Secondary data analysis

Introduction to Social Research Methodology

Author
Affiliation

Ben Stanley

Faculty of Social Sciences, SWPS University

Published

November 17, 2026

Foundations of secondary analysis

Primary data are collected by the researcher, for the researcher’s own question, using an instrument the researcher designed — a survey you commission, interviews you conduct, an experiment you run. Secondary data are data that already exist, collected by someone else and usually for some other purpose: government statistics, a consultancy’s survey, or your own company’s HR records. The distinction concerns origin and purpose, not quality — excellent studies are built on each kind. Its key consequence is that with secondary data, every design decision — sampling, question wording, definitions — was made by someone else, before you arrived.

Figure 1 sets the two kinds of data side by side: primary data are collected by you, for your question; secondary data already exist, collected by someone else for another purpose. The distinction concerns origin and purpose, not quality, and its key consequence is that every design decision in secondary data was made before you arrived.

Primary and secondary data Two columns. Primary data are collected by you, for your own question, with an instrument you designed: a survey you commission, interviews you conduct, an experiment you run. Secondary data already exist: collected by someone else, usually for some other purpose: government statistics, a rival consultancy's survey, your own company's HR records. Under both, the distinction is about origin and purpose, not quality: excellent studies are built on each. The key consequence: with secondary data, every design decision, who was sampled, how questions were worded, how concepts were defined, was made by someone else, before you arrived. Everything attractive, and everything risky, about secondary analysis flows from that fact. PRIMARY DATA Collected by you for your own question, with an instrument you designed A survey you commission Interviews you conduct An experiment you run SECONDARY DATA Already exist collected by someone else, usually for some other purpose Government statistics A rival consultancy's survey Your company's HR records The distinction is about origin and purpose — not quality: excellent studies are built on each THE KEY CONSEQUENCE With secondary data, every design decision was made by someone else, before you arrived: who was sampled how questions were worded how concepts were defined Everything attractive — and everything risky — about secondary analysis flows from this one fact.
Figure 1: Primary and secondary data.

Secondary analysis is a form of research in which data collected and processed by one researcher are reanalysed, often for a different purpose, by another (Babbie). The analysis is new; the data are not. It counts as a genuine data collection method: “collection” here means locating, obtaining, and preparing existing data rather than generating fresh observations. The practice has grown enormously as governments, organisations, and large academic projects have opened their datasets to reuse. For a manager it is usually the first method to consider, because the answer may already be sitting in a filing cabinet or a public database.

Figure 2 draws Babbie’s definition as two studies. The original study runs from its question to its data, analysis and findings; your study starts from a new question, reuses the same data and runs a new analysis to new findings. Collection, here, means locating, obtaining and preparing existing data.

What secondary analysis is At the top, Babbie's definition: secondary analysis is research in which data collected and processed by one researcher are reanalysed, often for a different purpose, by another. Below, two lanes. The original study runs from its question, to the data collected, to its analysis, to its findings. Your study starts from your own question, reuses the same data, and runs a new analysis to new findings: the data are not new; the analysis is. Bottom left: collection, here, means locating, obtaining and preparing existing data, which makes secondary analysis a genuine data collection method. Bottom right: for a manager it is usually the first method to consider, because the answer may already be sitting in a filing cabinet or a public database; the method has grown enormously as governments, organisations and academic projects have opened their datasets. SECONDARY ANALYSIS Data collected and processed by one researcher are reanalysed — often for a different purpose — by another Babbie THE ORIGINAL STUDY Their question Data collected Their analysis Their findings YOUR STUDY Your question The same data A new analysis New findings the data are not new the analysis is new COLLECTION, HERE, MEANS locating obtaining preparing existing data — not generating fresh observations: a genuine data collection method FOR A MANAGER Usually the first method to consider: the answer may already be sitting in a filing cabinet or a public database. It has grown enormously as governments, organisations and academic projects open their data.
Figure 2: What secondary analysis is: the data are not new; the analysis is.

Not the same as a literature review

Secondary analysis must be distinguished from the literature review covered in session 5. A review reuses other researchers’ findings and conclusions; secondary analysis reuses their raw data. The two are different operations with different outputs and different dependencies.

Figure 3 compares the two operations row by row: what you reuse, what you do, what you produce, and what you depend on.

Secondary analysis is not a literature review A comparison in four rows. What you reuse: a literature review reuses other researchers' findings and conclusions; secondary analysis reuses their raw data. What you do: read, evaluate and synthesise, against reanalyse, with new questions, new comparisons and new statistics. Output: a synthesis of what is known, against a new empirical result. You depend on: the authors' interpretations, against the quality of the original data collection. In short: a review tells you what others concluded; secondary analysis lets you check, extend or contradict them with the evidence itself. Literature review session 5: other researchers' conclusions Secondary analysis today: other researchers' data What you reuse What you do Output You depend on Others' findings and conclusions Read, evaluate, synthesise A synthesis of what is known The authors' interpretations Others' raw data Reanalyse: new questions, new comparisons, new statistics A new empirical result The quality of the original data collection IN SHORT A review tells you what others concluded; secondary analysis lets you check, extend or contradict them with the evidence itself.
Figure 3: Secondary analysis is not a literature review.

In short: a review tells you what others concluded; secondary analysis lets you check, extend, or contradict those conclusions with the evidence itself.

