited by 589 publications

(947 citation statements)

References 0 publications

Supporting

Mentioning

872

Contrasting

Unclassified

Order By: Relevance

“…Our analytic plan followed the multi-method approach applied in the PROsetta Stone Project and recommended by linking experts. 20 This approach includes methods based on IRT and one commonly used non-IRT method (equipercentile linking). IRT is a family of mathematical models that allow researchers to assign unique values (i.e., parameters) to each item based on how likely people with different levels of the measured construct are to endorse reach response category.…”

Section: Discussionmentioning

confidence: 99%

Establishing a Common Metric for Physical Function: Linking the HAQ-DI and SF-36 PF Subscale to PROMIS® Physical Function

Schalet

Revicki²,

Cook

et al. 2015

J GEN INTERN MED

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Establishing a Common Metric for Physical Function: Linking the HAQ-DI and SF-36 PF Subscale to PROMIS® Physical Function

Schalet

Revicki²,

Cook

et al. 2015

J GEN INTERN MED

View full text Add to dashboard Cite

“…The equipercentile equating method then was used to create the crosswalk between the UCLA and LEAS scores [11]. In simple terms, the crosswalk was created by identifying the scores on both scales that have the same percentile ranks.…”

Section: Methodsmentioning

confidence: 99%

A Crosswalk Between UCLA and Lower Extremity Activity Scales

Ghomrawi

Lee

Herrero

et al. 2017

Clinical Orthopaedics &Amp; Related Research

View full text Add to dashboard Cite

Background The University of California, Los Angeles (UCLA) activity scale and the Lower Extremity Activity Scale (LEAS) are the two most-widely used and rigorously developed scales for assessing activity level in patients having joint replacement. However, the two scales are not convertible, and the level of correlation between the two is not clear. Creating a crosswalk between these scales; that is, a concordance table to convert scores from one scale to the other and vice versa, will help compare results from existing studies using either scale, and pool those results for meta-analyses. It also will facilitate pooling data from multiple registries and data sources. Questions/Purpose To create a crosswalk between the UCLA and the LEAS activity scales for patients having THA or TKA. Methods Preoperative and 2-year postoperative UCLA and LEAS scores for a cohort of patients undergoing primary TKA or THA at the Hospital for Special Surgery between

show abstract

“…It is only necessary that linking items be sufficiently similar to other items that inflation generalizes substantially to the linking items. Given that linking items are typically chosen to be representative of the test as a whole (Kolen & Brennan, 2004), this degree of similarity is likely.…”

Section: Implications For Linkingmentioning

confidence: 99%

“…At the same time, these items could not be too novel, for two reasons. First, linking items should be representative of the test as a whole (e.g., Kolen and Brennan, 2004). Second, if items are too novel, they may be too salient and therefore too memorable to serve as uncorrupted linking items.…”

Section: Possibilities For Changes In Linkingmentioning

confidence: 99%

Adapting Educational Measurement to the Demands of Test-Based Accountability

Koretz¹

2015

Measurement: Interdisciplinary Research and Perspectives

View full text Add to dashboard Cite

Accountability has become a primary function of large-scale testing in the U.S.The pressure on educators to raise scores is vastly greater than it was several decades ago.Research has shown that high-stakes testing can generate behavioral responses that inflate scores, often severely. I argue that because of these responses, using tests for accountability necessitates major changes in the practices of educational measurement.The needed changes span the entire testing endeavor. This paper addresses implications for design, linking, and validation. It offers suggestions about possible new approaches and calls for research evaluating them.Adapting to accountability 1 Over the past several decades, accountability has become a primary functionarguably, the single most important function-of large-scale educational testing in the U.S. The transition has been gradual, dating back at least to the minimum-competency testing movement of the 1970s, and the nature of the accountability systems and the characteristics of the assessments used for this purpose have varied through several waves of policy initiatives (Koretz & Hamilton, 2006). Nonetheless, the pressure on educators to raise scores has increased markedly from one wave of initiatives to the next.The current situation, in which rewards and sanctions for schools based on scores are ubiquitous and consequential evaluations of teachers based on students' scores are in place or planned in a large number of states, represents a tremendous change from the typical uses of tests 40 years ago.The premise of this paper is that the current uses of tests for accountability require major changes to several aspects of educational measurement. The needed changes span the full sequence of measurement practices, beginning with test design, continuing with the activities needed to maintain testing programs, in particular, linking, and ending with validation.

show abstract

Test Equating, Scaling, and Linking

Cited by 589 publications

References 0 publications

Establishing a Common Metric for Physical Function: Linking the HAQ-DI and SF-36 PF Subscale to PROMIS® Physical Function

Establishing a Common Metric for Physical Function: Linking the HAQ-DI and SF-36 PF Subscale to PROMIS® Physical Function

A Crosswalk Between UCLA and Lower Extremity Activity Scales

Adapting Educational Measurement to the Demands of Test-Based Accountability

Contact Info

Product

Resources

About