How to Conduct Secondary Research: Process, Techniques & Best Practices

What you might not know is that according to the International Association of Scientific, Technical and Medical Publishers, around 3 million articles were published in more than 33,000 peer-reviewed English-language journals in 2018 alone (Johnson et al., 2018). However, most students spend weeks collecting original info and are looking for answers in verified, peer-reviewed sources. Whether you are writing a dissertation, business report, or policy analysis, secondary research is the quickest way to a credible and evidence-based conclusion. Yet, most students get it wrong by confusing simple background reading with secondary research. When done properly, it is a systematic, reproducible process for collecting, reviewing, and analysing existing data to answer new research questions. The ability to use and evaluate the quality of the sources, as well as analyse what they contain, is one of the most transferable skills for academic and professional life. This guide will walk you through the entire secondary research process, optimal methods, and the professional best practices that will distinguish your work.

Key takeaways

• Secondary research (also called desk research) analyses data that someone else has already collected, so you can answer a new question without running your own surveys, interviews or experiments.
• It is the standard method behind literature reviews, systematic reviews, meta-analyses and secondary data analysis of national datasets such as those held by the ONS and the UK Data Service.
• A defensible process has seven steps: define the question, choose source types, search systematically, check access and ethics conditions, evaluate each source, extract and organise the data, then analyse and write up.
• Examiners judge secondary research on how transparently you document the search, the data's original purpose and its limitations, not on how many sources you list.
• Most UK universities still require an ethics form for secondary research, even when the data is anonymised and publicly available.

What is Secondary Research?

Secondary research is data compiled by others for different purposes. Also known as desk research or secondary data analysis, secondary research is the systematic collection, evaluation and analysis of information that already exists in order to answer a new research question. Instead of conducting new research such as interviews, surveys or experiments (primary research), you will use existing sources, including published studies, government data, institutional reports, statistical databases, historical documents, and academic publications. More than 600 datasets are released every year by the UK's Office for National Statistics, of which the majority are never fully analysed by researchers, presenting a huge scope for secondary research (ONS, 2021).

This is important because it impacts the whole approach to your methodology, the sources you can cite, the questions you can ask, the restrictions you need to recognise, and the conclusions you can make. Secondary research does not involve the creation of new empirical data. What it can do is combine what you already know, identify trends across studies, analyse large datasets that would not realistically be collected by a single person, and build the evidence base for further research.

In the academic setting, secondary research is used for the literature review, which is the part of a dissertation or research paper that outlines what is already known and what is needed for your research. In the business world, it facilitates market analysis, collecting competitive intelligence and strategic planning. In policy research, it provides the government with information for decision-making based on demographic data, economic statistics, and programme evaluations.

Types of Secondary Research

Secondary research is not one method but a family of them. Which one you use depends on whether your existing material is text or numbers, and on how formal your review needs to be.

Literature review. The most common form in undergraduate and master's work. You read, compare and synthesise published studies to establish what is already known, where the evidence conflicts and where the gap is that your project fills. A narrative literature review is flexible; a structured one follows a documented search protocol. The same synthesis skill underpins the literature section of any paper; see how to write a research paper step by step for how it fits into a shorter piece.

Systematic review and meta-analysis. A systematic review answers a tightly focused question by searching several databases with pre-registered inclusion and exclusion criteria, appraising every included study for quality and reporting the whole process, usually against the PRISMA checklist. A meta-analysis goes one step further and statistically pools the results of comparable studies into a single effect estimate. Both sit at the top of the evidence hierarchy in health, psychology and education.

Secondary data analysis. Here the raw material is a numerical dataset rather than published papers: a national survey, census extract, company financials or an administrative record set. You run your own statistical analysis on data collected by someone else, which is why this method is popular with quantitative dissertations in economics, sociology, business and public health.

Content and documentary analysis. Qualitative secondary research that codes existing texts, speeches, policy documents, media coverage, archived interview transcripts or social media posts for recurring themes. It can be counted (how often a term appears) or interpreted (what the language reveals about attitudes).

Case study and scoping review. A case study built from existing records examines one organisation, event or policy in depth. A scoping review maps how much research exists on a broad topic and where the gaps are, without appraising every study in detail, which makes it a common choice when a topic is too new or too wide for a full systematic review.