Why managers should care

Organisational questions rarely arrive with a budget for a bespoke study attached. Before commissioning anything, a competent research team asks what the organisation already has and what already exists outside it. Existing data can also establish context: is the organisation’s problem unusual, or is the whole sector experiencing it? In the Meridian case — turnover up from 14% to 27% in two years — a great deal can be learned from the company’s own records and from Poland’s official statistics before a single employee is surveyed.

Figure 4 shows why context matters. In both charts Meridian’s turnover rises from 14% to 27%. Where the sector’s turnover stays flat (left), Meridian is an outlier and the problem is its own; where the sector rises too (right), Meridian is riding a sector-wide wave, which calls for a different diagnosis and remedy. The charts are schematic.

Why managers should care about existing data Organisational questions rarely arrive with a budget for a bespoke study, so before commissioning anything a competent research team asks two questions: what do we already have, inside the organisation, and what already exists out there? Existing data establish context. Two small charts show Meridian's turnover rising from 14% to 27% in two years. In the first, the sector's turnover stays flat: Meridian is an outlier, and the problem is Meridian's. In the second, the sector rises too: Meridian is riding a sector-wide wave, which calls for a different diagnosis and a different remedy. At Meridian: before surveying a single employee, ask what the company's own records, and Poland's official statistics, already reveal. No budget for a bespoke study comes attached — so before commissioning anything, ask: INSIDE THE ORGANISATION What do we already have? OUTSIDE IT What already exists out there? CONTEXT: IS OUR PROBLEM UNUSUAL, OR SECTOR-WIDE? Meridian is an outlier Meridian sector two years ago now 14% 27% a Meridian problem: look inside A sector-wide wave Meridian sector two years ago now 14% 27% a sector problem: a different remedy AT MERIDIAN Before surveying a single employee: what do Meridian's own records — and Poland's official statistics — already reveal?
Figure 4: Why managers should care: is the problem unusual, or sector-wide?

Sources of secondary data

Internal organisational sources

Most organisations hold far more data than they ever analyse. The main internal sources are summarised below. All were collected for administration rather than research, which is both their weakness (they record only what the process needed) and their innocence (nobody shaped them to please a researcher).

Internal source What it typically contains What it could show at Meridian
HR records Headcount, contracts, tenure, absence, pay grades, exits by store and month Who leaves: which stores, roles, tenure bands, and managers
Sales and CRM data Transactions, complaints, satisfaction scores, loyalty-card behaviour Whether high-turnover stores also show falling sales
Exit interviews Leavers’ stated reasons for departing Patterns in reasons given — if interviews were conducted consistently
Past employee surveys Engagement or satisfaction results from previous years, often CAWI Whether warning signs preceded the turnover rise — if the same questions were asked each year

Analysing any of these is secondary analysis: the marginal cost is analyst time, not fieldwork.

Figure 5 shows the four main internal sources and the double-edged character they share: collected for administration, they record only what the process needed, but nobody shaped them to please a researcher.

Internal sources: the organisation's own records Most organisations sit on more data than they ever analyse. Four internal sources, one at a time. HR records: headcount, contracts, tenure, absence, pay grades, and exits by store and month. Sales and CRM data: transactions, customer complaints, satisfaction scores, loyalty-card behaviour. Exit interviews: routinely conducted, rarely systematically analysed. Past employee surveys: engagement or satisfaction studies from previous years, often run online. At the bottom: all of them were collected for administration, not research. That is their weakness, because they record only what the process needed, and their innocence, because nobody shaped them to please a researcher. Most organisations sit on more data than they ever analyse HR records headcount contracts tenure absence pay grades exits by store and month Sales and CRM data transactions complaints satisfaction scores loyalty-card behaviour Exit interviews Routinely conducted — rarely systematically analysed: a qualitative dataset waiting for an analyst Past employee surveys Engagement or satisfaction studies from previous years, often run online (CAWI) — results on a shared drive COLLECTED FOR ADMINISTRATION, NOT RESEARCH THEIR WEAKNESS They record only what the process needed — your variable may not be there THEIR INNOCENCE Nobody shaped them to please a researcher: no social desirability in a rota log
Figure 5: Internal sources: the organisation’s own records.

Figure 6 applies them to Meridian. Each row pairs a source with the question it could answer for the board, and a caveat in yellow where one applies: exit interviews are comparable only if conducted consistently, past surveys only if the same questions were asked each year, and correlations with sales or rota data suggest a mechanism without proving one.

What Meridian's own records could tell the board At Meridian, before any new data are collected, four internal sources each answer a different question, with a caveat. HR records: who leaves, from which stores, roles, tenure bands and managers. Exit interviews: what reasons leavers give, but only if the interviews were conducted consistently enough to compare. Past engagement surveys: did warning signs appear before the turnover rise, but only if the same questions were asked each year. Sales and rota data: do high-turnover stores also show falling sales or heavier weekend scheduling? Such correlations suggest a mechanism but do not prove it. Every one of these is analysis of existing data: the marginal cost is analyst time, not fieldwork. AT MERIDIAN: WHAT COULD ITS OWN RECORDS TELL THE BOARD — BEFORE ANY NEW DATA? HR records Who leaves — which stores, roles, tenure bands, managers? Exit interviews What reasons do leavers give? only if conducted consistently enough to compare Past engagement surveys Did warning signs appear before the rise? only if the same questions were asked each year Sales and rota data Do high-turnover stores show falling sales or heavier weekend rotas? a correlation suggests a mechanism; it does not prove one NOTICE Every one of these is analysis of existing data. The marginal cost is analyst time, not fieldwork.
Figure 6: What Meridian’s own records could tell the board.

