Ann Kitchen

Summary of ‘Electronic reporting and its use for business surveys’.

By: Janet Sear, Statistics Canada

This paper overviews aspects of the way in which Statistics Canada believes its data collection of business surveys should and will develop in the future. The need for such an investigation stems from the availability of a number of reporting options that are likely to increase efficiency and accuracy of surveys compared to previous methods. The benefits of the new methods come from the elimination of the need to key the data, and from the reduction of response burden.

There are two aspects of electronic reporting that the paper considers: data collection and data transfer. Numerous forms of data collection are discussed, including questionnaires on diskettes, on the web, and in commercial software, electronic files, and touch-tone data entry. The options available for data transfer discussed in the paper involve mailing of diskettes, direct dial-up, internet (FTP and e-mail), and value added networks.

The paper outlines an example of a successful use of electronic reporting; an electronic collection vehicle developed for a retail chain and department store survey. The use of this method was found to be low cost, and it reduced the response burden. It allows the respondent to easily import the data from a spreadsheet or database. Interactive edits were also a feature of the collection, and this was found to be advantageous as the personalized instructions reduced errors associated with respondents not being fully aware of the survey coverage and definitions.

This method involved the respondents returning the information on diskette by mail. As this was an annual survey, this was no problem, however for a more frequent survey, return time would be an issue. This paper was published in 1997, so at the time of writing, access to the internet was probably less that that of today; the data collection could likely be done using the internet now.

Overall, Statistics Canada found that this methodology worked very well, and that respondents were in favor of it. They found it very easy to use as well as time-saving. The quality of the data improved from that acquired using alternative methods, and costs and return time of data were both reduced.