Secondary Research vs Primary Research

It is important to distinguish between secondary and primary research when planning any study. Primary research is research that involves gathering original data directly from the source (interviews, experiments, surveys, or observations). The researcher bases their design, sample, and data collection methods on the question they want to answer, and controls every stage of that process. Secondary research involves data from elsewhere. The researcher is not in control of the manner in which the original data has been collected and thus obtains both the attributes and the shortcomings of that data (Tantawi, 2021). For instance, a government census gathers nationally representative data that no researcher can replicate, but it restricts the analysis strictly to the questions the census designers chose to ask. Similarly, a published meta-analysis combines the results of several dozen studies and reflects the methodology of the original investigators.

This is not to say that secondary research is bad; it's a different kind of research. Secondary data is used to answer many research questions. Looking at trends in unemployment over time, secondary data, such as that provided by the Office for National Statistics (ONS), is much better suited than any primary data that you could gather. Primary research is essential if you wish to gain insight into the experiences of unemployment in a particular community.

At a glance: primary vs secondary research

When should you use secondary research?

Secondary research is the right choice when the question is about patterns rather than individual experiences, when the data you need already exists at a scale you could never collect yourself, when time or budget rules out fieldwork, or when your department will not grant ethical approval for primary work with a vulnerable group. It is the wrong choice when no dataset measures the thing you care about, or when the existing evidence is so dated that it no longer describes the world you are writing about. Many strong dissertations combine both: a secondary analysis to establish the pattern, followed by a small primary study to explain it.

Secondary Research Sources: Where to Look

The quality of the secondary research will be directly related to the quality of the sources you use. Not every piece of available information is reliable, up-to-date, or pertinent, and the ability to locate and assess reliable sources is essential for research. In most subjects, the best quality secondary sources are academic journals and peer-reviewed publications. Web databases such as JSTOR, Scopus, Web of Science and Google Scholar provide you with millions of peer-reviewed articles from all subject areas. Google Scholar is currently the source for more than 389 million documents, and Scopus has more than 90 million documents across 200+ disciplines. Systematic database searching is an important ability for all researchers. These databases are used by most UK universities for free via their library portals, rather than searching the general internet.

Government data is among the most reliable secondary research data available, especially in social, economic, and demographic research. The Office for National Statistics (ONS), the Home Office, the Department for Education, and the National Health Service (NHS) all have their own websites that feature a wealth of open-access data in the UK. These sources have high credibility because they are collected systematically at the national level using a transparent methodology.

Official secondary data sources, such as the World Health Organisation, the World Bank, the OECD, CIPD, and Think Tanks worldwide, produce institutional and organisational reports on issues related to the international community and sectors of interest. These are especially good for business and policy research.

There are online information databases designed specifically for research that collate data from a variety of sources and are searchable by topic, date and geographical area, such as Statista, the UK Data Service and EBSCO. They can be effective places to begin in a new topic area and quickly get an overview of where resources can be found for this topic.

Secondary research sources include historical records, government archives, newspapers and published books. For historical or qualitative research, these may be primary sources, but in the context of an understanding of existing knowledge about a topic, they become secondary sources of data.

UK datasets students can use for a secondary-research dissertation

If you are a UK student, start with the UK Data Service. It holds more than 10,000 data collections and you can register with your university login. Its data sits in three access tiers: open data, which anyone can download; safeguarded data, which needs registration and acceptance of an End User Licence (no sharing, no commercial use, correct citation); and controlled data, which can only be analysed inside its SecureLab environment after accreditation. For most dissertations, open and safeguarded data are more than enough.

Datasets that come with quick-start teaching guides include the British Social Attitudes survey, the Health Survey for England, the Community Life Survey and the Understanding Society teaching dataset. The ONS publishes the Labour Force Survey, Census 2021 tables and the Annual Population Survey. Beyond the UK, Eurostat, World Bank Open Data and the OECD data portal give comparable indicators for cross-country work.

Two other categories are easy to overlook. Grey literature (government evaluations, NHS and charity reports, think-tank papers from bodies such as the Institute for Fiscal Studies, working papers and conference proceedings) is not peer reviewed but is often the only source of recent, applied evidence, so appraise it carefully and say that you have. Internal data (a company's sales records, HR figures or customer feedback) is the backbone of business and management dissertations written with an employer's permission, and is treated as secondary data because it was collected for operational reasons, not for your study.

The Secondary Research Process: Step by Step

To make effective and valuable use of secondary research, a systematic process must be followed. Reading aimlessly and praying you bump into something useful is not a strategy; it's a time sink. A structured search strategy makes your data collection reproducible, eliminates selection bias, and immediately elevates the credibility of your work.

