tesifyUK
Start free
How it worksFAQBlogHelpContactLog in

Tag: ONS

  • 40 Economics Dissertation Topics for UK Students, Each With a Research Question and a Dataset (2026)

    40 Economics Dissertation Topics for UK Students, Each With a Research Question and a Dataset (2026)

    The economics dissertation topics that get marked up in UK departments share one property: each is attached to a dataset the student can actually download in October. The forty topics below are grouped by sub-field, and every one names the research question in a single line and the UK or international dataset that answers it. Topics without an accessible dataset are not on the list.

    This guide is for UK undergraduate and MSc economics students, and for students on business, PPE or finance programmes who need an applied economics project. The sources are the ones UK economics departments expect to see: the Office for National Statistics, the Bank of England, the UK Data Service, Nomis, HM Revenue and Customs, the Department for Work and Pensions and the World Bank, all opened while this list was being written. The general narrowing method is in our pillar on dissertation topic ideas by subject, which has no economics section; this is that section, with the data attached.

    Where the data come from

    Source What it gives an economics dissertation Access
    ONS (Labour Force Survey, ASHE, CPI, GDP, regional accounts) Labour market, earnings, prices, output; ASHE 2025 release published as Employee earnings in the UK Free; LFS microdata via UK Data Service
    UK Data Service (Understanding Society, LFS microdata, Living Costs and Food Survey) Household panel and survey microdata Free registration through your university; see our access guide
    Nomis Official labour market and census statistics by local authority Free, no login
    Bank of England database Bank Rate history, quoted rates, money and credit, exchange rates Free, downloadable series
    HMRC and DWP statistics (Stat-Xplore) Tax receipts, benefit caseloads, Universal Credit by area Free; Stat-Xplore needs a free account for some tables
    HM Land Registry and the UK House Price Index Price paid data and monthly house price indices by local authority Free, open data
    World Bank World Development Indicators Cross-country panels for development and growth questions Free
    Companies House bulk data Firm births, dissolutions and filed accounts Free monthly snapshot

    The UK Data Service registration route, which most of the microdata below require, takes a few days and is set out step by step in our guide to getting UK Data Service data for a dissertation. Start it in the first week.

    The Economics Network on what an undergraduate economics dissertation is for; the topics below are chosen so the data exist.

    Labour economics

    1. Minimum wage and hours. Did the 2024 and 2025 National Living Wage increases reduce hours worked in low-paying sectors? ASHE, ONS.
    2. The gender pay gap by occupation. How much of the UK gender pay gap in 2025 is explained by occupation and hours rather than pay within jobs? ASHE, ONS, with the Blinder–Oaxaca decomposition.
    3. Graduate premium by subject. Has the earnings premium for a degree changed for the 2015 to 2025 graduating cohorts? Labour Force Survey microdata, UK Data Service.
    4. Working from home and wages. Do hybrid workers earn more than fully on-site workers with the same characteristics? LFS and Understanding Society.
    5. Youth unemployment by region. Which local authorities have the highest NEET rates and what local characteristics predict them? Nomis and Explore Education Statistics.
    6. Economic inactivity and long-term sickness. How much of the rise in inactivity since 2020 is explained by long-term sickness, by age group? LFS, ONS.
    7. Union membership and pay. Is there still a union wage premium in the UK private sector? LFS microdata, UK Data Service.
    8. Migration and local wages. Did regions with higher post-2021 migration inflows see different wage growth in low-skilled occupations? Nomis and ASHE by local authority.

    Macroeconomics and monetary policy

    1. Pass-through of Bank Rate. How quickly did quoted mortgage rates follow Bank Rate in the 2022 to 2024 tightening compared with earlier cycles? Bank of England database.
    2. Inflation expectations and outcomes. Did household inflation expectations lead or lag CPI during the 2022 inflation? Bank of England Inflation Attitudes Survey and ONS CPI.
    3. Energy prices and core inflation. How much of UK core inflation in 2023 is attributable to second-round effects of energy prices? ONS CPI component series.
    4. Quantitative tightening and gilt yields. Did announced gilt sales move the yield curve? Bank of England yield-curve data and announcement dates.
    5. Regional productivity. Why has output per hour in the North East diverged from London since 2010? ONS regional and sub-regional productivity.
    6. Fiscal multipliers at the local level. Did areas receiving more Levelling Up Fund money see faster employment growth? Nomis and published fund allocations.
    7. The Phillips curve after 2020. Has the relationship between UK unemployment and wage growth changed? LFS and Average Weekly Earnings, ONS.
    An economics student comparing two downloaded ONS time series in a spreadsheet beside printed notes on a library desk
    Every topic on the list has a dataset you can open in October. The ones that do not are the ones that stall in March.

