Assessment: reports from the ATN Conference (II)
Authentic Assessment of Authentic Tasks
ATN Assessment Conference Keynote; Jan Herrington
Opens with the maxim “We assess what we value and we value what we assess”. Uses it to show how assessing time online, numbers of posts to forums, doing MCQs values lower-order knowledge repetition, the time spent online, and quantity of participation.
Cites Angelo “educative assessment tasks” – that should be the focus of our attention. Anything which is ‘to do’ – the task – that matters most. Tasks and assessment are inseparable.
Reprises the classic ‘from this to that’ movement for online learning – eg from instructivist to constructivist; individual to collaborative. Suggests that there are now further moves towards connectivism, Web 2.0, and so on.
Herrington outlines 9 elements of authentic learning:
- Authentic Context, reflecting the way something will be done in the real world; embracing the complexity of real world; provides purpose and motivation of learning;
- Authentic tasks, which have real-world relevance; may take the whole semester; complex and ill-defined (because time has to be taken to learn what the problem is);
- Expert performance, which Herrington linked to Web 2.0 – the crowd has the expertise; the expert knows more than you and can mentor – not necessarily an acknowledged expert; Herrington justifies the lecture when it is expert performance, not because it is transmission;
- Multiple perspectives
- Collaboration, including joint problem-solving and social support;
- Articulation, where online learning can be silent, it becomes authentic as a learning environment where people speak to their learning, present publicly and defend their positions;
- reflection, that gives opportunity to consider post-facto the choices made by students (reflection in action and on action differ);
- scaffolding and coaching role, which to be authentic can include others;
- authentic assessment, which through the Internet, provides public audience and thus motivates people to work harder at preparing for that greater exposure – the product at the end of the assessment has to be authentic beyond being just something for assessment.
Herrington gives an example of a virtual environment involving interviews, artefacts which mimic the real world, a screen design that cues people to the ‘reality’ in a manner not unlike a game. Emphasises that authenticity is not linked necessarily to high-resource, intensively produced environments.
More on authentic assessment: list of either ors which does beg the question of whether authentic is serving as a euphemism for ‘better’.
- Context factors: fidelity / transferability to world beyond classroom
- Student factors: production of knowledge; problem solving; collaboration with others; conversation; performance of knowledge
- Task factors: ill-structured challenges; wide range of responses; assessment integrated with the activity
Some problems with authentic assessment:
- minimum number of assessments challenges authenticity
- restrictions on group work or amount of grades for group work
- invigilation requirements
Herrington emphasises the alignment of assessment to the task – e.g. don’t use exam to assess authentic task. Instead, assess work from the perspective of the ‘assessor’ in the real world – role play the role from the real-world context. Also emphasises the need for scenarios to be consistent with what students expect and understand, to prevent the cognitive challenge being to ‘accept’ the authenticity. Cites Savery and Duffy: problem must be real – but challenges it to argue that students can manage with moderately real, or believably real.
Herrington concludes with a substantial example (Virtual Records) which dates from some year ago but shows many of the key elements – combining relatively artificial representations of reality, but with astute – quite dramatic / produced – cues which turn them into reality for students once inside the system. A key point which emerges to guide us: Alesio’s (1998) concept of cognitive realism.
A critical commentary
It is clear from Herrington’s presentation that the term authenticity is growing in its reach to serve as a synonym for multi-dimensional, high-quality learning experiences. Within her 9 elements, it is true that all are necessary and that, indeed, all can be linked to the world outside of the artificial confines of ‘study’. Yet, to some extent, this extension lowers the precision of application of authenticity by essentially saying all study is just like not-study; in fact it is not…authenticity at moments, even quite extended moments, is vital – but the artificiality is also important.At times the assumptions – that authenticity involves, say, collaboration – really expose the partial definition of authenticity. What, for example, if the task is (in the real world) individual and not collaborative? Would it not be authentic to limit collaboration? Developing this point to consider Web 2.0, one of the challenges for the emphasis on collaboration is that, to be honest, Web 2.0 is also about networked individualism and thus, to be authentic, different forms of collaboration would be required.
One of the examples Herrington gives is a ‘virtual environment’ which is authentic in her judgment. The primary markers of authenticity here are visual display (imagines of labs in this case), interactive navigability (as if they were inside the images) and so on. While clearly different to ‘read this about…’, its authenticity might also be seen as quite limited; that said, what makes it authentic is the assessment task which is very much ‘real world’. Thus authenticity is a complex – made up of more and less real components whose totality allows students to believe themselves to be ‘in the real world’ even when they are not. Indeed, belief, not actuality, matters – else the mistakes they migh make would not be possible and it is through mistakes that we learn,
Herrington’s examples – for example from educational research units / courses – show that some kinds of content / learning outcomes do suit better the authenticity approach, either because of the cohort or similar factors, or because making a viable realistic scenario or similar works easily; perhaps, because there is alignment between what the academics ‘know how to do’ (eg research methods) and what is being taught. Therefore they can model their practice. In some other areas, perhaps, the learning concerns broader and more shallow material – not ‘inhabited’ by the teacher. Of course, authenticity can still be generated, but perhaps takes more work, or perhaps requires a focus on that particular element of being at university which makes it most especially ‘artificial’ and avoid that aspect, so as to reduce the ‘unreality’ overload.