Here is an established framework to execute secondary research like a pro (seven steps; the two on access conditions and data documentation are the ones students most often skip):

Step 1: Establish a research question first.

Without a specific question, the information can become a mass rather than evidence. Your question should be specific and direct enough to direct the choice of source and should be limited so that you can answer the question within the scope of your research. "What factors are affecting the employment of graduates in the UK?" is a research question that can be answered and is searchable. "Employment" is not. If your question is still fuzzy, writing a short proposal usually sharpens it; this guide to what makes a good research proposal for a dissertation shows how the question, aims and sources should line up before you search.

Step 2: Determine which of the types of sources your question is most likely to be answered by.

Various kinds of questions require different sources. The question about public health trends requires government health data and epidemiological studies. There is a need for market research reports and psychological literature to address a question on consumer behaviour. Business case studies and management journals are required for a question on organisational strategy. By matching the source type to the question you're asking, you save considerable time and get more relevant results. There is a companion guide on how to conduct research to support your dissertation topic that goes deeper into choosing sources for each chapter.

Step 3: Search systematically using keywords and Boolean operators.

For academic databases, use exact wording in a search and Boolean operators (AND, OR, NOT) to narrow your search. Searching for "secondary research AND dissertation AND methodology" will return more relevant results than searching for "secondary research". Maintain a log of all search terms used, for replication and transparency of the process. Set date limits and language filters deliberately, and search at least two databases, because no single index covers every journal in a field. If you are doing a systematic review, record the number of results at each stage so you can draw a PRISMA flow diagram later.

Step 4: Check access conditions and ethics requirements before you download anything.

Secondary research is often described as having "no ethics hurdle". That is not quite true. Most UK universities still ask you to complete an ethics form, even if the outcome is a light-touch approval, and a full review is normally required where the data contains personal information about identifiable living people. Three checks save trouble later: confirm that your intended use falls within the licence or End User Licence attached to the dataset; check whether the original participants' consent covers re-use for research like yours; and make sure that combining two anonymised sources could not re-identify anyone. Data protection law (UK GDPR) applies to secondary data exactly as it does to primary data.

Step 5: Critically assess all sources before using.

Use the CRAAP test (Currency, Relevance, Authority, Accuracy, and Purpose) to determine if a source is suitable for a study.

Currency asks whether the information is up to date for the topic.

Relevance questions whether it is directly answering your research question.

Authority would inquire if the source is credible and if the author(s) have the appropriate qualifications.

Accuracy is a question of evidence supporting the claims.

Purpose constitutes whether there are any prejudices or interests that can impact the aim of the data.

For numerical datasets, add three questions the CRAAP test does not ask: how was the sample drawn and is it representative of the population you are writing about; how were the key variables defined and measured; and how much data is missing and why. The answers live in the dataset's documentation, sometimes called the codebook or user guide, and quoting them in your methodology is what turns a download into a defensible method.

Step 6: Extract and arrange your data in an orderly fashion.

When using your sources, make structured notes to document the key findings, the method used to create the key findings, the date the source was published, and a critical assessment of the strengths and weaknesses of your source. When you start writing, it will be much easier to put all the information you found into a cohesive document, since your notes are organised by topic and not by the source.

A simple extraction sheet with one row per source and columns for author and year, country, sample, method, key finding, limitation and the theme it belongs to is enough for most dissertations. Reference-management software such as Zotero, Mendeley or EndNote will store the full citation alongside your notes and format the reference list in Harvard, APA or whichever style your department requires; see this network guide to in-text citations in APA, MLA, Harvard and Chicago and the citation styles guide for UK universities for the conventions.

Step 7: Analyse, synthesise and write up.

The final step is covered in the next two sections: analysing what you have found, and presenting it in a methodology chapter that an examiner can follow and reproduce.

Secondary Research Data Analysis: Making Sense of What You Find

Gathering secondary sources is just half the job. It is a well-done literature review and rigorous data analysis that will make a contribution to knowledge and provide an original insight.

The most common method for analysing qualitative secondary sources such as published studies, policy documents and reports, is thematic analysis. You can recognise and note patterns and/or tensions in your sources, and organise your analysis around themes rather than summarising each source in turn. A strong literature review differs from an annotated bibliography in that it is themed.

