21890 Chap.6 Secondary Data: Sources, Validity and Coverage
Secondary Data: Sources, Validity and Coverage
Why secondary data
Students do not have ethics clearance for primary research, so projects reuse data that others gathered for a different purpose. The Week 5 lecture shows it is everywhere, from government surveys to meeting minutes and payroll records.
Two fundamental questions
Is the measurement valid for your concept, and does the data cover your population, period and variables?
Merits such as low cost and long time series come with challenges such as poor quality or a different original purpose.
Handling it well
Go back to original sources, build your own tables, respect proprietary and private data, and check reliability before you rely on a figure.
What this chapter covers
- 01
Primary and secondary data
- 02
Why the project uses secondary data
- 03
Merits and challenges
- 04
Valid measurement
- 05
Sufficient coverage
- 06
Finding data through the library
Worked example · free
Test whether sick days are a valid measure of staff mental health
- 1State what the concept needs: a measure of mental health or wellbeing among staff.
- 1Test the proxy: sick days record absence for any cause, so they are at best an indirect measure, which is the lecture's warning about proxies.
- 1Recommend: justify the proxy and state its limits, or add a closer measure such as published staff wellbeing survey results.
Key terms
- Proxy Measure
- An indirect indicator used in place of the concept a study really wants to measure.
- Data Coverage
- The extent to which a data set spans the population, period and variables a question needs.
Secondary Data: Sources, Validity and Coverage FAQ
What counts as secondary data for this project?
Existing data gathered by others for a different purpose, such as government statistics, industry database reports, company annual reports, minutes of meetings or published survey results. Your literature review discusses research; secondary data is the material you analyse to produce findings.
What are the two questions to ask of every data set?
Following Saunders, Lewis and Thornhill, ask whether the measurement is valid for your concept and whether the data has sufficient coverage of the population, time period and variables your research question needs. A source that fails either needs a stated limit or a replacement.
Can I use figures quoted in a news article?
Treat the article as a signpost, not the data. Trace the figure to the survey or report that produced it, check who collected it and how, then cite and analyse the original. If you cannot find the original, the figure should not carry your argument.
What are the merits of using secondary data?
The Week 5 lecture lists four: it takes less time and money to access and analyse, it is less obtrusive, it allows longitudinal analysis over years, and it helps you put your study in context. Each merit comes with the condition that the data fits your question.
What are the main challenges of secondary data?
It was gathered for someone else’s purpose, it can be difficult or costly to access, and it can be of poor quality. Those challenges are why every data set needs a validity check and a coverage check before it supports an argument.
Why go back to the original source of a table?
A chart or table printed in a report may not suit your study, and the lecture suggests building your own from the original figures. Going back also lets you check how the data was collected and whether it is still current.
Can I use a company's internal data?
Some company data is proprietary, so you need approval to use it, or it must be anonymised first. The lecture links this to the ethics topic in Week 6 and asks you to consider what is publicly available and what is private.
Assessment move
For every data source in your proposal, write one line on validity and one on coverage before you commit to it. Do the Week 5 library exercise on your own topic, locating a statistic and a company document, and note the challenges each would raise in your report. Keep a short data log with the publisher, the period and population covered, the variables and any limits you found.
When a data set falls short, decide explicitly whether to narrow the question or add another source, and record the reason. That log feeds Section 4 of the proposal and the method section of the report.
Working through Secondary Data: Sources, Validity and Coverage in 21890? Sia is AskSia’s AI Management tutor — ask any 21890 Secondary Data: Sources, Validity and Coverage question and get a clear, step-by-step explanation grounded in how 21890 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.