    Housing and urban economics

    1. Interest rates and house prices. How did local house price growth respond to the 2022 to 2023 rate rises, and did highly leveraged areas fall further? UK House Price Index and Bank of England data.
    2. Stamp duty thresholds and transactions. Did the 2025 return of the lower stamp duty threshold shift transaction timing? HM Land Registry price paid data.
    3. Private rents and the benefit cap. Where does Local Housing Allowance fall furthest below market rents? ONS private rental prices and DWP Stat-Xplore.
    4. New supply and prices. Do local authorities that build more see slower price growth? UK House Price Index and government housing supply statistics.
    5. Housing quality and energy costs. How does dwelling energy efficiency relate to household energy spending? English Housing Survey.

    Public economics and inequality

    1. Universal Credit and employment. Did the rollout of Universal Credit change employment rates in the areas that received it first? Stat-Xplore and Nomis.
    2. The two-child limit and child poverty. What is the association between the two-child limit and relative child poverty by family size? Households Below Average Income, DWP, and Stat-Xplore.
    3. Income inequality over the 2010s. Did UK income inequality rise or fall between 2010 and 2024, and by which measure? Family Resources Survey via UK Data Service.
    4. Council tax regressivity. How regressive is council tax relative to property value in one region? Valuation bands and UK House Price Index.
    5. Fuel duty freezes. What has the real value of fuel duty done since 2011 and who benefited? HMRC receipts and ONS CPI.
    6. Wealth inequality by age. How has the wealth gap between under-35s and over-65s changed? Wealth and Assets Survey, UK Data Service.

    Industrial organisation and firms

    1. Firm births after Covid. Which sectors saw the largest net company formation in 2021 to 2025? Companies House bulk data.
    2. Supermarket prices and competition. Do grocery prices for a fixed basket differ by local market concentration? ONS item-level price data and store location data.
    3. Corporate insolvencies and interest rates. Which sectors’ insolvency rates responded most to the 2022 to 2024 rate rises? Insolvency Service statistics and Bank of England data.
    4. Energy price cap and small business. Did SME dissolution rates rise more in energy-intensive sectors in 2022 to 2023? Companies House and ONS business demography.
    5. Market entry in retail banking. Has the market share of challenger banks changed lending margins? Bank of England quoted rates and Financial Conduct Authority data.
    A hand-drawn map of UK regions annotated with unemployment figures beside a laptop showing a bar chart
    Local authority data from Nomis turns a national question into a regional one, which is usually where an undergraduate finds something new.

    Development and international economics

    1. Remittances and growth. Do remittance inflows predict growth in low-income countries after controlling for aid? World Development Indicators.
    2. Trade after the Trade and Cooperation Agreement. Which UK goods sectors saw the largest fall in EU exports after 2021? HMRC overseas trade statistics.
    3. Mobile money and financial inclusion. Is mobile money adoption associated with higher household saving in sub-Saharan Africa? World Bank Global Findex.
    4. Aid and education outcomes. Does education aid per child predict primary completion rates? World Development Indicators.
    5. Exchange rate pass-through. How much of the 2016 sterling depreciation reached UK import prices? ONS import price indices and Bank of England exchange rates.

    Behavioural, health and environmental economics

    1. Sugar levy and consumption. Did the Soft Drinks Industry Levy change household purchases of sugary drinks? Living Costs and Food Survey, UK Data Service.
    2. Cost of living and mental health. Is financial strain in 2022 to 2024 associated with reported mental health in the household panel? Understanding Society.
    3. Clean air zones and traffic. Did the introduction of a clean air zone change traffic volumes at monitored sites? Department for Transport road traffic counts.
    4. Carbon pricing and emissions. Did the UK Emissions Trading Scheme price track sector emissions after 2021? Government emissions statistics and scheme auction data.