If you are using quantitative secondary data (statistical datasets), you use the same statistical methods as you would with primary data, and you analyse the data from the questionnaire. Data in the dataset is summarised using descriptive statistics. Inferential statistics can be used to test relationships and make inferences beyond the data. Whereas in primary research it is important to explain how the data were obtained, in secondary research it is important to explain how the data were obtained and whether this influences the data you are using. Two practical points trip up first-time users of survey data: apply the survey weights supplied with the dataset, otherwise your estimates will not represent the population; and decide up front how you will treat missing values and recoded variables, then report that decision. A correlation found in a large national dataset is not evidence of cause, and saying so plainly earns more credit than overclaiming. If you are new to the software, the CEWUK blog has step-by-step walkthroughs such as running the McNemar test in SPSS and partial correlation in R, both written for non-statisticians.

In either case, critical engagement is a fundamental aspect. Secondary research analysis is not a report of the findings of your sources. Evaluating the strength of evidence, deciding where evidence agrees and where it is in dispute, understanding the scope of limitations of the existing evidence, and forming one's own reasoned stance based on the evidence that is collectively gathered.

How to Write Up Secondary Research in a Dissertation Methodology

A secondary-research dissertation still needs a full methodology chapter, and this is where many students lose marks by writing two paragraphs about "using existing literature". An examiner is looking for six things, in roughly this order.

1. A justification for choosing secondary research over primary, tied to your research question rather than to convenience.
2. A description of each dataset or body of literature: who collected it, when, why, how the sample was drawn and how large it is.
3. Your search strategy in enough detail to be repeated: databases, search strings, date range, inclusion and exclusion criteria, and how many sources you started and finished with.
4. Your analytical approach: thematic analysis, descriptive and inferential statistics, content analysis, or a framework such as PRISMA for a systematic review, with a sentence on why it suits the data.
5. An honest limitations paragraph covering the age of the data, its original purpose, geographic scope, and anything you could not measure.
6. An ethics and data-management statement, even if the outcome was that no formal approval was required, naming the licence you used the data under.

Students who are unsure how a secondary methodology chapter should be structured can look at the guidance on dissertation writing support from Cheap Essay Writing UK, which walks through the chapter-by-chapter expectations of UK examiners. The methodology has to agree with what you promised in the introduction, so it is worth re-reading the guide to the structure and format of a dissertation introduction before you finalise either chapter; students still at the proposal stage can use the dissertation proposal help page to see what supervisors expect a secondary-research proposal to contain.

Common Mistakes in Secondary Research (and How to Avoid Them)

The same errors appear in draft after draft that comes across our editors' desks. None of them is about intelligence; all of them are about process.

• Summarising instead of synthesising. Writing one paragraph per source produces an annotated bibliography, not a literature review. Organise by theme, argument or variable and let the sources argue with each other.
• Cherry-picking. Only citing studies that support your position is the most common form of bias in secondary research. Report the contradictory evidence and explain it.
• Ignoring the data's original purpose. A survey designed to measure job satisfaction cannot be quietly re-used as a measure of wellbeing. State what the variable actually measured.
• Using dated evidence without saying so. Pre-2020 data on remote working or student mental health describes a different world. Old data is acceptable; unacknowledged old data is not.
• Citing a secondary source as if you had read the primary one. If you found a finding quoted in a textbook, cite it as "cited in" or go and read the original. Examiners check.
• Treating blogs, Wikipedia and lecture slides as sources. Use them to find the real sources, then cite those.
• Leaving the limitations section to the last night. Limitations you identify yourself are a sign of rigour; limitations the examiner identifies for you are lost marks.

Several of these appear in the committee-room list of top reasons dissertations get rejected, which is worth reading before submission.

Best Practices for Secondary Research

There's a set of consistent habits that researchers who produce accurate secondary research follow.

They record all the sources they come across, so that you don't have to go through the process of finding out the information later.

They evaluate source quality and skip the time-consuming task of writing a thorough, in-depth analysis of the material if it isn't a good source.

They draw clear distinctions between the conclusions they come to and the information provided by their sources, a distinction that is of great importance to academic integrity.

They state the extent to which they are limited in their secondary data, even if it is in fact limited, such as being outdated, geographically limited or methodologically limited.

They use multiple, independent sources to support each key claim, and they triangulate, not just on one study or data set.

Methodology Note: If you are writing a systematic review or a high-level dissertation, document your search strategy completely. Clearly state your inclusion/exclusion criteria, target databases, and—where applicable—follow standard frameworks like PRISMA for total transparency. For qualitative reviews, the CASP checklists are the equivalent quality-appraisal tools most UK health and social-care departments expect to see named.

