Assessing Speaking, Sari Luoma, CUP, 2004


Image source: Pixabay

Summary notes

This is another text that I would class as an ‘if-you-have-time’ text or an MA text on discourse rather than an essential text for Delta purposes. Therefore, I’m not going to provide all of the notes I made on this text, but just cover a few points that could be useful. The book is part of a series with titles commencing with ‘Assessing …’. I have only read Assessing Speaking so far.

The book covers important research in the areas of speaking, and flags up the influence of people such as Bygate, Hasselgren, and Hymes through examples of how test design has been shaped by them.

A couple of interesting items that I learned included the following:

Speaking tests conducted in pairs of students have had the concern that one speaker will heavily influence the other speaker, but Luoma lists several researchers who have concluded that the influence is negligible on overall results.

Luoma also mentions Hasselgren (1998) and Towell (1996), in reporting that speakers’ use of ‘smallwords’ – that is, common set phrases that fill, bridge, and keep the conversation going – can improve a listener’s (and similarly test rater’s) perception of fluency and competency of the speaker.

The book serves as a useful overall guide to speaking test construction and considerations. Rather than focusing on ‘teaching’ testing terms, the book shows how test concepts can be applied and critiqued in practice, and offers some examples of actual tests to illustrate these points. In terms of future direction (keeping in mind this book is already nearly two decades old), Luoma points to rating checklists. The ‘new’ concept of sociocultural theory also has implications on speaking being interactional, so does not lend itself well to the traditional speaking test one-on-one candidate-interlocutor/interviewer mode.

A new term that I haven’t seen in the books so far read, but which I don’t think would be likely to appear in the Delta test, is setting cut scores. Also called standard scores, this involves cutting raw scores into ranges and bands that determine pass, fail, etc.

Another item that I thought was a good reminder from a regular teaching point of view, was to explain tails and topicalisation to students, as this is an area that I don’t think I have ever seen covered in class course books, but would be worth including. This refers to the very normal spoken feature of presenting key information about the topic at the end or at the beginning of a sentence in a non-standard grammatical way. Luoma gives the example of Joe, his name is from Quirk and Greenbaum (1976).


Language Testing and Assessment – an advanced resource book, Glenn Fulcher and Fred Davidson, Routledge Applied Linguistics, 2007

language testing and assessment

I started reading this book, but then stopped after the first section.

At over 400 pages, and some ten years more current than the other books on testing that I’ve summarized so far, this title is a nice addition to the field of language testing study. It  considers how testing terms and concepts are presented in prior literature and goes into much more academic depth in discussing their meaning and significance. It also reminded me how there are some terms that haven’t been satisfactorily explained by some of the other books I’ve read so far – such as construct validity and concurrent validity, and which I need to seek more clarity on. However, it quickly became clear that the content of this book is arguably more suited to a Master’s candidate reader than a Delta reader, and because time is limited, I am therefore putting this book away for the time being. Hopefully I will have another opportunity to come back to it in the future, as I think it freshens the debate.