4 Data exploration
4.1 Data exploration tools

Software applications for charts, programming, databases and spreadsheets are commonly used to explore data. Here are some examples of applications:

• Spreadsheets are programs that allow adding columns and rows of figures, to calculate means and to perform descriptive statistical analyses. They can be used to create summaries of results as well as charts and graphs to better understand relations between variables. These can be displayed in a number of ways: bar charts, line charts, and pie charts are just a few examples of the data visualizations that can be produced.
• Data is sometimes stored in databases for easy access and to allow the production of summaries, aggregate data or reports. A database program should be able to store, retrieve, sort and analyze data.
• Specialized programs can be developed to edit, clean, impute and process the final tabular output. They offer the full service in one module and can be used each time the same survey is completed and entered within the system. These programs will produce results ready to be published.
• Statistical software applications are used for data processing and to produce summaries and data visualizations, but they can also be used to carry advanced statistical analyses such as modelling.

One example of a very popular data exploration tool is R software. R is a programming language and open-source software that anyone can download and install on their personal computer to transform, explore and analyze data. All charts and graphs presented in the upcoming sections were created with R.

Computer output obtained from these data exploration tools may be used in a variety of ways. They can be saved for future retrieval and use, be sent to other teams in electronic files or be disseminated online to communicate statistical information to users. Output is usually governed by the need to communicate specific information to a specific audience. To help determine the best output type for the information you have produced, ask yourself these questions:

• For whom is the output being produced?
• How the audience will best understand it?