The increasing emphasis on policy decisions based on evidence has led to a demand for data and the application of statistical analysis to:

• identify the scale of the problem;
• analyse its causes and suggest ways to address them; and
• monitor the impact and evaluate the outcomes of policy implementation.

In addition to identifying the scale of the problem, descriptive statistics are used to describe the baseline situation and to monitor the changes perceived after the introduction of the policy.

Inferential statistics uses statistical techniques and an analytical framework to draw inferences about unknown aspects. In a policy context, inferential statistics can be used to analyse the causes of a problem and thereby suggest methods for addressing them. The statistical analysis can also be used to evaluate the outcome of a policy by separating the outcomes that can be attributed directly or indirectly to the policy itself from those that would have occurred even if the policy had not been implemented.

We often find that the exact data required are not available. An important component of our statistical work relates to advising on the relevance of existing datasets and on methods to develop more useful data. We also find that qualitative data analysis can add value to quantitative data by providing an understanding of processes.