    What the marker is looking for in an economics topic

    Economics markers read a proposal for three things before they read the topic: whether the question has a causal shape or a descriptive one, whether the data can distinguish the two, and whether the student has noticed the difference. A topic phrased as “the effect of X on Y” has promised a causal claim, and the proposal then has to say where the variation in X comes from that is not itself caused by Y. That is why the strongest undergraduate topics cluster around policy changes with dates and thresholds: the minimum wage uprating, the stamp duty threshold, the Universal Credit rollout, the sugar levy. Each supplies a before and an after, or a treated group and a comparison group, and the method follows from the design rather than the other way round.

    The second thing markers notice is whether the dataset matches the unit of analysis. A question about firms needs firm-level data, which in the UK means Companies House or a licensed database rather than ONS aggregates; a question about households needs the household panel; a question about regions can be answered from Nomis in an afternoon. The mismatch, a household question answered with regional averages, is the commonest reason an economics dissertation reports a correlation it cannot interpret.

    How to choose one and turn it into a question

    Three tests. First, can you download the data this week; if the dataset needs a special licence or a Freedom of Information request, choose another. Second, is there variation to exploit: a policy that arrived in some places before others, a threshold, a shock with a date. A topic with no variation produces a description, not a finding. Third, does the question fit a method you have been taught: difference-in-differences, panel fixed effects, a decomposition or an event study. The four narrowing moves that turn a topic into a testable question are the same across subjects and are set out in our topic-ideas pillar; the economics-specific step is to write the identification strategy in one sentence before the proposal, because the marker will ask for it.

    Once the dataset is chosen, the data sources most economics students use overlap with those in our comparison of finance dissertation data sources, and the sample-size and power argument in our guide to sample size for an undergraduate dissertation applies to a regression on 40 regions as much as to a survey.

    Where economics students lose marks on the topic

    On the identification sentence that never gets written. A topic with a dataset is a start; a question with a source of variation and a method is a dissertation. If the data are downloaded and the introduction and methodology are still blank, Tesify can draft both from your own question, your own dataset and your own identification strategy, in the order an economics marker checks them. Every sentence stays yours and is 100% written by you; what you get is a chapter that exists before the deadline.

    Frequently asked questions

    What makes a good economics dissertation topic?

    A question with a downloadable dataset, a source of variation such as a policy change or a threshold, and a method you have been taught. Topics that fail one of the three stall at the data stage.

    Which datasets do most UK economics dissertations use?

    The Labour Force Survey and Understanding Society through the UK Data Service, ONS earnings and price series, Nomis for local labour markets and the Bank of England database for monetary questions.

    How long does UK Data Service registration take?

    Usually a few days through a university login, longer for safeguarded datasets that need a project description. Start in the first week of term so the data are ready when the proposal is approved.

    Can I use international data instead of UK data?

    Yes. The World Bank World Development Indicators and Global Findex support cross-country panels, and a UK department will accept them provided the question is framed carefully and the limitations of country-level data are stated.

    Do I need panel data?

    Not necessarily. A well-identified cross-section or a time series with a clear policy date can support a strong undergraduate project. Panel data help when the question is about change within units over time.

    Is a purely descriptive economics dissertation acceptable?

    It can pass, but it rarely scores in the top band. Markers look for a question, a source of variation and a method, even a simple one, rather than a set of charts.

    How many observations do I need?

    Enough for the method: a difference-in-differences with 40 local authorities over 10 years gives 400 observations, which is workable. State the count after cleaning, not before.

    Should I choose a topic from current news?

    Only if the data exist. A 2026 policy has no outcomes to measure yet; choose an earlier version of the same policy or a comparable earlier shock with published data.

  • Where to Find UK Data for Your Dissertation: Free Sources by Subject (2026)

    Where to Find UK Data for Your Dissertation: Free Sources by Subject (2026)

    Secondary data analysis is the most underused route through a UK undergraduate dissertation. It removes the recruitment problem, removes most of the ethics burden, and gives you sample sizes no student survey will ever reach. The obstacle is not availability — it is knowing which sources an undergraduate can actually get into.

    That is the organising principle here. Every source below is real, currently live, and free at the point of use. What differs is the access tier, and one of those tiers is effectively closed to you.

    First, understand the three access tiers

    The UK Data Service — the UK’s largest collection of economic, population and social research data, funded by UKRI through the Economic and Social Research Council — operates a three-tier model that most other UK data holders mirror in some form.