External sources

External sources fall into three broad groups. First, official statistics: Statistics Poland (GUS), the national statistical office, publishes employment, wage, and labour-demand data by sector and region, much of it through the Local Data Bank; Eurostat, the EU’s statistical office, publishes harmonised labour-market data (including the EU Labour Force Survey) comparable across member states; and the OECD publishes cross-national indicators on employment, job tenure, and earnings. These are collected under statutory mandates, with published methodologies and large, carefully designed samples. Their strength is credibility and comparability; their limit is that they describe sectors and countries, not individual firms.

Figure 7 places the three official sources on a scale from national to cross-national, names what unites them — a statutory mandate, a published methodology and large, carefully designed samples — and draws their limit: Polish retail is one figure made of thousands of firms, in which Meridian is a sliver. GUS can tell the board about Polish retail, and nothing about store 17.

External sources: official statistics Three official sources, from national to cross-national. Statistics Poland, GUS, the national statistical office: employment, wages and labour demand by sector and region, much of it in the Local Data Bank. Eurostat, the EU's statistical office: harmonised labour-market data, including the EU Labour Force Survey, comparable across member states. The OECD: cross-national indicators on employment, job tenure and earnings across rich democracies. What unites them: a statutory mandate, a published methodology, and large, carefully designed samples, which give them credibility and comparability. Their limit: they describe sectors and countries, not your firm. A bar of Polish retail made up of thousands of firms contains Meridian as one sliver: GUS can tell the board about Polish retail, and nothing about store 17. national European cross-national Statistics Poland (GUS) Employment, wages and labour demand by sector and region much of it in the Local Data Bank Eurostat Harmonised labour-market data, incl. the EU Labour Force Survey comparable across member states OECD Cross-national indicators: employment, job tenure, earnings across rich democracies WHAT UNITES THEM a statutory mandate a published methodology large, carefully designed samples credibility, comparability THEIR LIMIT They describe sectors and countries — not your firm Polish retail: thousands of firms in one figure Meridian GUS can tell the board about Polish retail — and nothing about store 17
Figure 7: External sources: official statistics.

Second, industry-association and consultancy reports: retail federations, HR consultancies, and job portals publish studies of turnover, pay, and engagement. They are often timely and sector-specific, but the reader should always ask who paid for the study and what the publisher is selling.

Third, large academic surveys, designed for reuse from the outset. The European Social Survey (ESS) measures attitudes and behaviour across Europe every two years; the European Values Study (EVS) is a long-running study of values, including attitudes to work. Academic surveys publish their questionnaires, sampling designs, and data files openly — the gold standard of documentation.

Figure 8 compares the remaining external sources and places all of them on a scale of documentation, from a headline and a chart to the full questionnaire, sampling design, response rates and data files: industry and consultancy reports vary across the lower half, official statistics sit high, and academic surveys such as the ESS and EVS sit at the top.

External sources: reports and academic surveys Two more external sources. Industry-association and consultancy reports, from retail federations, HR consultancies and job portals, are often timely and sector-specific, but you must ask who paid for them and what they are selling. Large academic surveys are designed for reuse from the outset: the European Social Survey, on attitudes and behaviour across Europe every two years, and the European Values Study, a long-running study of values including attitudes to work. Underneath, a scale of documentation runs from a headline and a chart to the questionnaire, sampling design, response rates and data files. Industry and consultancy reports spread across the low and middle of the scale; official statistics sit high, with published methodologies; academic surveys such as the ESS and EVS sit at the top: the gold standard of documentation. INDUSTRY AND CONSULTANCY REPORTS retail federations HR consultancies job portals + often timely and sector-specific − ask who paid for them — and what they are selling A job portal's report on how hard it is to keep staff is also an advert for a job portal. ACADEMIC SURVEYS, DESIGNED FOR REUSE European Social Survey (ESS) attitudes and behaviour across Europe, every two years European Values Study (EVS) a long-running study of values, incl. attitudes to work + questionnaires, sampling designs and data files published openly HOW WELL DOCUMENTED? a headline and a chart questionnaire, sampling design, response rates, data files industry and consultancy reports: it varies official statistics published methodology ESS, EVS the gold standard
Figure 8: External sources: reports and academic surveys, and how well each is documented.

Digital trace and big data

A newer category of secondary data is generated as a by-product of behaviour rather than by asking anyone anything: website clickstreams, till and loyalty-card transactions, staff rota and badge-swipe logs, social media posts, and online job-board activity. Such data are enormous in volume, fine-grained, and collected continuously, recording behaviour as it happens rather than as it is recalled. But they are found data, not designed data: there is no sampling frame, no consent for research, and coverage extends only to those who leave traces. Digital trace data are powerful raw material that demand more methodological caution, not less. A useful one-line summary: big data tells you what people did at scale, but is usually silent on why.

Figure 9 lists the kinds of trace that behaviour leaves, its attractions in green and its cautions in red, and ends with the contrast between what people did, which big data records at scale, and why, on which it is usually silent.

