This procedure consists of projecting sample survey results to the total market through the use of published market data. With industrial products, the Standard Industrial Classification (SIC) system provides a unique and effective way of projecting survey data.
Because the Standard Industrial Classification system as developed by the federal government is by far the most widely used system of industrial classification, it is described here in some detail.
The SIC system is intended to cover the entire field of economic activity: agriculture; forestry and fisheries; mining; construction; manufacturing; transportation; communication; electricity, gas, and sanitary services; wholesale and retail trade; finance, insurance, and real estate; services; and government. Reporting units are establishing, not legal entities or companies. Each establishment is classified according to its major activity. The SIC distinguishes two broad classes of establishment – operating establishments or economic units, which produce goods or services and central administrative offices and auxiliary units, which manage or provide services for other establishments of the same company. The latter type is only partly measured by the SIC system.
All manufacturing industries are combined into about 20 major groups, 150 subgroups, and 450 industries. The following is an illustration:
Group no. 358 – Service industry machines.
Industry no. 3581 – Automatic merchandising machines. Includes establishments primarily engaged in manufacturing automatic merchandise units, also referred to as vending machines (excluding music, amusement, or gaming machines) and coin operated mechanisms for such machines.
Forms that are grouped together in the SIC system have a considerable amount in common. The advantage of using this system is that the government publishes a wide variety of data by such groupings (e.g. the Census of Manufacturers) and the classification has been kept reasonably current to reflect changes in the American manufacturing scene. But the system is not without its drawbacks.
1. When an establishment produces two or more products, the SIC is based on the principal product that is so determined on the basis of sales. Thus, the data on the primary product are inflated.
2. When an establishment is integrated (produces components parts, assembled into a final unit), it is not shown as part of the industry that produces a component part.
3. When an establishment is part of a company complex that engages in centralized buying, that fact is not taken into account in the SIC data.
4. When a firm makes a specialized product that is not important enough to constitute an industry, it may be grouped with producers pf other unrelated types of products in a miscellaneous category (for example, furniture casters are classified in SIC 3429, hardware not otherwise classified).
The need to relate SIC groups to geographical areas can be met through the use of the federal government publication entitled Country Business Patterns, which contains statistics based on information reported under the Federal Old Age and Survivors Insurance Program. Data are presented by SIC within counties on the number of employees, payrolls, total number of establishments, and number of establishments by employment size. Only for the large countries the information for the 150 industry subgroups otherwise the data are limited to the 20 major industry groups.
When using the survey procedure to estimate potential, researchers use mail, telephone, or personal interviews to obtain information on purchases made by firms that have been identified as belonging to the appropriate SIC categories. This information is then projected to the universe of all firms in those SIC categories.
The following steps illustrate the use of the survey method combined with the SIC system to determine potentials for an industrial product:
1. From a sample of companies within each of the industry classes that could conceivably buy the product, determine the amount of the product purchased and the number of employees for each company.
2. Using the sample data, compute average purchases per worker for each SIC group.
3. Multiply the average purchases per worker in each SIC group by the total number of workers in that group. This gives an estimate of the national market potential for each SIC group.
4. Allocate the national market potential for each SIC group among the various counties according to the proportion of that SIC group workers in each county. Thus, if Los Angeles County employs 10 percent of all the workers in a given SIC group, Los Angeles will have a potential of 10 percent of the national potential represented by that SIC group.