    • Open. Neither login nor registration is required. Published under the Open Government Licence or Creative Commons. Download and go.
    • Safeguarded. You register and accept an End User Licence, agreeing not to share the data with unregistered users, not to attempt to identify individuals, and to cite the data correctly. This is the undergraduate route, and for students at UK institutions registration usually runs through your university login.
    • Controlled. Accessed only in a secure environment, requiring a detailed project application, accredited researcher status, training and an institutional legal agreement. The UK Data Service states directly that these data “are not suitable for use by inexperienced researchers, such as undergraduates, and should only be used if absolutely necessary.”

    Read that last line before you build a project around a controlled dataset. Students lose weeks discovering this at the application stage. Plan for open and safeguarded data, and treat anything controlled as out of scope.

    A related trap: the Office for National Statistics Secure Research Service provides access to de-identified unpublished data under the Five Safes framework, but full accredited researcher status requires an undergraduate degree or higher including a significant proportion of maths or statistics, or several years of quantitative research experience. It is not an undergraduate route either.

    Psychology, sociology and social policy

    UK Data Service is your first stop. It holds the major UK social surveys, and its safeguarded tier is designed for exactly the kind of secondary analysis an undergraduate project needs. Register through your institution, accept the End User Licence, and you have access to survey data with sample sizes in the thousands.

    UK Data Archive, based at the University of Essex, is the lead partner of the UK Data Service and has curated the UK’s social, economic and population data for over fifty years. It was the first academic department in a university to be awarded ISO 27001 certification, and was accredited in 2020 by the UK Statistics Authority under the Digital Economy Act 2017. In practice you will usually arrive at its holdings through the UK Data Service catalogue.

    Office for National Statistics publishes population, labour market, wellbeing and social survey outputs, all under the Open Government Licence v3.0. No registration, no licence negotiation.

    Politics and international relations

    The British Election Study has been explaining Britain’s electoral behaviour for sixty years and releases its data openly to researchers. Its two main strands are a large internet panel, currently released through Wave 30, and a random probability survey following the 2024 general election. For a dissertation on voting behaviour, partisanship or turnout, this is the field’s standard evidence base, and using it puts you on the same data as published political science.

    Check the study’s own data pages for the current registration requirements before planning around it, as access conditions differ between the panel and the probability survey.

    Criminology

    data.police.uk publishes street-level crime data, outcome data, and stop and search records, alongside police force and neighbourhood-level information and the Police Annual Data Requirement covering matters such as arrests and 101 call handling. It is released under the Open Government Licence v3.0 and offers an API as well as bulk downloads.

    For a criminology dissertation this is unusually good material: it is geographically granular, it is longitudinal, and it needs no ethical approval because it concerns recorded incidents rather than identifiable individuals. Be careful with the well-known limitation, though — recorded crime measures what was reported and recorded, not what occurred, and your analysis should say so.

    Education

    Explore Education Statistics, operated by the Department for Education and regulated by the Office for Statistics Regulation, is the single best entry point. It provides statistical summaries, an open data catalogue, a custom table builder and an API. The table builder matters for undergraduates: it lets you construct exactly the extract you need without writing code.

    This is also the appropriate source for attainment, absence, workforce and school-characteristics data — figures that circulate widely in secondary form but should be cited from the publisher.

    Health and nursing

    NHS England is now the custodian of England’s national health and social care datasets. Note the organisational change, because citing the wrong body dates your work immediately: NHS Digital legally merged into NHS England on 1 February 2023, and the regulations that effected the merger transferred NHS Digital’s functions to NHS England and abolished NHS Digital. The digital.nhs.uk domain still resolves, now branded NHS England Digital, which is why the old name persists in student bibliographies.

    A hard constraint sits alongside this for health students. The Health Research Authority states that standalone research at undergraduate level requiring ethics review or HRA approval cannot take place, and that undergraduate applications are no longer accepted for Research Ethics Committee review. Published aggregate NHS statistics are unaffected — you can analyse them freely — but a project involving NHS patients or staff as participants is not available to you. Secondary analysis is the HRA’s own suggested alternative, which makes this section the pragmatic route rather than the consolation prize. Our guide to choosing a survey platform and getting ethics right sets out that restriction in full.