Digital trace and big data Digital trace data are found, not designed: behaviour leaves traces as a by-product, and nobody is asked anything. Examples: website clickstreams, till and loyalty-card transactions, staff rota and badge-swipe logs, social media posts, online job-board activity. Their attractions: enormous volume, fine grain, continuous collection, and behaviour recorded as it happens rather than recalled afterwards. But, being found rather than designed, they have no sampling frame, no consent for research, and they cover only those who leave traces, so they need more methodological caution, not less. The line to remember: big data tells you what people did, at scale; it is usually silent on why. FOUND DATA, NOT DESIGNED Behaviour leaves traces as a by-product — nobody is asked anything website clickstreams till and loyalty-card transactions staff rota and badge-swipe logs social media posts online job-board activity ATTRACTIONS enormous volume fine-grained collected continuously Behaviour recorded as it happens — not recalled afterwards, so no memory slips and no effort to look good BUT: FOUND, NOT DESIGNED no sampling frame no consent for research covers only those who leave traces needs more methodological caution, not less WHAT people did at scale, in fine detail big data tells you this WHY they did it motives, meanings, intentions it is usually silent on this
Figure 9: Digital trace and big data: found, not designed.

Benefits and risks

The benefits of secondary analysis are chiefly practical; the risks are chiefly methodological. Secondary analysis trades control for convenience.

Benefits

The first three benefits are the ones that are most appealing to managers. Cost: the most expensive phases of research — sampling, fieldwork and data entry — have already been paid for by someone else. Speed: analysis can begin in days, whereas a bespoke survey takes months to design, field and clean. Scale: national statistical offices and academic surveys achieve sample sizes and coverage no single organisation could fund. For a student, a start-up or a junior research team, secondary analysis therefore makes otherwise impossible studies feasible.

Figure 10 draws the three: the stages of a study with the expensive middle already paid for, the months of a bespoke survey against days for secondary data, and a national survey’s sample beside one organisation’s own.

Benefits of secondary analysis: cost, speed, scale Three benefits, one panel each. Cost: of the stages of a study, the expensive ones, sampling, fieldwork and data entry, have already been paid for by someone else; your cost is analyst time. Speed: a bespoke survey takes months to design, field and clean before analysis begins; with secondary data, analysis can begin in days. Scale: national statistical offices interview tens of thousands of households, and academic surveys cover whole countries, samples no single organisation could fund. At the bottom: for a student, a start-up or a junior research team, secondary analysis makes otherwise impossible studies feasible. COST Design Sampling Fieldwork Data entry Analysis already paid for by someone else your cost: analyst time SPEED A bespoke survey design field clean months before the first analysis Secondary data analysis begins in days When a board wants preliminary answers by the next meeting, the difference decides SCALE national surveys: tens of thousands one organisation's own survey samples no single organisation could fund FOR A STUDENT, A START-UP OR A JUNIOR RESEARCH TEAM Secondary analysis makes otherwise impossible studies feasible: you cannot commission a national labour force survey — but you can download one.
Figure 10: Benefits: cost, speed, scale.

Two further benefits matter for particular questions. Longitudinal reach: existing series stretch back years or decades, so they allow the study of change — trends, turning points, before-and-after comparisons — and no new study, however lavish, can go back in time and measure last decade. No participant burden: nobody is interrupted, surveyed or observed again, which is ethically lighter (no new consent to seek, no fatigue imposed on over-surveyed employees) and practically valuable at Meridian, where staff already distrust head-office questionnaires. But none of these benefits answers the decisive question: were these data collected in a way that can answer your question?

Figure 11 shows an existing series running across the past decade, where a new study could only begin today, and the people a secondary analysis does not have to ask again — then the decisive question that no benefit answers.

Benefits of secondary analysis: reach and burden Two further benefits. Longitudinal reach: existing series stretch back years or decades, so you can study change, trends, turning points, before-and-after comparisons; a line runs across the past decade, while a new study can only begin today, because no new study can go back in time and measure last decade. No participant burden: nobody is interrupted, surveyed or observed again. That is ethically lighter, with no new consent to seek and no fatigue imposed on over-surveyed employees, and at Meridian, where staff already distrust head-office questionnaires, it saves the one good chance at a survey for the questions that need it. Then the decisive question, which none of these benefits answers: were these data collected in a way that can answer your question? LONGITUDINAL REACH Existing series let you study change: trends, turning points, before-and-after comparisons a decade ago today an existing series a new study starts here No new study, however lavish, can go back and measure last decade. NO PARTICIPANT BURDEN Nobody is interrupted, surveyed or observed again Ethically lighter: no new consent to seek, no fatigue for over-surveyed staff Meridian's staff already distrust head-office questionnaires BUT NONE OF THESE BENEFITS ANSWERS THE DECISIVE QUESTION Were these data collected in a way that can answer your question?
Figure 11: Benefits: reach and burden.

Risks

The risks all flow from the fact that the design decisions were made by someone else, for someone else’s purpose. Fit: the variable you need may be missing, or measured in a way that misses your concept — a national “job satisfaction” item is not a measure of Meridian employees’ satisfaction. Fit fails in three typical ways: the right name with the wrong content, the right content with the wrong population, or the right population in the wrong year. Unknown quality: response rates, interviewer conduct and data cleaning were outside your control, and sometimes outside your knowledge, so validity and reliability must be assessed at second hand, through the documentation — if it exists.

Figure 12 draws fit as two pieces that do not join, lists the three ways it fails, and puts the unknowns of quality behind a screen.

