1. Introduction
Many analyses of firm performance are based upon self-reported measures. However, these are likely to be subject to reporting error and/or perception biases. Respondents are often under legal obligations to provide correct data for official purposes, such as tax reporting. However, even when it is a compulsory requirement to return survey questionnaires, the incentives and thus the time taken to fill them in are lower (particularly if one member of staff does not hold all of the required information). Because of this, one might expect self-reported measures of firm performance to be more subject to general reporting error than alternative official sources. In addition, self-reported measures of relative performance may be subject to the biases observed in the psychology literature. One of these is that most people tend to think they are above average.
In this paper we shall consider both absolute and relative performance, reported in the Business Operations Survey (BOS), with alternative measures taken from official sources, brought together under the Improved Business Understanding via Longitudinal Database Development (IBULDD) project in the prototype Longitudinal Business Database (LBD). This database provides us with a unique opportunity to compare self-reported objective and subjective measures of performance with those from administrative sources. This will enable not only users of the BOS itself, but also other similar surveys to understand the strengths and limitations of such data and how to interpret their results. It also allows us to consider the appropriateness of previous analyses, as well as providing better quality research and policy advice in the future. Another implication of such work is for the construction of surveys themselves. If financial information can be just as well or better gleaned from administrative sources – and there are no legal obstacles to this happening – it may be possible to remove these questions from the surveys. This would reduce respondent load and/or allow time for other important questions.
In the following section, we briefly consider the issues surrounding the collection of data through survey and administrative sources. At the heart of our analysis is the data upon which we base our comparisons. These are described in section 3. We describe our results in section 4. Section 5 concludes.
Back to Top