    Economics, business and geography

    Nomis, a service provided by the Office for National Statistics, is the specialist labour market and census portal. It holds census data from 1921 onwards, labour market profiles, employment figures, population estimates and claimant counts, and most of the site can be used without registering. For any dissertation with a local or regional dimension — comparing labour markets across local authorities, mapping deprivation, tracking employment change — Nomis will do in minutes what would otherwise take days.

    data.gov.uk is the general catalogue of UK public data, published under the Open Government Licence v3.0. It is broad rather than deep: use it to discover that a dataset exists, then go to the publishing body for the authoritative version and the documentation.

    How to choose a dataset without wasting a fortnight

    Work in this order.

    1. Find the documentation before the data. Read the user guide and the questionnaire. If you cannot tell from the documentation which variable answers your research question, the dataset is not the right one.
    2. Check the access tier immediately. Open or safeguarded means you can proceed. Controlled means choose a different dataset.
    3. Check the unit of analysis. Individuals, households, schools, police force areas and local authorities are not interchangeable, and a mismatch between your research question and the unit will not be fixable later.
    4. Check the years available. A question about change over time needs comparable measures across waves, and survey questions get reworded more often than students expect.
    5. Download and open it before committing. Confirm the file format works with your software and that the variables you need are actually populated rather than missing for most respondents.

    Whichever package you plan to use, check it can read the file format — most UK archives distribute in formats that all the mainstream options handle, but confirming early avoids an unpleasant surprise. Our comparison of SPSS, R and jamovi covers the trade-offs if you have not settled on one.

    Open government datasets displayed as tables and charts on a monitor
    Read the documentation before the data — the variable list decides whether a dataset can answer your question.

    The advantage nobody mentions: your sample size problem disappears

    A student survey that reaches sixty people is powered to detect only large effects. A national social survey gives you thousands of cases, which means your project can address questions a primary-data undergraduate study genuinely cannot.

    That changes what you should write in your methods chapter. With secondary data you are not justifying a small convenience sample; you are justifying your analytic sample — the cases you retained after applying inclusion criteria and handling missing data — and explaining any weighting the survey requires. Our guide to sample size for an undergraduate dissertation covers how to frame that argument, and the same chapter-level discipline described in our guide to writing a methodology chapter applies: describe the decision, then defend it.

    Cite the data properly

    Datasets are citable objects with their own reference formats, and the End User Licences you accept normally require correct citation as a condition of access. Cite the data creator, the year, the title, the edition or wave, the distributor and the persistent identifier. Most archives publish the exact citation string on the dataset’s landing page — use theirs rather than composing your own.

    Also cite the version. Survey datasets are revised, and a reader who cannot tell which edition you analysed cannot reproduce your results.

    Once the data are in hand, the remaining work is writing the analysis up so the argument is legible. If that is where you are stuck, you can draft those chapters in Tesify from your own output and your own interpretation — 100% written by you.

    Frequently asked questions

    Do I need ethical approval for secondary data analysis?

    Usually a lighter process rather than none. Many departments require a short ethics form even for anonymised secondary data, and analysing existing data does not exempt you from your institution’s review procedures. Ask your supervisor before assuming you are exempt.

    Can undergraduates register with the UK Data Service?

    Yes, for open and safeguarded data, normally through your institutional login. Controlled data is a different matter — the service states explicitly that it is unsuitable for inexperienced researchers such as undergraduates.

    Is secondary data analysis seen as an easier dissertation?

    No, and it is marked on the same criteria. What you save in recruitment you spend on data management: understanding a complex dataset, handling missing values, deriving variables and applying weights correctly are substantial analytical tasks in their own right.

    Should I cite NHS Digital or NHS England?

    NHS England, for anything current. NHS Digital was abolished when it merged into NHS England on 1 February 2023. Cite NHS Digital only for material it published before that date, under the name it carried at the time.

    What does the Open Government Licence let me do?

    Broadly, it permits you to copy, adapt and use the information, including commercially, provided you acknowledge the source with the required attribution statement. For a dissertation this means you can analyse and reproduce the data freely as long as you cite it.

    Can I combine two different datasets?

    Sometimes, and it can produce a genuinely original project — for example joining local-authority-level crime and education data. Check that the geography, time period and definitions genuinely match, and note that some licences restrict linking datasets, so read the terms before you merge.

    What if the dataset I want is controlled?

    Redesign around an open or safeguarded alternative. Applications for controlled access involve accreditation, training and institutional agreements on timescales that do not fit an undergraduate project, and the holders themselves advise against it at this level.