Evaluating the Adjacent Levels Model for Differentiated Assessment

Literaturnachweis - Detailanzeige

Trefferliste

Autor/in	Bramley, Tom
Institution	Cambridge Assessment (United Kingdom)
Titel	Evaluating the Adjacent Levels Model for Differentiated Assessment
Quelle	(2014), (20 Seiten) PDF als Volltext Verfügbarkeit
Sprache	englisch
Dokumenttyp	gedruckt; online; Monographie
Schlagwörter	Foreign Countries; Exit Examinations; Alternative Assessment; High Stakes Tests; High School Students; Grades (Scholastic); Grading; Decision Making; Models; Student Evaluation; Test Construction; Test Items; Item Response Theory; United Kingdom (England) + Suchen Sie Ihr Suchwort? Ausland; Final examination; Abschlussprüfung; High school; High schools; Student; Students; Oberschule; Schüler; Schülerin; Studentin; Notenspiegel; Notengebung; Schulnote; Decision-making; Entscheidungsfindung; Analogiemodell; Studentische Bewertung; Testaufbau; Test content; Testaufgabe; Item-Response-Theorie
Abstract	The aim of this study was to compare models of assessment structure for achieving differentiation between examinees of different levels of attainment in the GCSE in England. GCSEs are high-stakes curriculum-based public examinations taken by 16 year olds at the end of compulsory schooling. The context for the work was an intense period of debate among politicians, regulator, examination boards, teachers and public about the appropriate content and assessment structure for the reformed GCSEs announced by the coalition government in 2012. The focus was on the 'adjacent levels' model for differentiation. In this model, papers are targeted at three specific non-overlapping ranges of grades. The majority of examinees enter for a pair of papers at adjacent levels and receive the highest grade achieved. There is no aggregation of marks across papers. In contrast, in the standard GCSE tiered model examinees take a pair of papers either at foundation tier or at higher tier. The range of grades available overlaps, with C, D and E available on both tiers. A similar tiering model could be applied to the structure with three adjacent papers described above, if the marks on each paper were added together and grade boundaries applied to the aggregate scale, with overlapping grades available as in the standard tiered model. The adjacent levels model has two important differences from the tiered models: i) there is only one route to each grade; and ii) there is no compensation across papers -- the results on one paper are effectively discounted in grading. A review of tiering methods by Baird et al. (2001) discussed the strengths and weaknesses of the model and concluded it was worthy of further consideration. This study used simulation to compare the three models of differentiation described above in terms of: i) suitability of grade boundary locations; ii) raw score distributions; and iii) reliability (in particular classification accuracy). The simulations were based on an item bank constructed from a mathematics GCSE taken in June 2012. Three papers at adjacent levels of difficulty were constructed by selecting easy, medium and difficult questions from the bank. The results showed that using an adjacent levels model instead of a tiered model would be likely to reduce reliability in psychometric terms. However, two aspects of validity could increase. The lowest grade boundaries on each paper in the adjacent levels model were at a higher proportion of the paper total than those on the standard model, implying that knowledge of the content of the question paper would give a better indication of what examinees with a given grade would know and be able to do. Furthermore, the single route to each grade removes the ambiguity about overlapping grades inherent in the current system -- whether they imply reasonable achievement on a restricted domain of knowledge/skills, or lesser achievement on the full domain. The discussion considers further the various trade-offs that must be made in choosing a model for differentiated assessment that best serves the interests of the different stakeholders in the system. (As Provided).
Anmerkungen	University of Cambridge Local Examinations Syndicate (Cambridge Assessment). The Triangle Building, Shaftesbury Road, Cambridge, CB2 8EA, UK. Tel: +44-1223-55331; Fax: +44-1223-460278; e-mail: info@cambridgeassessment.org.uk; Web site: https://www.cambridgeassessment.org.uk/
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2024/1/01