The REF is wrong: books are not inferior to papers

Actually, it is largely the H and not the SS that publish books, as any publication database can show you. Books suffer from a lack of definition as to what the object covers and also a vast number of pay-to-publish outfits that print on demand, and the demand is sometimes very small. As long as books are necessary for promotion, as in many countries not so far from the UK where never mind the quality feel the width approach is adopted, Mickey Mouse publishing will continue, and it is not only the UK that will be dragged down. Part of the problem lies in quality assessment. Journal standards and their peer review methods are known, and give a good proxy of quality. Books are another kettle of fish. Books are often the highest level of achievement,and this must be praised and defended, but low achievement books are rife. The analysis of evaluation protocols and their strengths and weaknesses is being actively undertaken within the ENRESSH COST network. We have SIG looking at books both from a bibliometrics and non-bibliometric aspect. Many of us are in fact humanities scholars keen to promote good quality outputs. Evaluation and books are a Europe-wide question, and beyond as we have observers from outside the hallowed EU. Unfortunately we have little UK input, it is as if you have already left Europe and have confused Europe and European values with those of the EC.

But in the last REF, across the humanities (panel D), those books that people did put in did extremely well. They were much more likely to score 4* than articles. That was true for the ones that were double weighted too. The take-home message was 'the book is back'. If people haven't realised this, the problem is not due to the REF, but to false beliefs about the REF.

In 2014 I saw no evidence of books being undervalued in sub-panel 34 Art and Design: History, Practice and Theory. Some sub-panels may choose to (mis)interpret the messages of the REF, but it seems unfair to suggest that the REF itself is flawed in this respect. Main panel D reported that "almost all the requests for double-weighting – 98 per cent – were accepted." (REF 2014: Overview report by Main Panel D and Sub-panels 27 to 36, January 2015).

There are high and low quality book publishers - just as there are high and low quality journals. What matters is not the medium, length or format but the standards. It is not brain surgery to distinguish: University presses always have a strict review process before and after the m/s is completed. Trade presses do not always follow the same process. A university press book is usually worth 100 articles in terms of impact on the discipline and citations. Use the evidence of citations to evaluate.