Key points at a glance
- Existing data must actually support your research question.
- Read the codebook and original design before analysing.
- Document the selected subset, processing and limitations.
What is secondary data analysis?
Secondary analysis uses previously collected data for a new investigation. The material may include survey records, documented interviews or other suitable research data. The UK Data Service describes reanalysis of existing material as a research approach in its own right.
The advantage goes beyond avoiding your own data collection. Existing material may cover periods or populations you could not realistically reach. At the same time, your work is constrained by available measurements and the original selection decisions.
Match the question to the data
Before downloading, list the concepts, population, periods and comparisons you need. Then inspect the codebook, exact question wording and collection documentation. A promising variable name does not guarantee that your intended concept was measured.
Original example: You want to study time spent on digital learning, but the dataset records only days of use. This variable can address frequency, not duration. Adapt the question transparently or find more suitable data rather than ignoring the distinction.
Check access and data quality early
Investigate usage conditions, access procedures and file formats in advance. A publicly visible catalogue entry does not mean that every file is immediately downloadable. Allow time for applications and for understanding the accompanying documentation.
Examine how the original sample was selected and which groups are absent. For surveys, inspect available weights and the sampling design before applying standard procedures. Different analyses may require different weighting decisions; explain the choice you actually make.
For qualitative material, collection context is especially important. Without information about the guide, interview setting or participant selection, apparently clear statements may be difficult to interpret. Record these gaps as limitations rather than filling them with assumptions.
Make your own contribution traceable
Describe the original collection separately from your selection and processing. Identify the version, variables, exclusions, joins and derived measures. Readers should understand how the supplied resource became the dataset used in your analysis.
Your contribution may be a new question, a justified comparison or a transparent reanalysis. Do not automatically claim to be the first to investigate the issue. Check earlier publications using the dataset and explain what your work adds or examines again.
Final submission context
Formal details can feel like a separate writing task, but they matter just as much for a printed submission. Anything missing, misplaced or inconsistently formatted in the document will also appear in the bound copy.
Use this guide together with your cover page, table of contents, page numbers, source notes and appendices. Prepare the final PDF for printing and binding only after the complete file has been checked.
Practical check before PDF export
- Do headings, chapter structure and page numbers match?
- Are sources, figures and tables included completely?
- Are required elements such as declarations, appendices or the cover page included where required?
- Did you open and check the final PDF after exporting it?
Checklist
- Identify the required concepts.
- Check question wording and the codebook.
- Clarify access and permitted use.
- Understand the original design and relevant weights.
- Document your analysis and the data version.
Common mistakes
- Treating a suitable variable name as proof of suitable measurement.
- Checking access deadlines only when analysis should begin.
- Describing the original collection as your own fieldwork.
Print and bind your finished thesis
Once you have checked the content and final PDF, configure the printed copies to match your submission requirements.
Configure printing and bindingRelated content
Explore related guidance on thesis structure, sources, formatting and final submission.
Frequently asked questions
Does secondary analysis count as empirical research?
It can, because you analyse data to address a question. Confirm with your supervisor that the particular approach meets the requirements of your programme.
Must I reanalyse the entire dataset?
The question determines the scope. Select and justify an appropriate subset. Simply retelling an existing finding is different from conducting your own analysis.
What if an important variable is missing?
Consider whether a defensible alternative operationalisation is available. Otherwise revise the question or data source. An approximately related variable must not silently be treated as an identical measure.