Step 14: Secondary Data, Mixed Methods and Systematic Reviews
MatchesDesk-based dissertations · mixed-methods designs · systematic and scoping reviews (PRISMA 2020)
In 30 seconds
- Secondary data were collected by someone else (national surveys, administrative records, the World Bank, the ONS). They are large and cheap, but may not measure exactly what you need.
- Mixed methods combines quantitative and qualitative data. The three core designs are convergent (both at once), explanatory sequential (numbers first, then interviews to explain) and exploratory sequential (interviews first, then a survey to test).
- Triangulation means checking a finding with more than one kind of evidence.
- A systematic review answers a question by finding, screening and appraising all relevant studies with a pre-set, reported method.
- The PRISMA flow diagram reports how many records were found, screened, excluded and included.
Video lesson coming soonA short animated lesson for this step will appear here.
Secondary data
Many excellent dissertations collect no new data at all. National household surveys, labour force surveys, university administrative data and international datasets give samples no student could collect alone, often over many years.
| Advantages | Disadvantages |
|---|---|
| Large, often representative samples | Variables may not match your concepts exactly |
| Saves time and money | You did not control how the data were collected |
| Often covers many years (time series or panels) | Definitions can change between years |
| Fewer ethical issues if anonymised | Access may need registration or a licence |
In your dissertationFor secondary data, report the source, who collected it, when, how it was sampled, the exact variables you used (with their original names), any recoding, and why the measures fit your concepts. Read the dataset’s documentation (the “user guide” or “codebook”) before you start.
Chapter 14 of Research Methods Made Simple covers this in full, with a worked mixed-methods study and a PRISMA template.Get the textbook · $15
Mixed methods designs
| Design | Order | Part-time jobs example |
|---|---|---|
| Convergent | QUAN + QUAL at the same time, then compare | A survey with open questions, analysed side by side with interviews run the same month |
| Explanatory sequential | QUAN → QUAL | The survey finds marks fall above 20 hours; interviews with those students explain why |
| Exploratory sequential | QUAL → QUAN | Interviews reveal “work as structure”; a new survey item measures how common it is |
Worked example
Plan an explanatory sequential study and show how the two phases connect.
Phase 1Survey 300 students. Result: marks are lower on average among students working over 20 hours, but a quarter of them score above the overall median.
LinkUse the survey to select interviewees: 5 long-hours students with high marks and 5 with low marks (purposive sampling).
Phase 2Interviews show that high achievers tend to have flexible employers and fixed weekly study routines.
IntegrationReport a joint display: the quantitative pattern, the qualitative explanation, and what they mean together. Integration is what makes it mixed methods rather than two separate studies.
Systematic reviews
An ordinary literature review (Step 3) supports your own study. A systematic review is the study: it tries to find every relevant piece of evidence on a question using a method set in advance, so that another researcher could repeat it and get the same set of studies.
- Write a focused question (often with PICO, Step 2).
- Set inclusion and exclusion criteria before searching.
- Search several databases with a documented search string.
- Remove duplicates; screen titles and abstracts; then read full texts.
- Appraise the quality of each included study.
- Extract data into a table and synthesise (narratively, or statistically in a meta-analysis).
Worked example
Complete a PRISMA flow. A search finds 1,240 records. 310 are duplicates. Title and abstract screening excludes 842. Full-text reading excludes 61.
Identified1,240 records
Screened\( 1{,}240 – 310 = 930 \) after duplicates removed
Full text\( 930 – 842 = 88 \) assessed for eligibility
Included\( 88 – 61 = 27 \) studies in the review
The PRISMA diagram also lists the reasons for full-text exclusions (wrong population, no relevant outcome, not empirical), with numbers for each.
Common mistakeCalling a literature review “systematic” because you searched Google Scholar carefully. A review is systematic only if the question, criteria and search are set in advance and reported, screening follows those criteria, and the numbers at each stage are documented.
Practise until it sticks: Research Methods Made Simple Practice Questions has 1,004 questions, each with a worked answer.Get the practice book · $9
Practise: 6 questions
Try each one on paper first, then check the worked answer.
EasyQ1. Give two advantages and two disadvantages of secondary data. Show worked answer
Advantages: large samples; cheap and fast; often long time periods. Disadvantages: variables may not match your concepts; no control over collection; definitions may change; access restrictions.
EasyQ2. What is triangulation? Show worked answer
Checking a finding using more than one source, method, researcher or theory, so the conclusion does not rest on a single kind of evidence.
MediumQ3. Name the design: interviews with 12 nurses identify five sources of burnout, then a survey of 400 nurses measures how common each is. Show worked answer
Exploratory sequential: qualitative first to discover, quantitative second to measure.
MediumQ4. A search finds 860 records; 140 are duplicates; 655 are excluded at title and abstract; 48 are excluded at full text. How many studies are included? Show worked answer
\( 860 – 140 = 720 \) screened; \( 720 – 655 = 65 \) full texts; \( 65 – 48 = 17 \) included.
MediumQ5. Why must inclusion criteria be set before searching? Show worked answer
Setting them afterwards allows the reviewer, consciously or not, to include studies that support a preferred answer and exclude those that do not. Pre-set criteria make the review unbiased and repeatable.
ExamQ6. Evaluate the use of mixed methods for studying how part-time work affects students. [6] Show worked answer
Strengths: the survey shows how common and how strong the association is [1]; interviews explain the mechanisms behind it [1]; triangulation strengthens confidence in the findings [1]. Weaknesses: more time and two sets of skills needed [1]; findings may conflict and must be integrated, not just placed side by side [1]. Judgement: well suited to a “how and how much” question if time allows and integration is planned [1].
Check yourself
1. A survey followed by interviews to explain its results is
Quantitative first, then qualitative to explain.
2. PRISMA is used to report
It shows records identified, screened, excluded and included.
3. Statistically combining results from several studies is a
Meta-analysis pools effect sizes across studies.
Your score: 0 out of 3
The textbook and the practice book together: the Research Methods Made Simple bundle for $19 (was $24).Both books · $19
The short version
- Secondary data give scale cheaply; check that the measures fit your concepts.
- Mixed methods: convergent, explanatory sequential or exploratory sequential, always integrated.
- Systematic reviews set the question, criteria and search in advance and report every stage.
- Use a PRISMA diagram to show how studies were selected.
Studying more than one subject? The Complete Book Bundle includes every Economics, Statistics, Mathematics, Econometrics and Research Methods book.
Want more questions like these, with worked answers?
The practice book has 1,004 more, from quick recall to exam style.