Historical Statistics of the United States is a familiar part of public and academic reference collections. Twenty‐two years after the appearance of the bicentennial edition, Cambridge University Press presents a valuable electronic version of this classic. Edited by a team of scholars, librarians, and computer experts, this excellently designed CD‐ROM greatly enhances access to the rich historical data on American society contained in this work. It also stands as a model for electronic reference tools. The program consists of a single disc plus a well‐designed user manual. It operates with no problems on both IBM compatible PCs (386 or higher processors, Windows 3.1 or later, 8 MB of RAM) and Apple Macintoshes (System 7 and 4 MB RAM.) An option for network access is available. Installation is very simple on either PC or Apple systems.
As with Cambridge University Press’s other electronic products (Chaucer’s Wife of Bath’s Prologue, The World Shakespeare Bibliography, and Johnson’s Dictionary) Historical Statistics uses DynaText software for searching and viewing data. The 12,500 tables and all descriptive texts from the 1975 edition have been converted into electronic format and encoded in Standard Generalized Markup Language (SGML).
DynaText uses a split main screen to display the table of contents on one screen and data tables and notes on the other. Users browse the table of contents and view data by clicking on chapter headings. Clicking on icons next to main chapter headings displays its subheadings. Through this method, users may browse through the notes and data series.
Data for each chapter contains source notes and tables. Hypertext links placed with the notes allow direct access to data sets. Similar links appear with the data tables for reference back to source notes. Information can be downloaded in either Lotus 1‐2‐3 or compatible spreadsheet or text formats. Movement through the texts is very simple and straightforward, with the exception of a lack of explanation of how to close source notes accessed from data tables via hypertext links.
Along with browsing, simple and advanced keyword search modes are available from a search window which appears at the bottom of the main screen. Boolean connectors, proximity operators, and use of wild cards provide flexible and powerful searching of tables and text. Also available are search forms methods for searching by data table, table title or column headings, table titles, or data series numbers.
Electronic formats, either in CD‐ROM or other formats, are the best methods for accessing statistical data sets. This version of Historical Statistics is a fine example of a well design tool that greatly enhances access to the valuable information contained in this work. The next step is for the academic community to produce an updated version in the same easy to use format.