Risks of secondary analysis: fit and quality Two risks. Fit to your question: the data were collected for someone else's purpose, so the variable you need may be missing, or measured in a way that misses your concept; drawn as two shapes that do not fit. A national job-satisfaction item is not a measure of Meridian employees' satisfaction. Fit fails in three ways: the right name but the wrong content, the right content but the wrong population, the right population but the wrong year. Unknown quality: you did not oversee the collection, so response rates, interviewer conduct and data cleaning sit behind a screen, outside your control and sometimes outside your knowledge. At the bottom: validity and reliability must be assessed at second hand, through the documentation, if it exists. FIT TO YOUR QUESTION the variable you need what was measured ≠ a national “job satisfaction” item is not Meridian employees' satisfaction Fit fails in three ways: right name, wrong content right content, wrong population right population, wrong year UNKNOWN QUALITY You did not oversee the collection: response rates? interviewer conduct? data cleaning? outside your control — and sometimes outside your knowledge SO Validity and reliability must be assessed at second hand — through the documentation, if it exists
Figure 12: Risks: fit and quality.

Three further risks deserve elaboration.

Unit-of-analysis mismatch. The data may describe one level (the sector, the country, the region) while your question concerns another (one firm, one store, one team). Retail turnover for Poland as a whole cannot reveal whether one store’s manager is the problem, and inferring individual-level facts from aggregate data is the road to the ecological fallacy.

Missing documentation. Without the questionnaire, the sampling design, and the definitions, you cannot judge what the numbers mean. An undocumented dataset is a rumour with decimal places.

Figure 13 shows both risks. On the left, a ladder of levels from country to team: the data describe the sector, the question concerns one store, and sliding from one to the other is the ecological fallacy. On the right, a report with a headline figure and a chart but no questionnaire, sampling design or definitions.

Risks of secondary analysis: mismatched units and missing documentation Two more risks. Unit-of-analysis mismatch: a ladder of levels runs from country, to sector, to firm, to store, to team. The data describe one level, here the sector; the question concerns another, here one store. Sliding from the sector to the store is the ecological fallacy, inferring facts about lower-level units from aggregates: retail turnover for Poland as a whole cannot tell you whether store 17's manager is the problem. Missing documentation: a report shows a headline figure, retail turnover hits 31 per cent, and a chart, but the questionnaire, the sampling design and the definitions are missing. Without them you cannot judge what the number means: an undocumented dataset is a rumour with decimal places. UNIT-OF-ANALYSIS MISMATCH Country Sector Firm Store Team the data describe this your question concerns this ✗ The ecological fallacy inferring facts about stores, teams or people from aggregate data Polish retail turnover cannot tell you whether store 17's manager is the problem MISSING DOCUMENTATION Retail turnover hits 31% questionnaire? sampling design? definitions? …and one sentence about an “online survey of employees” An undocumented dataset is a rumour with decimal places.
Figure 13: Risks: mismatched units and missing documentation.

Changing definitions. Long time series are the great gift of secondary data, and definitions quietly shifting underneath them are the great trap. Statistical offices revise classifications and methods; what counts as “employed”, “vacancy”, or “retail” can change between years. Organisations do the same: if Meridian redefined “leaver” two years ago — say, from voluntary quits only to all departures including seasonal contract expiries — part of the rise from 14% to 27% may be an artefact of the definition. The rule: before interpreting any trend, ask whether the measure changed before believing the world did.

Figure 14 illustrates the trap with Meridian’s turnover. If “leaver” was redefined early in the two years, the jump at the break (yellow) may be an artefact of the definition, and only the rest of the rise (green) is a candidate for real change. The chart is illustrative, not real data.

Risks of secondary analysis: definitions change over time A schematic chart of Meridian's turnover over two years, rising from 14 to 27 per cent. A dashed line marks a moment when the HR system was updated and the definition of a leaver changed: before it, voluntary quits only; after it, all departures, including the planned expiry of seasonal contracts. The jump at the break may be an artefact of the definition, not a change in the world; only the rest of the rise is a candidate for real change. Statistical offices revise their classifications too: what counts as employed, a vacancy, or retail can change between years. The rule: before interpreting any trend, ask whether the measure changed before believing the world did. The chart is illustrative. AN ILLUSTRATION: MERIDIAN'S TURNOVER 10% 20% 30% two years ago now 14% 27% before: voluntary quits only after: all departures, incl. seasonal contract expiries “leaver” redefined an artefact of the definition? a real change? STATISTICAL OFFICES DO IT TOO Classifications and methods are revised, for good reasons. What counts as… “employed” a “vacancy” “retail” …can change between years — and a naive reading takes the break for a real-world event. THE RULE Before interpreting any trend, ask whether the measure changed — before believing the world did.
Figure 14: Risks: definitions change over time.

The balance

Figure 15 pairs each practical benefit with a methodological risk, row by row, and names the pattern: secondary analysis trades control for convenience. The fitness-for-purpose checklist is how that trade is audited.

