Tools for Data Sharing
Data cannot be shared effectively unless users can make sense of them. That means data have to be properly managed. Datasets must be cleaned and coded. Questionnaires or data collection forms used to generate them must be available in a language the user can understand. Often, datasets collected in different locations or at different times need to be combined to allow for useful analysis. And yet data management is usually the most neglected part of public health research.
Ternyata’s data management expertise derives largely from the management of data generated by national behavioural surveillance systems for HIV. This involves combining similar but non-identical data collected from up to six different risk populations in dozens of cities over multiple years. We’ve written
First Things First: Guidelines on management and coding of behavioural surveillance data Family Health International, Bangkok, 2006
Detailed technical guidelines for data managers. Prepared at the request of the WHO/UNAIDS surveillance working group, these guidelines provide data managers with information on how to create datasets which will meet the needs of HIV prevention and care programme managers, so that surveillance data can easily be translated into improvements in services. They also contain useful information on questionnaire design and data entry.
While looking at issues surrounding access to data held in biobanks around the world, we created an online library of resources relating to data access. If you use the Firefox browser with the free Zotero reference management plug-in, you can access the library at https://www.zotero.org/groups/4900/