From Survey Data to Statistics
Once the survey data have been captured, coded, and cleaned (see Chapter 11), and, where relevant, estimated to the total population (see Chapter 5), analysis needs to be carried out to bring together the key messages contained within the data; then the derived outputs must be prepared for dissemination and disseminated. This chapter provides an introductory overview of the phases involved in moving from survey data to statistics: analysis and dissemination (see Figure 12.1). The chapter also reviews the archive phase in the generic statistical business process model (GSBPM) (see Chapter 1), which is another dissemination process. The final phase evaluation was covered in Section 10.7. This chapter presents an introductory overview that merely scratches the surface of some specialized areas; key sources of further information are cited throughout this chapter.
The ultimate aim of the analysis phase is to produce statistics that meet the users' (or the researchers') statistical needs in relation to the output quality dimensions (see Chapter 3), and in doing so, build or maintain trust in the produced statistics. Part of the trust dimension will include maintaining data security and confidentiality during analysis; this is especially important ...