Benefits and risks of secondary analysis: a summary A ledger of five paired rows. Benefits on the left, all practical: low cost, because collection was already paid for; speed, because analysis can start at once; scale, samples you could never fund; longitudinal reach, so you can study change over time; no participant burden, which is ethically lighter. Risks on the right, all methodological: fit, because the data were collected for someone else's question; quality, because collection was outside your control; unit of analysis, possibly the wrong level for your question; documentation, which may be thin or absent; definitions, which may shift over time. The pattern: secondary analysis trades control for convenience, and the checklist that follows is how you audit that trade. Benefits — practical Risks — methodological Low cost collection already paid for Speed analysis can start at once Scale samples you could never fund Longitudinal reach study change over time No participant burden ethically lighter Fit collected for someone else's question ↔ Quality collection outside your control ↔ Unit of analysis wrong level for your question ↔ Documentation may be thin or absent ↔ Definitions may shift over time ↔ THE PATTERN Secondary analysis trades control for convenience the checklist that follows is how you audit that trade
Figure 15: Benefits and risks of secondary analysis: a summary.

In full, the five pairs are:

Benefits Risks
Cost — the expensive phases (sampling, fieldwork, data entry) have already been paid for Fit — the data were collected for someone else’s question; the variable you need may be missing or measured differently
Speed — analysis can begin in days rather than months Unknown quality — response rates, interviewer conduct, and cleaning were outside your control
Scale — sample sizes and coverage no single organisation could fund Unit-of-analysis mismatch — data at the wrong level for your question
Longitudinal reach — series stretching back years or decades permit the study of change Missing documentation — without the instrument and definitions, numbers cannot be judged
No participant burden — nobody is surveyed or observed again; ethically lighter Changing definitions — measures may shift over time, producing artefactual trends

Assessing fitness for purpose

Before using any secondary dataset, interrogate it like a witness. The checklist has seven questions:

Question What to establish
Who collected it? Competence and independence: statistical office, university consortium, industry body, consultancy, vendor
Why was it collected? Purpose shapes design — data collected to administer, to persuade, and to research have different properties
When was it collected? Currency (a 2019 study describes a pre-pandemic labour market) and consistency of measurement across waves
From whom? Population, sampling method, and achieved response rate; beware self-selected panels and hidden composition in aggregates
How? Mode and procedure — PAPI, CAWI, CATI, administrative records, or digital trace — each with characteristic biases
With what instrument? The exact question wording; similar labels can hide different variables
Under what definitions? How key concepts are defined, and whether the definitions have changed — the decisive question for comparability

If the documentation cannot answer these questions, that silence is itself a finding.

Figure 16 groups the seven questions in the order of a study’s anatomy: provenance (who, why), sampling (when, from whom) and measurement (how, with what instrument, under what definitions).

The fitness-for-purpose checklist Interrogate any secondary dataset like a witness, with seven questions in three groups that follow the anatomy of a study. Provenance: one, who collected it, and are they competent and independent? Two, why, for what purpose and with what interest in the result? Sampling: three, when, is it current enough and consistent over time? Four, from whom, what population, sample and response rate? Measurement: five, how, by what mode and procedure: PAPI, CAWI, CATI, administrative records or digital trace? Six, with what instrument, what exact questions or measures? Seven, under what definitions, how are the key concepts defined, and have the definitions changed? At the bottom: if the documentation cannot answer these questions, that silence is itself a finding. Interrogate any dataset like a witness: PROVENANCE 1 Who collected it — competent and independent? 2 Why for what purpose — with what interest in the result? SAMPLING 3 When current enough — and consistent over time? 4 From whom what population, what sample, what response rate? MEASUREMENT 5 How mode and procedure: PAPI, CAWI, CATI, records, digital trace 6 With what instrument the exact questions or measures 7 Under what definitions how key concepts are defined — and whether that changed If the documentation cannot answer, that silence is itself a finding
Figure 16: The fitness-for-purpose checklist.

Figure 17 takes the provenance questions. Collectors range from bodies with no stake in the result to vendors with a product to sell; and purpose shapes design, whether the data were collected to administer, to persuade or to research.

Who collected the data, and why Who: collectors range from a national statistical office and a university consortium, through an industry body and a consultancy, to a vendor with a product to sell; each has different competence and different incentives. Why: purpose shapes design. Data collected to administer, such as payroll and till records, record what the process needed and no more. Data collected to persuade, as in some consultancy reports, may be sampled and worded conveniently. Data collected to research, such as GUS, Eurostat or the ESS, are designed for inference, but for someone else's inference. The quick test: would the collector be embarrassed by any particular result? If yes, read the methodology section twice. WHO COLLECTED IT? national statistical office university consortium industry body consultancy vendor with a product to sell no stake in the result a stake in the result different competence, different incentives WHY WAS IT COLLECTED? PURPOSE SHAPES DESIGN To administer payroll, tills, rotas Records what the process needed — no more To persuade some consultancy reports May be sampled and worded conveniently To research GUS, Eurostat, ESS Designed for inference — but for someone else's A QUICK TEST Would the collector be embarrassed by any particular result? If yes, read the methodology section twice.
Figure 17: Who collected the data, and why.

Purpose deserves particular attention: data collected to administer (payroll, tills) record only what the process needed; data collected to persuade (some consultancy reports) may sample and word questions conveniently; data collected to research (GUS, Eurostat, ESS) are designed for inference — but for someone else’s inference. A quick test of independence: would the collector be embarrassed by any particular result? If so, read the methodology section twice.

Definitions decide whether comparison is possible at all. Does “turnover” mean all separations, voluntary quits, or replacements hired? Does “retail” include e-commerce fulfilment? Does “employee” include agency staff? Only when definitions match can Meridian’s 27% be compared with anyone else’s figure.

