Secondary data analysis is analysis of existing data collected where its original purpose may not be for research. Its advantage is economy, breadth of data -- a lot of which is available for public use -- which may not be available to small research teams, particularly student researchers.
A secondary dataset should be examined carefully, however, to confirm that the data defined and coded allows for the desired analysis. Also, you need to appraise the quality of data carefully by reading documentation of data collection procedures, missing data, code book.
If you need to find a statistics such as total number of hospital bed in the United States in 2005-2015, consult the Finding Health Statistics guide.
Define the question |
Example:
|
Specify the population |
Examples:
|
Specify the variables to include in analysis |
Examples:
|
Specify the type of data most appropriate to answer the question |
Examples:
|
Create a list of datasets that include the information needed | This is an iterative process. You may have to revise your research question based on available data. |