Secondary data analysis
Ask a new question of existing evidence
Secondary data analysis reuses data collected previously for another study or purpose. Sources can include survey microdata, archived interviews, financial databases and digital traces. It is a data-source strategy rather than a specific estimator. Researchers must assess whether the original sampling, measurement, consent, timing and documentation support the new question before choosing an analysis.
Choose when a suitable existing dataset offers coverage or historical depth beyond feasible primary collection. Confirm lawful access, variable definitions, sample design and whether the desired construct was actually measured.
Strengths
- Can access scale and historical coverage economically
- Enables reanalysis and replication
Limitations
- Available variables constrain the research question
- Original selection and measurement problems persist
Know the boundary
Existing data are not automatically representative or adequate for causal identification.