Figure 18 takes the sampling questions. Vintage matters (a 2019 study describes a pre-pandemic labour market), and trends need the same question, definition and population in every wave. From whom means the population, the sample and the response rate — with two traps: a self-selected sample dressed as a general one, and an average that hides its composition.

When the data were collected, and from whom When: vintage is substantive, not a footnote: a 2019 study of retail employment describes a pre-pandemic labour market. For trends, you need consistent measurement across waves, the same question, the same definition, the same population, every time; a series that changed its question in the middle is two series wearing one label. From whom: identify the population, the sampling method and the achieved response rate. Two traps. A national survey of employees recruited from one job portal's users is a survey of that portal's users. And aggregates hide composition: an average retail wage mixes hypermarkets, boutiques and e-commerce warehouses, and may describe none of them. WHEN? 2019 2020 now a 2019 retail study describes a pre-pandemic labour market: vintage is substantive, not a footnote For trends: consistent measurement across waves wave 1 wave 2 wave 3 same question ✓ ✓ ✗ same definition ✓ ✓ ✓ same population ✓ ✓ ✓ A question changed in wave 3? Then it is two series wearing one label. FROM WHOM? population sample responses response rate = responses ÷ sample Trap 1: a self-selected sample dressed as a general one employees A “national survey of employees” recruited from one job portal is a survey of that portal's users Trap 2: aggregates hide composition hypermarkets boutiques e-commerce warehouses one “average retail wage” — which may describe none of them
Figure 18: When the data were collected, and from whom.

Figure 19 takes the measurement questions: the mode and its biases, the exact wording (two items reported under one label can be different variables), and the definitions that decide whether any comparison is possible.

How, with what instrument, under what definitions Three measurement questions, one row each. How: the mode shapes the answers: CAWI panels, CATI interviews and administrative registers each have characteristic biases; face-to-face and telephone modes invite socially desirable answers, more in session 8. With what instrument: get the exact wording, because two items reported under the same label, are you satisfied with your job, and would you recommend working here, are different variables wearing similar labels. Under what definitions, the decisive question for comparability: does turnover mean all separations, voluntary quits, or replacements hired? Does retail include e-commerce fulfilment? Does employee include agency staff? Only when definitions match can you compare Meridian's 27 per cent with anyone else's number. How? the mode CAWI panels CATI interviews administrative registers Each mode has characteristic biases: face-to-face and telephone modes invite socially desirable answers — more in session 8. With what instrument? exact wording “Are you satisfied with your job?” reported as “employee satisfaction” “Would you recommend working here?” also reported as “employee satisfaction” ≠ Different variables wearing similar labels — no exact wording, no comparison Under what definitions? the decisive question for comparability “turnover” all separations? voluntary quits? replacements hired? “retail” incl. e-commerce fulfilment? “employee” incl. agency staff? incl. civil-law contracts? Only when definitions match can Meridian's 27% be compared with anyone else's number
Figure 19: How, with what instrument, under what definitions.

The checklist applied

Suppose the team finds an industry report claiming that “retail staff turnover in Poland averages 31%” — apparently reassuring for Meridian’s board. The checklist deflates it: the publisher is a recruitment agency that profits from portraying turnover as universal (who/why); the sample is a self-selected online panel of 400 HR managers with no reported response rate (from whom); and the report’s definition of turnover includes planned seasonal contract expiries, which Meridian’s figure excludes (definitions). The comparison collapses, because the 31% and the 27% are not measuring the same thing. A number without its methodology is not evidence; fitness for purpose is established, never assumed.

Figure 20 runs the checklist on the 31% figure: three questions, three failures, and a comparison that collapses.

Applying the checklist at Meridian An industry report says retail staff turnover in Poland averages 31 per cent. Set beside Meridian's 27 per cent, it looks reassuring: the board's relief is that the problem must be normal. Then the checklist, one question at a time. Who and why: the report comes from a recruitment agency, which profits from portraying turnover as universal. From whom: an online panel of 400 HR managers, self-selected, with no reported response rate. Definitions: the report counts planned seasonal contract expiries, which Meridian's figure excludes. The comparison collapses: the 31 and the 27 per cent are not measuring the same thing. The lesson: a number without its methodology is not evidence; fitness for purpose is established, never assumed. THE COMPARISON 31% industry report 27% Meridian The board, relieved: “We're below average — the problem is normal!” Who and why ✗ A recruitment agency — it profits when turnover looks universal From whom ✗ An online panel of 400 HR managers, self-selected, response rate unreported Definitions ✗ Counts seasonal contract expiries — Meridian's 27% excludes them The comparison collapses: 31% and 27% are not measuring the same thing THE LESSON A number without its methodology is not evidence: fitness for purpose is established, never assumed
Figure 20: Applying the checklist at Meridian.

Combining secondary and primary research

In organisational diagnosis the efficient sequence is almost always to exhaust the existing data before collecting new data. Secondary analysis does three preparatory jobs: it establishes context (is the problem unusual or sector-wide?), it sharpens the question (internal records show where and among whom the problem sits, so primary research can ask why), and it provides benchmarks against which the organisation’s own results will be judged. It also disciplines spending: never pay to collect what already exists.

Figure 21 shows the sequence — exhaust the existing data, then collect new data — and the three preparatory jobs that make the primary study smaller, sharper and cheaper.

