CRANV1P1 Documents and Questions chapter Cyril Cleverdon Jack Mills 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. -19 - CHAPTER 3 Documents and Questions To provide the necessary basis for the test, we required a collection of docu- ments, a set of search questions, and a complete assessment to determine the documents relevant to each question. These aims were accomplished in three main stages: Stage 1. A letter was sent to authors of research papers, requesting search questions and a relevance assessment of the papers they cited. Stage 2. Using the collection of documents and a set of questions made UP from the replies to stage one, technically competent people examined every document in relation to every question to find any relevant documents in ad- dition to the authors' cited documents. Stage 3. The additional documents judged relevant in stage two were submitted to the authors, requesting their final assessment of relevance. First will be givendetails of the methods used in these three stages, and the response made by the authors. Then will follow a more detailed examination of the question-document assessment of relevance, and finally a brief analysis of the questions. Methodology and authors' responses 271 recent papers on the subject of high speed aerodynamics and aircraft structures were obtained. Although high speed aerodynamics had been chosen as the main subject for the test, a small set of documents dealing with aircraft struc- tures was introduced in order to examine the effect of including two dissimilar sub- jects in one collection. These papers were referred to as base documents, and in order to be accepted for the test a base document had to satisfy certain criteria; it had to be a paper published in the English language containing at least two references in a bibliography, these references being in English, dated 1954 or later and likely to be readily obtainable. Since aerodynamic papers contain on average about twelve references, neither this nor any of the other requirements caused the rejection of many papers. Most of the selected papers were published during 1962, and the first half of 1963; the articles from one prominent journal predominated, but some research reports were included. A list of the different sources of those which were finally used is given in Table 3.1. 76.9% of the papers are American publications, 22.5% British and 0.6% Swedish. To the author of each of these papers' was sent a form, quoting the title and reference of his own paper, and also listing up to ten of the papers which had been included as references. The authors were asked to do two things. 1. To state the basic problem, in the form of a search question, which was the reason for the research being undertaken leading to the paper, and also to give not more than three supplementary questions that arose in the course of the work, and which were, or might have been, put to an information service.