Conclusion

Secondary research is not an easy way to obtain original information, but it is a formal, scientific approach. When done systematically and with careful evaluation of sources and critical analysis, it yields insights that no single primary study could provide. It builds on knowledge from various studies, uses existing data on a scale for which primary data cannot compete, and lays the foundation for the evidence base on which all good research is built.

Stop confusing casual background reading with rigorous methodology. Master this process, and you'll save dozens of hours while producing sharper, higher-impact academic and professional writing.

Frequently Asked Questions

What are examples of secondary research sources?

Secondary research sources include peer-reviewed journal articles, government datasets, institutional reports, published books, statistical databases, historical records, and market research reports. In the UK, key sources include the ONS, the UK Data Service, and academic databases such as JSTOR and Scopus.

How is secondary research different from primary research?

Primary research involves collecting original data directly through interviews, surveys, or experiments. Secondary research analyses data already collected by others for a different purpose. Primary research generates new data; secondary research draws new insights from existing data.

What are the advantages of secondary research?

Secondary research is cost-effective, time-efficient, and gives access to large-scale datasets that no individual researcher could collect independently. It allows historical analysis, cross-national comparison, and rapid mapping of existing knowledge on a topic.

What are the disadvantages of secondary research?

The main disadvantages are a lack of control over data quality, potential obsolescence, and the risk that existing data may not precisely match your research question. The researcher must work within the methodological choices made by the original data collectors.

How do researchers collect secondary data?

Researchers collect secondary data by searching academic databases, accessing government open-data portals, consulting institutional reports, reviewing published literature, and using purpose-built research databases such as Statista or the UK Data Service.

What sources are commonly used in secondary research?

Commonly used sources include peer-reviewed journals accessed via JSTOR or Scopus; ONS and government statistics; WHO and OECD reports; think tank publications; Statista; the UK Data Service; and academic books from recognised publishers.

Is secondary research qualitative or quantitative?

It can be either. Analysing published studies, policy documents or media texts for themes is qualitative secondary research; running statistics on an existing survey or census dataset is quantitative secondary research. Many dissertations combine the two, using a literature review to frame a secondary data analysis.

Do I need ethical approval for secondary research?

Usually you still need to submit an ethics form, although the review is often lighter than for primary research. A full review is normally required where the data includes personal information about identifiable living people. Always check your university's policy and the licence conditions attached to the dataset you intend to use.

Can you write a dissertation using only secondary research?

Yes. Secondary-research dissertations are accepted across most UK disciplines, provided the methodology chapter justifies the choice, describes the datasets or literature in detail, documents the search and analysis process, and discusses limitations. Examiners assess the rigour of the process, not whether you collected the data yourself.

Is a literature review secondary research?

Yes. A literature review is the most common form of secondary research: it collects, evaluates and synthesises studies that other researchers have published. A systematic review and a meta-analysis are more structured versions of the same idea, and they are also classed as secondary research even though they may produce new statistical results.

What is desk research?

Desk research is another name for secondary research, used mainly in business and market research. It describes work that can be done at a desk, from existing reports, databases and publications, rather than in the field. Academic writers tend to say secondary research or secondary data analysis; the method is the same.

How long does secondary research take?

It depends on scope. A focused literature review for an undergraduate dissertation typically takes several weeks of searching, reading and note-taking. A secondary analysis of a national dataset adds time for cleaning and learning the data documentation. A full systematic review, with two independent screeners and quality appraisal, is a months-long project.

References

1. Johnson, R., Watkinson, A., & Mabe, M. (2018). The STM report: An overview of scientific and scholarly publishing (5th ed.). International Association of Scientific, Technical and Medical Publishers. https://www.stm-assoc.org/2018_10_04_STM_Report_2018.pdf
2. ONS. (2026). Statistics. https://www.ons.gov.uk/aboutus/whatwedo/statistics
3. Tantawi, R. (2021). Secondary data. EBSCO Research Starters. https://www.ebsco.com/research-starters/social-sciences-and-humanities/secondary-data
4. UK Data Service. (n.d.). Types of data access. https://ukdataservice.ac.uk/help/access-policy/types-of-data-access/
5. UK Data Service. (n.d.). Students: using data for your dissertation. https://ukdataservice.ac.uk/learning-hub/students/
6. University of St Andrews. (n.d.). Ethical guidance: secondary data. https://www.st-andrews.ac.uk/research/integrity-ethics/humans/ethical-guidance/secondary-data/
7. Page, M. J., et al. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372, n71. https://doi.org/10.1136/bmj.n71