Secondary first, primary second In organisational diagnosis the efficient sequence is to exhaust the existing data before collecting new data. Secondary analysis does three preparatory jobs for the primary research that follows. It establishes context: is the problem unusual, or sector-wide? It sharpens the question: internal records show where and among whom the problem sits, so primary research can ask why. It provides benchmarks: the yardstick against which your own results will be judged, as when a survey finds 40 per cent of staff dissatisfied with scheduling and the board asks whether that is a lot. All three feed a primary study that is smaller, sharper and cheaper. And it disciplines spending: never pay to collect what already exists. 1 Exhaust the existing data 2 then collect new data Establishes context Is the problem unusual — or sector-wide? national retail figures vs Meridian's 27% Sharpens the question Records show where and among whom — so primary research can ask why the rise sits in six stores with one regional manager Provides benchmarks The yardstick for your own results “40% dissatisfied with scheduling — is that a lot?” A primary study that is smaller, sharper and cheaper than it would otherwise be Never pay to collect what already exists
Figure 21: Secondary first, primary second.

Secondary data nonetheless have hard limits. They cannot answer questions that nobody’s data address — no public dataset contains Meridian employees’ reasons for leaving. They cannot rescue a mismatched unit of analysis — sector aggregates will never identify which stores, teams, or managers drive a problem. They cannot reveal meanings and motives — administrative records show that people left, not what leaving meant to them. And they inherit every flaw of the original collection permanently: no reanalysis can repair a biased sample or a leading question. The question is therefore never secondary or primary? but what must the primary study still do, once the secondary work is done?

Figure 22 pairs each of the four limits with what the primary study must therefore do.

What secondary data cannot do Four hard limits, each pointing at what the primary study must still do. Secondary data cannot answer questions nobody's data address: no public dataset contains Meridian employees' reasons for leaving, so ask Meridian's own people. They cannot rescue a mismatched unit of analysis: sector aggregates never resolve into stores, teams or managers, so collect data at the level of the question. They cannot reach meanings and motives: records show that people left, not what leaving meant to them, which calls for qualitative methods. And they inherit every flaw of the original collection, permanently: no reanalysis repairs a biased sample or a leading question, so design your own sample and instrument. The question is never secondary or primary, but what must the primary study still do, once the secondary work is done? Secondary data cannot… …so the primary study must answer questions nobody's data address no dataset contains Meridian employees' reasons for leaving ask Meridian's own people rescue a mismatched unit of analysis sector aggregates never resolve into stores, teams, managers collect at the level of the question reach meanings and motives records show that people left — not what leaving meant use qualitative methods (session 7) escape inherited flaws, permanently no reanalysis repairs a biased sample or a leading question design its own sample and instrument Never secondary or primary? but what must the primary study still do, once the secondary work is done?
Figure 22: What secondary data cannot do.

A defensible combined design for Meridian proceeds in four stages, each cheaper than the next and each making the next one smarter:

  1. Internal secondary analysis — analyse HR records: turnover by store, role, tenure, and manager; check whether the definition of “leaver” changed.
  2. External secondary analysis — benchmark against GUS and Eurostat retail employment and wage data: is Meridian an outlier, or riding a sector-wide wave?
  3. Primary qualitative research — interviews with recent leavers from the worst-affected stores, targeting the why that records cannot reach.
  4. Primary quantitative research — a survey of current staff, designed around what stages 1–3 revealed.

Figure 23 draws the four stages as a staircase: each costs more than the one before, and each tells the next whom to ask or what to ask.

A combined design for Meridian A four-stage research plan for Meridian's board, drawn as a staircase in which each stage costs more than the one before. Stage one, internal secondary analysis: HR records give turnover by store, role, tenure and manager, and first of all show whether the definition of a leaver changed. Stage two, external secondary analysis: a benchmark against GUS and Eurostat retail employment and wage data shows whether Meridian is an outlier or riding a sector-wide wave. Stage three, primary qualitative research: interviews with recent leavers from the worst-affected stores, targeting the why that the records cannot reach. Stage four, primary quantitative research: a survey of current staff, designed around what stages one to three revealed. The records tell the interviews whom to talk to; the interviews tell the survey what to ask. Each stage is cheaper than the next, and each makes the next one smarter. 1 Internal secondary HR records: turnover by store, role, tenure, manager — first, was “leaver” redefined? 2 External secondary Benchmark against GUS / Eurostat retail data: an outlier, or a sector-wide wave? 3 Primary qualitative Interviews with recent leavers from the worst-hit stores: the why 4 Primary quantitative A survey of current staff, designed around what stages 1–3 revealed the records tell the interviews whom to ask; the interviews tell the survey what to ask Each stage is cheaper than the next — and makes the next one smarter
Figure 23: A combined design for Meridian.

Conclusion

Secondary analysis reuses data, not findings — that is what separates it from the literature review and what makes it a data collection method in its own right. Its sources span the organisation’s own records (HR files, sales and CRM data, exit interviews, past surveys) and a rich external ecosystem (GUS, Eurostat, OECD, industry reports, and academic surveys such as the ESS and EVS). Its benefits — cost, speed, scale, longitudinal reach, and the absence of participant burden — are bought by surrendering control over design. The fitness-for-purpose checklist — who, why, when, from whom, how, with what instrument, under what definitions — is how a researcher decides whether someone else’s data can support their conclusions.