Analytical assessment rubrics to facilitate semi-automated Essay grading and feedback provision

Assessment is an essential part of the learning process, both in formative learning settings and traditional summative assessment. Both types are challenging, as it can be difficult to ensure consistency, reliability and absence of bias. In formative assessment the problem of workload and timely results is even greater, as the task is carried out more frequently. Information technology is able to assist teachers in these challenges to various degrees depending on the type of test items. The essay test item, besides the well-known application to train language skills and to acquire foreign languages, is widely used to test higher order thinking skills and therefore it can be applied in a great variety of subject domains at different educational levels. Evaluating essays is a time-consuming task hence supporting technologies can deliver great advantages. In this paper we introduce a semi-automated approach to essay grading based on analytical assessment rubrics, the use of which facilitate feedback provision. A prototype system is described in terms of requirements derived from the authors' own experience and the published literature, a workflow model. Reflection of the development experience and user feedback informs further development of the system.