Skip to main content

Educator, applied linguist, language tester.

Alphabet soup: AfL, AaL, LOA

2 min read

Last week, my post on formative assessment (and a subsequent tweet asking for suggestions) sparked a short conversation on Twitter with @ashley about Assessment for Learning and Assessment as Learning, as well as Learning Oriented Assessment. I'm still looking for suggestions for this blog (let me know!); in the meantime, here's my attempt at sorting out these concepts.

Assessment for Learning (AfL) is for all intents and purposes formative assessment. It's useful here to revisit Dylan Wiliam @dylanwiliam's table:

Assessment as Learning was originally proposed by Lorna Earl @lmearl. While often differentiated from AfL, if we accept Wiliam's definition of AfL, AaL is more accurately a subset of AfL:

Learning Oriented Assessment is the 'new' kid on the assessment block:

Figure from
Carless (2007)

Originally proposed by David Carless @carlessdavid and his colleagues, the concept should ring a bell for those of you who are familiar with the backward design approach to curriculum. This approach includes Understanding by Design (Wiggins @grantwiggins & McTighe @jaymctighe), popular in K-12:

(taken from here; original source unknown)

And also Biggs's Constructive Alignment (well-known in HE):

Diagram by UCD Teaching & Learning

I see LOA as a model that not only employs backward design, but does it in a way that foregrounds formative assessment (including AaL). It also deemphasises the distinction between summative and formative assessment in a way that might actually be constructive -- the key is to make summative assessment perform a learning-oriented service, in addition to institutional purposes. I say constructive because seeing the two assessments as a dichotomy (mutually exclusive) could put teachers and learners in a bind -- we can't do away with summative assessments because of institutional demands, and positioning them as the 'bad guys' doesn't necessarily eliminate washback. IMO, the distinction between formative and summative is still important, but the gap can be narrowed, and an assessment could be thoughtfully designed to serve both purposes, perhaps especially if it is an 'alternative' assessment rather than a traditional timed test. By aligning all assessments with the LOs, we can ideally ensure that both kinds -- summative and formative -- are pulling stakeholders in the same direction rather than opposing ones, and promote positive washback.

I've really only just started thinking about these concepts (and what they mean in relation to my own research), so any thoughts you might have on this are very welcome :)