Some problems in measuring manufacturing output

It is the purpose of this study to examine and evaluate some of the techniques and to consider some of the problems involved in measuring the output of manufacturing industries. In the course of this analysis, some of the weaknesses in measurement technique will also be indicated. Throughout the study, the emphasis will be on measurement employed by the Bureau of the Census in taking the Census of Manufactures, undoubtedly the most nearly complete and most comprehensive record of manufacturing production available from any source. The primary source materials for this analysis will be the volumes of the Census of Manufactures dating from 1899 to 1947. In addition, the more important writings in the field of manufacturing output research will be used, especially the works of Charles A. Bliss, Soloman Fabricant, Simon Kuznets, and William H. Shaw. The procedure to be used in this study is as follows: first, to summarize briefly the historical background of the manufactures census and to examine the techniques presently used by the Census in measuring output. Second, to evaluate these techniques by indicating (a) the principal uses of output data (b) the limitations of the data for these uses and (c) some alternative measurement techniques. Last, to summarize what has been shown and to indicate some current statistical gaps in output data as well as possible improvements in the Census program for manufactures statistics.