CRANV2 Introduction chapter Cyril Cleverdon Michael Keen Cranfield An investigation supported by a grant to Aslib by the National Science Foundation. Use, reproduction, or publication, in whole or in part, is permitted for any purpose of the United States Government. - 2 - which should be beyond the comprehension of a schoolboy of average intelligence, but, even so, it is suggested that Chapter 3 might be skipped by those who are not closely concerned with the particular problem of performance measurement. Important though the work in this Chapter is felt to be, yet the arguments may well be of marginal interest to many readers. At the beginning of Chapter 4, which presents the main set of test results, full information is given concerning the performance measures which are actually used; Chapter 3 explains in some detail why those measures were selected in preference to other possible measures. To a lesser extent, the same is true of Chapter 2 which discusses at length the variables which were being investigated and the environment in which the test was carried out. Again, we have tried to make Chapter 4 complete in itself in that such matters are briefly recorded therein. Only if the reader is puzzled as to why such seemingly unnecessarily tortuous actions have been taken, need he refer to Chapter 2 to find the possible justification. The test results presented in Chapter 4 make up the main bulk of the report. Some may cavil at the way in which, at the slightest provocation, we include plots of the results. Undoubtedly these add to the bulk, but we can only hope that they will allow readers more quickly to get a general idea of what has been happening. The following chapter presents substantially the same set of results in a simpler but probably more controversial manner. In Chapter 6, extracts have been taken from the main test results and presented in such a way as to illustrate different aspects of the investigation. Subsidiary to the main test was an attempt to make a comparative evaluation of citation indexing and bibliographic coupling. While there should be no serious problems in making such an evaluation under operational conditions, the value of testing this form of index in an artificial environment appears dubious. However, with considerable reservations the results are given in Chapter 7. Up to this stage the results have been presented without any attempt being made to draw conclusions. All such have been relegated to the final chapter of this volume, in which an attempt is also made to relate the results to other investigations in this field. There is one general apology that should be made and that is for the introduction into this report of yet more jargon. Many terms first used in reporting .Cranfield I now appear to have gained general acceptance, but it is unlikely that such phrases as ,maximum starting term coordination level method' or ,proportional coordination level method' will crop up very frequently in the literature - and we certainly hope they won't - but it has been necessary to find terms to describe certain procedures so that, in reports of other tests, one has a chance of knowing which of several