Fourteen abstracts, taken from a recent issue of the British Journal of Educational Psychology, were printed in four different versions. Version 1 was the original format. Version 2 matched Version 1, but its type-size was increased to that of the corresponding article. Version 3 matched Version 2, but paragraphs were introduced to clarify the underlying structure. Version 4 matched Version 3, but the text of the abstract was rewritten in an attempt to clarify the text even further.
The effects of these changes were assessed in four ways. Firstly, undergraduates each ranked for clarity Versions 1–4 of individual abstracts. The results showed that each change in design lead to a significantly greater preference - thus Version 1 was the least preferred and Version 4 the most preferred. Secondly, Flesch readability scores were computed for the original and the revised abstracts. It was shown using this measure that the revised abstracts were significantly easier to read than the originals. Thirdly, academic staff and postgraduates completed cloze (comprehension) tests on original and revised versions of two of the original abstracts. The results showed (i) significantly higher cloze test scores for the revised versions, and (ii) that experienced staff and postgraduates performed significantly better on both versions than did beginning postgraduate students. Finally British and overseas postgraduate students of librarianship completed a cloze version of an original and its revised abstract. The results showed that the British students did significantly better than overseas ones, and that both groups scored significantly higher on the revised abstract.
These four sets of data thus indicate that it is possible to improve the clarity of journal abstracts, and that this can be achieved for many different kinds of reader.