Politics
Studies public problems and proposes solutions
A policy researcher in a think tank - an organisation that studies public policy - investigates problems and proposes solutions. They review studies, analyse figures, interview people who run or use public services and examine whether programmes work. Their findings may appear in a report, short briefing, chart or response to a government consultation. Every important claim must lead back to evidence.
Think tanks sit between universities, politics and public debate. They are usually slower than newsrooms, faster than universities and more public than government analysts. Most have a mission or recognisable political outlook. The researcher must stop evidence becoming decoration for a decision already made. At the same time, policy choices are never completely free of values. A limited finding can still be useful when its limits are clear.
Routes include paid internships, research-assistant posts, graduate schemes and moves from universities, charities, journalism or public services. Politics, economics and social-science degrees are common. Organisations that rely heavily on statistics may expect data training, while specialists value knowledge of areas such as health or housing. Employers examine how you conduct research, how clearly you write and whether you understand current policy debates. Postgraduate study helps in some teams, but no professional qualification is required.
The lower line resembles charity-sector pay, while the upper line approaches the Civil Service. Economics institutes tend to pay more because they compete with government and consultancies for staff. The later figures describe the researchers who stay, moving from research assistant towards senior fellow or associate director; many others leave think tanks within ten years.
| When | Lowest 10% earn under | A typical earner | Top 10% earn over |
|---|---|---|---|
| On graduating | £22,000 | £28,000 | £38,000 |
| 1 year in | £24,000 | £30,000 | £42,000 |
| 2 years in | £26,000 | £33,000 | £46,000 |
| 3 years in | £28,000 | £36,000 | £51,000 |
| 4 years in | £30,000 | £40,000 | £56,000 |
| 5 years in | £32,000 | £43,000 | £61,000 |
| 6 years in | £34,000 | £46,000 | £66,000 |
| 7 years in | £36,000 | £49,000 | £72,000 |
| 8 years in | £37,000 | £52,000 | £78,000 |
| 9 years in | £39,000 | £55,000 | £83,000 |
| 10 years in | £40,000 | £57,000 | £88,000 |
Public arguments often begin with a solution already attached: tougher penalties, more financial support or a new national target. Research gives you permission to ask what is causing the problem and who is missing from the statistics. You can also test whether the proposed measure tackles the real cause. Discovering that the original question is wrong can save months of energetic campaigning for the wrong answer.
Records created while running a public service may show where it is failing. They may not reveal whether the cause is cost, fear, an inaccessible form or a timetable that ignores shift work. Interviews can suggest an explanation, while a larger set of figures tests whether the experience is widespread. The pleasure lies in using different kinds of evidence to build a stronger explanation.
A strong report may be questioned by a journalist, used in advice to a minister or challenged at an event. People may return to it months later when politicians become interested in the subject. You see how different audiences test the same analysis. A useful chart or idea can also give a scattered concern a common language. The work is public enough to be criticised, quoted and improved.
Think tanks depend on grants, paid projects and donors. Their interests affect which subjects receive staff time. Responsible organisations protect researchers’ independence, but pressure can be quieter than an order to change a result. An inconvenient project may not be renewed, or a funder may want its preferred wording in the title. Short funding periods also create temporary jobs and interrupt the slow building of expertise.
A government spending announcement is next week, but access to the public records will take three months. You must decide whether a quick review, a few interviews or an older substitute measure can support a useful conclusion. Think-tank research repeatedly forces a compromise between speed and confidence. “We do not yet know” attracts less attention than another organisation’s confident recommendation.
Readers may accept or dismiss a report because of what they believe about the organisation before opening it. A careful analysis can quickly be labelled “left-wing”, “free-market”, “industry-funded” or “campaigning”. Supporters may exaggerate the result because it suits their case. You cannot control that response, but you must name the funders, defend the research and correct people on your own side.
Usually 3–4 years
Politics, economics, social policy and statistics are useful when they teach you how to test a claim, not only how to discuss one. Choose methods modules that make you collect, clean or interpret evidence and keep your decisions visible in the final work. A master’s is common and may be expected for strongly quantitative posts, but it does not replace practical research. A dissertation using real interviews, documents or data gives a research director something concrete to question when you apply for the junior rung.
Usually 0–2 years’ experience
You’ll find previous studies, clean sets of figures, arrange interviews, take notes and check references. At the same time, you learn how a think-tank project is built. Good early work means recording where evidence came from, spotting a suspicious figure and understanding why “I cannot verify this” is a useful finding.
Usually 2–5 years’ experience
You can take responsibility for a defined part of a project, choose suitable research methods with guidance and draft substantial sections of a publication. You’ll begin presenting findings in group discussions and explaining them to journalists or policy professionals. The shift is from completing tasks to judging what the project can honestly claim.
