Each of the CBI’s qualitative business surveys is underpinned by a scientific weighting and aggregation structure using official Government data or other research sources and instruments
There is a specific process involved with the CBI surveys. Firstly, the percentage of answers to each reply option is calculated. Two alternatives become available at this stage: a simple unweighted counting of the answers and a weighted counting. In the first case, the numbers of positive and negative replies are counted, then expressed as percentages of the total number of firms in the sector. In the second case, a weighted coefficient is used for each firm representing a component of its official size ie employment or turnover or production. The answers to each option are then given as the sum of the coefficients of the firm that have given the same reply. The application of weighting measures at this point in the calculation process aims at reducing the bias introduced by the inclusion of different companies with different sizes. However, it is important to note that the weighting of business survey results does not follow the same reasoning as that applied to the calculation of official statistical indicators.
The most recently developed individual survey weighting system was for the CBI Service Sector Survey. This weighting system not only weights response by employment and turnover, but allows within the system for adjustments to be made where there are non-responses in the system and allows for the pre-non response and post -non response adjustment weight to be displayed for those companies that did respond.
The CBI undertook a major 'top-level' survey reclassification and reweighting exercise in 2010 and further information upon this important work will appear here in due course. The CBI is currently engaged in revising its weighting data based upon the latest official updates. Please note that the Survey Management Group is currently developing an online and hard-copy Technical Survey Guide which will provide a comprehensive picture of our surveys and how they operate. This will be available in March 2012. Further details available in due course.