A review of approaches to assessing writing at the end of primary education
Download 0.91 Mb. Pdf ko'rish
|
International primary writing review - FINAL 28.03.2019
Validity of outcomes in relation to assessment purpose and construct
The extent to which an assessment’s outcomes need to reflect the full breadth of the construct may help drive the choice between different methods of marking, grading, or judging. For example, secure-fit models can be used to ensure that assessors evaluate students’ writing across the entire breadth of the assessment construct A review of approaches to assessing writing at the end of primary education 28 because assessment objectives are clearly defined, and pupils must demonstrate proficiency in all criteria to achieve each standard. However, this approach can be quite restrictive and time-consuming (see Section 2). In addition, if pupils are meeting (or exceeding) expectations in the majority of areas, but fail to meet expectations in 1 area, they will receive the lower grade (as previously discussed by Cresswell & Houston, 1991). Outcomes from this approach therefore might not always provide a valid reflection of general levels of performance, which might give rise to tensions when outcomes are at odds with teachers’ professional judgements (see Section 2). Best-fit approaches are less strict in their focus on every individual criterion, thus can avoid this particular issue, but they do not guarantee such an all- encompassing coverage of assessment objects as secure-fit models. Comparative judgement approaches produce an overall rank-order of pupils without the need for potentially complicated marking criteria which may not be consistently interpreted. However, while the assessment construct can be defined, this method relies on assessors’ holistic understanding of what different levels of performance against that construct look like. In other words, instructions are not given to assessors under this method for how to differentiate between levels of performance in the same way as is done under secure-fit or best-fit models. It is therefore difficult under this approach to know how assessors are making their judgements, and therefore how closely those judgements may or may not reflect the desired depth and breadth of the construct under consideration (eg see van Daal, Lesterhuis, Coertjens, Donche, & De Maeyer, 2019). This could potentially raise concerns over the validity of judgements, and also limit opportunities for feedback. To help with this, assessors could be asked to provide annotated justifications for their decisions, which could be used as part of quality assurance processes. To increase the amount of feedback for pupils and/or schools, annotated exemplars of performance could be produced for various points on the ability scale that this process produces (although this is unlikely to provide the same level of detail on what pupils can do as a clear mark scheme would; Heldsinger & Humphry, 2010, 2013). The intended skill coverage of the assessment will have further implications for the choice between different methods. For example, where the intention is to focus on technical skills via multiple-choice or short-response type items, a points-based mark scheme may be appropriate (where every mark that can be awarded is clearly defined in terms of what the ‘correct’ response is). However, different methods may be better suited for the valid assessment of more subjective skills, such as writing for a purpose or audience. Accordingly, most assessments in the international literature have adopted a best-fit levels-based approach. However, while KS2 assessments in England have historically been assessed via best-fit, a secure-fit model was introduced in 2016, because it was felt that the previously adopted best-fit levels- based approach was too flexible (discussed in Section 2). Comparative judgement offers an innovative alternative for judging extended responses. However, while comparative judgement approaches may be effective at assessing writing in a uni- dimensional manner (ie to produce a single outcome score), multi-dimensional assessment (eg to produce separate scores for technical and compositional type skills) is not possible without multiple rounds of judging. A review of approaches to assessing writing at the end of primary education 29 Download 0.91 Mb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling