A rubric is a descriptive scoring (or marking) scheme developed by teachers (sometimes in consultation with students) to guide judgements about the products or processes of students' learning through an assessment task. A rubric is often presented as a type of matrix that provides scaled levels of achievement or performance for a set of criteria or dimensions of quality for a particular assessment task, e.g., a paper, an oral presentation, use of teamwork skills or a work placement. The descriptions of the possible levels of attainment for each of the criteria or dimensions of performance are described fully enough to make them useful for judgment of, or reflection on, progress toward intended learning outcomes. In simpler terms, a rubric is an extended two dimensional checklist of what is expected in an assessment response (criteria) matched against descriptions of how well the work presents each criterion (levels of performance).
Marking checklists may increase the efficiency of marking but they often provide only limited information to promote learning. An example of an annotated marking checklist (PDF 17KB) is available. It includes comments that identify some of the ISSes and limitations in using marking checklists as well providing ideas for potential improvements. Many of the limitations of simple checklists can be overcome through the development of rubrics containing carefully considered and clearly stated descriptions of levels of performance.
Types of rubrics
There are two main types of rubrics - analytical and holistic. The key difference between the two types is the number of discrete criteria addressed in the rubric.
- Analytical rubrics breakdown and disaggregate multiple criteria.
- Holistic rubrics present "lumped together" descriptions of performance to assist broader judgements about the whole assessment process or task.
Some analytical rubrics build a holistic judgement in as one of the criteria. This can accommodate a range of situations, e.g.:
- where the emergent characteristics of the whole of an assessment product may be different (greater or less) to the sum of the component parts
(e.g. where a student teacher being observed in a school placement does not use an expected method of teaching and therefore might be assessed at a lower point on an analytical scale, but nevertheless creates an effective learning experience for the class and therefore could be assessed at a high level of proficiency on an holistic "effectively promotes learning" scale).
- where multiple components of an assessment task contain multiple elements that are being assessed at different levels of aggregation
(e.g. an oral presentation on a particular topic where knowledge of the topic is the main assessment criterion might be assessed using a holistic scale for presentation skills but several analytical scales that assess elements of topic knowledge)
A rubric may be general or task specific i.e. designed to evaluate a category of tasks (e.g. oral presentation skills or laboratory skills) or a particular task (e.g. an oral presentation on the origins of the American War of Independence or laboratory analysis of a soil sample). If the development of a set of broad skills (e.g presentation skills) is an intended outcome from a topic or course then a general rubric for those skills could be used multiple times to help students track the development of their skill set over time. Allen and Tanner (2006) present an example of a full analytical rubric.
‘Generic' rubrics should be viewed with caution and appropriately modified to suit your particular learning context and assessment task. Using and re-using generic rubrics without modification can create significant challenges for those making judgements about a particular assessment piece as generic descriptions of performance need to be translated to the specific context anyway for meaningful judgements to be made. If multiple markers are involved with marking (e.g. with a large class) then there is potential for different translations to be made and for inconsistencies to arise across markers' judgements. Also generic rubrics can only provide generic feedback to students. Ideally a specific rubric should be created for each specific assessment task to help provide useful feedback to students.