The AP Research rubric awards 7 of its 50 points for justifying your method and another 7 for drawing out the implications of your findings. That means the topic matters less than whether you can collect data on it and say something consequential about the result. The thirty below are grouped by the method they need, so you can pick one you can actually execute between September and April.
Most AP Research topic lists are lists of subjects. That is not what you need. You need a question with a method attached that you can finish between September and April.
The rubric makes this explicit. The AP Research Academic Paper rubric has nine rows and 50 points, and the two most heavily weighted rows are both about execution rather than subject:
| Rubric row | Points | What the top band asks for |
|---|---|---|
| Research: Design | 7 | A rationale for your method in relation to the question |
| Establish: Argument | 7 | The consequences and implications of your conclusion |
| Understand and Analyze: Context | 6 | Why the topic matters to the wider field |
| Evaluate: Sources and Evidence | 6 | Why your evidence is relevant and significant |
| Select and Use: Evidence | 6 | Interpretation and synthesis, not just reporting |
Fourteen of fifty points go to why this method and so what. A topic you cannot gather evidence on forfeits the first. A topic nobody would act on forfeits the second.
So the thirty below are grouped by method, because the method is the constraint.
Survey-based (6 topics)
You write the instrument, you recruit the sample, you analyse the responses. Feasible for anyone with a school and a willing year group.
- Does the presence of a visible phone on the desk affect recall performance in sixth-form students?
- How do students in selective and non-selective schools differ in their estimates of their own university chances?
- What proportion of students can correctly identify the source of a news claim they have shared?
- Does framing a maths problem as "a puzzle" rather than "a test" change reported anxiety and measured accuracy?
- How does the language teachers use about effort versus ability relate to students' stated willingness to attempt harder work?
- Do students who revise with handwritten notes report different confidence from those using typed notes, and does confidence track performance?
The design point to make explicitly in your paper: say why a survey was the right instrument for your question, and name what it cannot tell you.
Interview and qualitative (5 topics)
Smaller samples, richer data. Strong on the Context and Argument rows because you are building an account rather than testing a number.
- How do first-generation university applicants describe the information they wished they had had?
- How do teachers in state and independent schools describe their role in the university application process?
- What do students who dropped a subject at sixth form give as their reasons, and how do those differ from their teachers' accounts?
- How do recent immigrants to the UK describe navigating the school system for their children?
- How do student journalists describe the limits on what they can publish?
Coding your transcripts properly is where these projects are won. Say what your coding scheme was and how you arrived at it.
Content and discourse analysis (6 topics)
No participants, no ethics complications, and the data already exists. The most under-used category.
- How has the language used to describe artificial intelligence in UK national newspapers changed over the last five years?
- What proportion of university subject pages describe entry requirements in terms a 16-year-old could act on?
- How do political manifestos from the last three general elections differ in their treatment of higher education funding?
- What claims do commercial revision platforms make about outcomes, and what evidence do they cite?
- How are female and male scientists described differently in school textbook biographies?
- What changed in the public health messaging of a single government department across one defined period, and what appears to have prompted each shift?
Secondary analysis of public data (6 topics)
You analyse data somebody else collected. This is how most professional social science works, and it is entirely legitimate here.
- Does the relationship between school funding per pupil and attainment hold once intake is controlled for?
- How closely do local air quality readings track reported respiratory admissions in the same areas?
- Has the gap between male and female participation in A-level physics narrowed, and at what rate?
- What is the relationship between constituency-level turnout and median age across recent elections?
- Do countries with longer compulsory schooling show different patterns in tertiary participation?
- How do publicly reported waiting times vary between comparable NHS trusts, and what explains the variance?
Our guide to free datasets for student research lists where to get data that is clean enough to use.
Experimental and quasi-experimental (4 topics)
Possible without a lab if the manipulation is informational rather than physical.
- Does the order in which arguments are presented change which one readers rate as stronger?
- Does adding a named author to a short text change how credible readers judge it?
- Does labelling a product "sustainable" change stated willingness to pay, and by how much?
- Does the font a passage is set in affect reading time or comprehension in a within-subjects design?
Replication (3 topics)
Underrated, and strong on the Context row because you can explain exactly where your study sits in the literature.
