Secondary analyses of existing data

Secondary analysis of existing data involves analysing (quantitative or qualitative) data which has been collected by someone else for some other purpose; whereas primary data has been generated by an individual (or research team) for a specific purpose. The methods of analysis could be the same for primary and secondary analyses, but the question that is being answered with secondary analyses of existing data is not what the data was specifically collated for. For example, if a team conceives of and develops a research project, collects the data to address the specific questions to be answered by the study and do their own analyses of the data it is then primary analysis. Another research team may then pose a question that could be answered through analysis of the study data – this is secondary data analysis. Secondary data analyses are often performed using routinely nationally-collected data such as electronic health records and registry data.  Examples of research questions include:

Data archives exist to provide access to high quality quantitative and qualitative data from previously conducted studies and are a good source for data to undertake secondary analyses. For example UK Data Archives provides a platform to access social, economic and population data and UK Data Service access data around health, and includes studies such as the Health Survey for England and Surveys of Psychiatric Morbidity.  


Other national and international repositories include:  

These datasets may allow you to answer your research question, please be aware you will need to register and possibly request permission to access the dataset, and if you proceed you will need to agree to certain conditions.  

Useful references: