I: Theoretical and Conceptual Foundations of Learning-Oriented Assessment

Theoretical and Conceptual Foundations of Learning- Oriented Assessment

2 Learning-Oriented Assessment

Learning-Oriented Assessment: Basic Concepts and Frameworks in Using Assessment to Support Language Learning

Nick Saville


In this chapter, I explore the concept of learning-oriented assessment (LOA) when applied to language learning in particular. I suggest a number of common dimensions that are needed in developing a comprehensive, theoretical LOA framework. The challenge is to link all types of language assessment into a coherent model that values learning processes and learning outcomes. In doing so, I consider the origins of the LOA concept, including the ways in which the terminology has been used by different scholars, and how it relates to other assessment concepts that seek to integrate assessment and learning. Recent developments of LOA are placed within a wider historical context, and a number of framework proposals that have emerged in different parts of the world, including the UK, the USA, Hong Kong, and Australia since the 1990s, are reviewed. In providing this perspective, a case is made for a systemic model of LOA - one that provides a coherent and comprehensive framework to language education and seeks to promote better learning across the lifespan. That is, starting in formal education and extending beyond that as new opportunities for learning arise in the course of normal life (in work, travel, leisure activities, sport, etc.). This requires enhanced ways to integrate assessment into all contexts of learning and goes beyond the design of the assessment instruments and procedures themselves.

While the systemic model prioritizes learning, at the same time it needs to meet high-level policy goals that address the issue of educational standards. The LOA framework, therefore, needs to incorporate the sociopolitical considerations that have led to large-scale testing regimes and the adoption of high-stakes, external examinations within educational reform programs at a national or regional level. In other words, the systemic model should enable coherent links between the macro-level policies and the micro-level educational contexts where the teaching and learning take place. Ultimately the aim should be to bring about improved teaching and learning practices in classrooms and other language learning environments, and to avoid the possible negative impacts that can occur when learning and assessment practices are divergent.

The positive alignment of various kinds of assessment to support learning has hitherto proved challenging, and in concluding this chapter, I highlight some ways in which digital technology can provide transformative opportunities to create successful ecosystems of learning. In achieving this, it is anticipated that language teachers will need to engage with the relevant technology and to develop higher levels of language assessment literacy in using the new forms of data that can be generated. In the future, these concerns will become essential components in teacher education and professional development.

Learning-Oriented Assessment (LOA)

LOA is one of several terms used in recent years to identify a shared concern: to promote better learning. Other terms that are in use include teacher-based assessment (TBA), classroom-based assessment (CBA), alternative assessment, assessment for learning, and dynamic assessment. There is, however, no single, coherent LOA movement and the various approaches differ according to their precise motivation, reflecting the geographical and educational context in which they arose. So, while much of the foundational literature on the subject is shared, a review of the different contexts where research has been conducted reveals both similarities and differences in approach.

Turner (2012), for example, points out that the various terms have also been used with different meanings by different scholars, so in this chapter, I attempt to be consistent in the use of some key words. Large-scale assessment is used for external examinations of the kind produced by examination boards and testing agencies. For approaches that prioritize learning, I use the adjective “learning-oriented.” Where specific movements have adopted different terminology, I use that.

A Historical Perspective: The Impact of Large-Scale Examinations on Learning

Contemporary assessment practices in schooling have evolved over the past 150 years since their origins in the Victorian era. From the beginning, emerging educational systems made explicit links between learning goals and the assessment of learning using external examinations. In England, public examinations had their origins in the schools themselves, where there was a perceived need to set educational standards using external evaluation of pupils’ achievement. The foundation of the University Examination Boards, the Oxford Delegacy, and the University of Cambridge Local Examinations Syndicate (UCLES) in the 1850s was a response to this need (Raban, 2008; Roach, 1971). The “local examinations” were thought to extend educational opportunities by providing external assessment locally for pupils in their own schools. The aim was to support learning, to certificate achievements based on what had been taught, and to accredit educational standards for societal purposes, e.g., to select applicants to go on to university.

There was a requirement from the start for the examinations to be learning-oriented in a general sense, and in this respect, the original conception of school examinations might be considered an early example of “backward design,” with the intentional alignment of the learning objectives in the curriculum to the assessment of outcomes in the examinations. As Boud (2000) points out, assessment has traditionally performed this “double duty” - it is about “grading and about learning” and has to handle “the tensions and compromises” that have to be dealt with to achieve this.

The effects of the examinations on teaching and learning continued to be a concern for the UK examination boards in the 20th century, as noted by Spolsky (2004, p. 305). He describes how:

from its beginning UCLES accepted the key role to be played in test development by the “stakeholders” in particular those schools in various countries of the world that wished to establish examination centres, mainly for their own students. From the earliest years, the Cambridge test writers and their various committees saw themselves as sharing with the schools not so much an examination as the culmination of a teaching process.

He also notes that before “the word backwash had been coined,” the governance committees of UCLES regularly queried whether modifications being proposed would be accepted by the schools.

By the end of the 20th century, various forms of educational assessment had emerged around the world, linked to diverse educational policies and practices at a regional or national level (Eckstein 8c Noah, 1993). From the 1920s onwards, the emerging field of psychometrics had introduced the concepts of validity and reliability that became increasingly important in the design and delivery of standardized educational assessments, especially in the USA. In reporting on the Cambridge-TOEFL Comparability Study, Bachman et al. (1995) observed the differences between the British and US traditions with respect to psychometric features of the testing systems. Elsewhere, school-based assessments persisted with little reference to psychological measurement, as in many parts of Europe (Newton & Shaw, 2014).

How and where the assessment takes place, and who has responsibility for the administration and the outcomes, are key issues that have led to significantly different policies and practices (Vladingerbroek &

Taylor, 2009). Whichever stance is adopted, however, it is now widely believed that all kinds of assessments have the potential for unintended or unplanned side-effects and consequences, including negative impacts on teaching and learning (noted by Latham as early as the 1870s; Roach, 1971). Eckstein and Noah (1993), for example, outline eight persistent “dilemmas of examination policy” that examination authorities have to live with. Each of these “represents a potential problem associated with use of the examination system to control either individual destinies or what occurs in schools” (p. 244).

Some commentators argue that the balance has tilted too far away from learners and their learning goals and towards policy-related processes and the summative use of results for accountability purposes (see the discussion below for a UK perspective). In response to the perceived negative aspects of large- scale assessment regimes, alternative solutions have emerged over the past 30 years, seeking to identify learning-oriented features that differ from this trend. For example, Davison and Leung (2009) suggest that LOAs “signify a more teacher-mediated, context-based, classroom-embedded assessment practice,” that is “explicitly or implicitly defined in opposition to traditional externally set and assessed large-scale formal examinations” (p. 395).

Turner (2012), in her review of CBA, sees it as a separate field of activity that is distinct from large-scale assessment and suggests that it requires a different conceptualization of the fundamental concepts of validity and reliability. She highlights a number of recurrent themes in the literature related to learning orientation, including the following areas: assessment criteria within a socio-cultural framework; learning and assessment tasks in classroom contexts; the role of feedback and reflection; the roles of participants, especially the teachers and learners themselves; the alignment between different assessment types (internal/external); and the importance of teacher education with a need for greater “assessment literacy.” These points are discussed below in relation to the models of LOA that have emerged.

< Prev   CONTENTS   Source   Next >