- A direct replication of a well-known priming study with a UK school sample.
- A replication of a published finding about testing effects, with your own year group.
- A replication of a published content analysis using more recent material.
Examiners do not penalise replication. It is a recognised contribution, and the rubric's Context row specifically rewards connecting your work to the scholarly community — which a replication does by construction.
What a scoring topic looks like in practice
College Board publishes marked exemplars, and they are the best calibration available for judging whether your idea is the right size.
Its 2024 Sample D is a content analysis of FDA warning letters issued to herbal remedy manufacturers between 2005 and 2023, using a coding scheme adapted from earlier research. It scored a 4, and the reader commentary credits it specifically for a clear, narrow focus, for identifying a gap in the literature, and for a replicable method.
Notice what that topic is and is not. It is not novel science. It is a defined corpus of public documents, read systematically, with a coding scheme borrowed and adjusted. A sixth-former with a laptop could have done it.
That is the shape to aim for: a bounded set of evidence you can actually get hold of, examined with a method someone else could repeat. Several of the content-analysis and secondary-data topics above are deliberately built to that template.
What caps your score
Three failure modes account for most low marks, and none of them are about the subject.
Access you do not have. Clinical populations, children under 16 without elaborate consent, restricted datasets, specialist equipment. Students pick these in September and discover the problem in January, by which point the timetable has gone.
No measurable outcome. "Is gene editing ethical?" is a good question and a bad AP Research project, because there is no evidence you could collect that would answer it. You will write a literature review, and the Design row will cost you most of its seven points.
Too broad. The commonest problem by far, and the easiest to fix. Our guide to narrowing a research question covers the mechanics.
One thing worth knowing about the format
AP Research has no end-of-course written exam. College Board states this directly: your score is the 4,000–5,000 word academic paper at 75% plus a 15–20 minute presentation and oral defence at 25%, submitted through the AP Digital Portfolio.
That changes how you should pick a topic. You are not revising for a paper in May; you are managing a project to an April deadline. A topic whose data collection runs into February is a topic that will produce a rushed analysis. Pick for feasibility first and interest second — you will be living with it for eight months either way.
After the deadline
A finished AP Research paper with original data is close to the length and structure that student journals publish. Several students each year revise theirs for submission afterwards. The paper needs reworking rather than forwarding — journals want a different framing from an examined piece — and you should check each venue's policy on previously submitted work.
Our guides to where high school students can publish and how to publish research in high school cover what that involves, and our comparison of AP Research and AP Seminar covers how the two courses differ if you are still choosing.
For how the paper is actually marked, see our breakdown of the AP Research rubric.
Frequently asked questions
What makes a good AP Research topic?
One where you can collect your own data with the time, access and equipment you actually have, and where the result would matter to somebody. The rubric awards 7 points for the rationale behind your method and 7 for the consequences and implications of your conclusion. A topic you cannot gather evidence on loses the first; a topic nobody would act on loses the second.
Can I do an AP Research project without a lab?
Yes, and most strong papers do. Survey work, interview studies, content analysis, archival work and secondary analysis of public datasets all produce original data without equipment. The rubric does not reward laboratory work specifically — it rewards a method matched to the question and justified clearly.
How specific does my AP Research question need to be?
Specific enough that you know exactly what you are measuring and on whom. 'Does social media harm teenagers?' cannot be answered in a 5,000-word paper. 'Does the presence of a phone on the desk affect recall performance in a sample of 60 sixth-formers?' can. Narrowing is the single highest-return thing you can do to your score.
What AP Research topics should I avoid?
Three kinds. Topics requiring access you do not have — clinical populations, restricted data, specialist equipment. Topics with no measurable outcome, such as broad ethical questions. And topics so well covered that you cannot add anything, where you end up writing a literature review rather than a research paper.
Does my AP Research topic need to be in my intended major?
No. It helps for application purposes if it is adjacent to what you want to study, because you can then talk about it, but the rubric is indifferent. A well-executed project outside your field beats a weak one inside it.
Can I use my AP Research paper for something else afterwards?
Often, yes. A finished 4,000–5,000 word paper with original data is roughly the length and structure student journals expect. Some students submit a revised version after the AP deadline. You should check each journal's policy on prior submission and adapt the paper rather than sending it unchanged.