Test Equating

Kolen, Michael J.; Brennan, Robert L.

doi:10.1007/978-1-4757-2412-7

Cited by 304 publications

(158 citation statements)

References 0 publications

Supporting

Mentioning

151

Contrasting

Unclassified

Order By: Relevance

“…We again used the SAS program EQUIPERCENTILE (Price et al, 2001), a realization of the algorithms for equipercentile linking described by Kolen and Brennan (1995) in order to compare the BPRS/PANSS with the CGI. For further details of the methodology used, please consult Leucht et al (2005a, b).…”

Section: Discussionmentioning

confidence: 99%

Linking the PANSS, BPRS, and CGI: Clinical Implications

Leucht

Kane

Etschel

et al. 2006

Neuropsychopharmacol

259

172

View full text Add to dashboard Cite

To understand what given scores of the Brief Psychiatric Rating Scale (BPRS) and the Positive and Negative Syndrome Scale (PANSS) mean from a clinical point of view is important for the translation of research results into practice. We therefore (a) compared the absolute change of the BPRS/PANSS with the Clinical Global Impressions Ratings (CGI) -improvement score and the change of the CGI severity score, (b) analyzed whether the severity of illness at baseline had an impact on the latter association, and (c) attempted to replicate previous BPRS findings using a completely different data set based upon the PANSS-derived BPRS. The method used was equipercentile linking of BPRS and CGI ratings from 14 drug trials in acutely ill patients with schizophrenia (n ¼ 5970). An absolute reduction of the BPRS/PANSS by approximately 10/15 points corresponded to a CGI change of 'minimally improved' and to a change of the CGI severity score by one severity step. However, the latter associations depended on the severity of symptoms at baseline. Less severely ill patients required less BPRS/PANSS total score reduction to achieve the same CGI-improvement score than more severely ill patients. This effect of initial severity was attenuated using percentage rather than absolute BPRS/PANSS reduction scores.The linking analysis between the absolute BPRS/PANSS reduction and the CGI may have an implication for the interpretation of efficacy differences found in clinical trials, and for sample size estimations. Clinicians seem to base CGI ratings on relative change rather than on absolute change of symptoms.

show abstract

Section: Discussionmentioning

confidence: 99%

Linking the PANSS, BPRS, and CGI: Clinical Implications

Leucht

Kane

Etschel

et al. 2006

Neuropsychopharmacol

259

172

View full text Add to dashboard Cite

show abstract

“…Guidelines have been suggested for practitioners to determine appropriate sample sizes and anchor sizes in real world testing situations (Kolen & Brennan, 2004). Studies have been conducted to explore how equating results would be impacted by such factors as anchor sizes (Klein & Kolen, 1985;Ricker & von Davier, 2007), anchor items' content representation of the test (Klein & Jarjoura, 1985), or the combined effects of anchor size, content representation, and some other factors, such as anchor difficulty relative to test difficulty, as well as the distribution of anchor item difficulty (Liao, 2009;Sinharay & Holland, 2007).…”

Section: Overviewmentioning

confidence: 99%

Exploring Alternative Test Form Linking Designs With Modified Equating Sample Size and Anchor Test Length

Wang

Qian

Lee

2013

ETS Research Report Series

View full text Add to dashboard Cite

Since its 1947 founding, ETS has conducted and disseminated scientific research to support its products and services, and to advance the measurement and education fields. In keeping with these goals, ETS is committed to making its research freely available to the professional community and to the general public. Published accounts of ETS research, including papers in the ETS Research Report series, undergo a formal peer-review process by ETS staff to ensure that they meet established scientific and professional standards. All such ETS-conducted peer reviews are in addition to any reviews that outside organizations may provide as part of their own publication processes. Peer review notwithstanding, the positions expressed in the ETS Research Report series and other published accounts of ETS research are those of the authors and not necessarily those of the Officers

show abstract

“…The estimation that involves these parameters in the item response theory is known as the 3PL Model. On the other hand, several methods that can be implemented in order to perform equating are namely mean and mean method, mean and sigma method, and item characteristics curve method which includes the Stocking and Lord method [9,10].…”

Section: Introductionmentioning

confidence: 99%

“…This means that there has been relationship between the measurement scales from both tests. Thereby, if the scale of test 1 is equated to that of test 2 for the 3PL Model then the relationship between the item parameter and the participants' abilities for both scales can be stated as follows [10]. q is the abilty of i participant on the test 1 scale after having been equated to test 2.…”

Section: Introductionmentioning

confidence: 99%

“…q is the abilty of i participant on the test 1 scale after having been equated to test 2. The c parameter is not transformed because its value does not depend on the metric q or, in other words, the parameter c is free from the scale transformation [10]. Moreover, the constant that links the score of test 1 and that of test 2 can be calculated using multiple methods known as test score linking method.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The Equating of Battery Test Packages of Mathematics National Examination 2013-2016

et al. 2018

View full text Add to dashboard Cite

Abstract. When a test implements several test instruments, there is an assurance that the test instruments that will be implemented are equal and this equality is an urgent matter. Therefore, this study aimed at confirming the equality of the test instruments that had been implemented in Mathematics National Examinations (MNE) 2013-2016. This study was conducted using quantitative approach. The data were gathered using students' response documentation technique and MNE test instruments; the test instruments were drawn from the packages that had been administered to the national examinations for junior high school students in the Province of Yogyakarta Special Region. The data were analyzed by using the stages of item parameter estimation, designing the equating equation, equating through concordance model, drawing the test characteristics curve, and interpretation. The results of the analysis showed that the instruments that have been administered are almost equal, both from one package to another and from one year to another, with the standards of test instrument 2013.

show abstract

Test Equating

Abstract: Test equating : methods and practices / Michael J. Kolen, Robert L. Brennan. p. cm.-(Springer series in statistics) Includes bibliographical references.

Cited by 304 publications

References 0 publications

Linking the PANSS, BPRS, and CGI: Clinical Implications

Linking the PANSS, BPRS, and CGI: Clinical Implications

Exploring Alternative Test Form Linking Designs With Modified Equating Sample Size and Anchor Test Length

The Equating of Battery Test Packages of Mathematics National Examination 2013-2016

Contact Info

Product

Resources

About