Usually 5–8 years’ experience
You’ll design and lead projects, combine interviews with statistics, manage junior researchers and defend conclusions under public or political challenge. You may write funding applications and build relationships with outside experts. The problems become less clear. A possible funder may know the answer it likes but not the question that could test it.
Usually 8–12 years’ experience
A principal fellow may become an expert in a research method or policy field. A programme manager takes greater responsibility for people, budgets and several projects. Both routes require checking the quality of other people’s research. You challenge weak conclusions, protect independence and decide when the evidence is strong enough for the organisation to take a position.
Usually 12+ years’ experience
You’ll choose the organisation’s research priorities, secure funding, build partnerships and protect its reputation for serious work. Directors create the conditions for careful research and decide where limited resources can add knowledge. Senior specialists may lead through deep expertise without managing a large team. Influence rests increasingly on choosing questions that will still matter after the immediate argument has moved on.
These subjects explain public institutions, political ideas and the history behind present disputes. Graduates still need strong enough research methods to test a political explanation. Informed commentary or a persuasive essay is not the same as research.
Economics develops reasoning about rewards, pressures, trade-offs and numerical effects. Its models can sharpen a policy question. Researchers must also notice how public services work in practice and whose experience is hidden by a convenient average or money-based measure.
These subjects examine public services, place, inequality and how people experience systems. They often provide useful training in both interviews and statistics. The gap may be learning to turn a careful academic discussion into a clear finding for a live debate without erasing important differences.
Strong numerical training helps with surveys, public records, experiments and analysis that other researchers can repeat. Policy work adds the challenge of deciding what the available measure leaves out. You must explain uncertainty to readers who may want a simple verdict.
Close reading, an understanding of context and sustained argument are valuable when evidence is spread across laws, archives, interviews and disputed accounts. Their weaker ground is often study design and statistics. Researchers must make the route from source to conclusion explicit enough for somebody else to challenge or repeat.
Education, health, engineering and community-development subjects can stop a project misunderstanding the system it studies. They bring knowledge of classrooms, care, infrastructure or local institutions. Researchers must turn that perspective into questions evidence can answer, understand the limits of personal examples and remain open to a result their own field may not welcome.
Hover a subject to see why it helpsTap a subject to see why it helps
08:45 – 09:30Question the headline
You reopen a draft report on a council trial of cheaper bus fares. The summary says passenger journeys rose by 18%. However, a note from the travel-card company reveals that changes between buses were counted differently halfway through the trial. The strongest number in the report may not describe a real change at all.
09:30 – 10:30Reconstruct the measure
You call the council analyst and bus operator to find when the counting changed. You rebuild the figures from the original records of people tapping onto buses. The increase falls to 8%. You ask a colleague to check because a believable correction can be just as wrong as the original figure. The earlier chart can no longer be used.
10:30 – 11:20Revisit what is known
A quick search of previous transport studies shows that lower fares often increase use. However, frequent and reliable buses matter greatly to people working unusual hours. You record where you searched, which words you used and why you included each study. You also separate evidence about actual journeys from surveys asking what people intended to do.
11:20 – 12:30Hear where the timetable breaks
Warehouse and care staff describe finishing after the final connecting bus. One participant supports cheaper fares but says they have not changed her journeys because she still pays for a taxi home. Her account suggests an explanation to test in the wider figures, not proof that everybody has the same experience. You protect her identity and do not claim more than the interview can show.
12:30 – 13:00Lunch
Cheese toastie.
13:00 – 14:10Split the result
You examine journeys by hour, route and ticket type. Most growth happened during the day. Evening use barely changed on routes where the final bus left before workers finished. The trial made available journeys cheaper, but it did not create missing buses. The figures now support the morning interview, although they cannot show how many people used taxis instead.
14:10 – 15:00Invite an unfriendly reading
A senior economist reviews the analysis. Roadworks disrupted one of the routes used for comparison, so you cannot claim that cheaper fares caused the whole increase. You remove that claim and simply describe what changed. A colleague’s challenge has made the conclusion smaller and the report stronger.
15:00 – 16:00Test the recommendation
Council officers, a disability organisation and the bus operator examine two proposals together. Extending every late route would be costly. Adding buses around the largest shift changes could be tested, but may not help disabled passengers who need connecting services. You propose a temporary trial with a review of disabled access rather than pretending the research has designed a complete network.
16:00 – 17:45Refuse the easy launch
The communications lead has already written the headline “Cut fares transform access to work”. You explain that the corrected figures do not support “transform” or prove that cheaper fares caused the increase. A more honest finding is that cheaper travel increased use mainly where a suitable bus already existed. Publication slips by a day while the charts and summary change. The result is less exciting, but it identifies the next test the council actually needs.
+3 hours compared with the average UK job