What this rubric measures
The AP Research Academic Paper Rubric is the official scoring guide used to evaluate student writing on AP Research assessments. It is an Holistic (5-point scale) rubric that scores responses across 1 distinct criteria, allowing teachers to give precise, targeted feedback on each area of writing.
All 1 scoring criteria
Click any criterion to expand its score level descriptors. The language below is taken verbatim from the official College Board AP Research scoring guide.
1 Academic Paper (overall holistic score)
Rich Analysis of a New Understanding Addressing a Gap in the Research Base. The 5-level paper does all of the following:
- Focuses a topic of inquiry with clear and narrow parameters, which are addressed through the method and the conclusion.
- Explicitly connects a topic of inquiry to relevant scholarly works of varying perspectives AND logically explains how the topic of inquiry addresses a gap.
- Logically defends the alignment of a detailed, replicable research method to the purpose of the inquiry.
- Justifies a new understanding or conclusion through a logical progression of inquiry choices, sufficient evidence, explanation of the limitations of the conclusion, and an explanation of the implications to the community of practice.
- Enhances the communication of the student's ideas through organization, use of design elements, conventions of grammar, style, mechanics, and word precision, with few to no errors.
- Cites AND attributes sources, with a consistent use of an appropriate discipline-specific style (in both bibliography/works cited AND in-text), with few to no errors.
Well-Supported, Articulate Argument Conveying a New Understanding. The 4-level paper does all of the following:
- Focuses a topic of inquiry with clear and narrow parameters, which are addressed through the method and the conclusion.
- Explicitly connects a topic of inquiry to relevant scholarly works of varying perspectives AND logically explains how the topic of inquiry addresses a gap.
- Logically defends the alignment of a detailed, replicable research method to the purpose of the inquiry.
- Supports a new understanding or conclusion through a logically organized line of reasoning AND sufficient evidence. The limitations and/or implications, if present, of the new understanding or conclusion are oversimplified.
- Competently communicates the student's ideas, although there may be some errors in grammar, discipline-specific style, and organization.
- Cites AND attributes sources, with a consistent use of an appropriate discipline-specific style (in both bibliography/works cited AND in-text), with few to no errors.
Ineffectual Argument for a New Understanding. The 3-level paper does all of the following:
- Carries the focus or scope of a topic of inquiry through the method AND overall line of reasoning, even though the focus or scope might still be narrowing.
- Situates a topic of inquiry within relevant scholarly works of varying perspectives, although connections to some works may be unclear.
- Describes a reasonably replicable research method, with questionable alignment to the purpose of the inquiry.
- Conveys a new understanding or conclusion, with an underdeveloped line of reasoning OR insufficient evidence.
- Competently communicates the student's ideas, although there may be some errors in grammar, discipline-specific style, and organization.
- Cites AND attributes sources, using a discipline-specific style (in both bibliography/works cited AND in-text), with few errors or inconsistencies.
Report on Existing Knowledge with Simplistic Use of a Research Method. The 2-level paper does all of the following:
- Presents a topic of inquiry with narrowing scope or focus, that is NOT carried through either in the method or in the overall line of reasoning.
- Situates a topic of inquiry within a single perspective derived from scholarly works OR through a variety of perspectives derived from mostly non-scholarly works.
- Describes a nonreplicable research method OR provides an oversimplified description of a method, with questionable alignment to the purpose of the inquiry.
- Summarizes or reports existing knowledge in the field of understanding pertaining to the topic of inquiry.
- Generally communicates the student's ideas, although errors in grammar, discipline-specific style, and organization distract or confuse the reader.
- Cites AND/OR attributes sources (in bibliography/works cited and/or in-text), with multiple errors and/or an inconsistent use of a discipline specific style.
Report on Existing Knowledge. The 1-level paper does all of the following:
- Presents an overly broad topic of inquiry.
- Situates a topic of inquiry within a single perspective derived from scholarly works OR through a variety of perspectives derived from mostly non-scholarly works.
- Describes a search and report process.
- Summarizes or reports existing knowledge in the field of understanding pertaining to the topic of inquiry.
- Generally communicates the student's ideas, although errors in grammar, discipline-specific style, and organization distract or confuse the reader.
- Cites AND/OR attributes sources (in bibliography/works cited and/or in-text), with multiple errors and/or an inconsistent use of a discipline specific style.
The Academic Paper rubric is HOLISTIC: each score level requires meeting all five rows of attributes simultaneously. Each row describes the same scoring attribute (topic of inquiry, perspective from scholarly works, research method, conclusion/new understanding, communication, citation) at each performance level. Readers assign the score that best fits the paper as a whole.
How to score with the AP Research Academic Paper Rubric.
A practical guide for teachers and norming teams. How to apply each descriptor consistently, the pitfalls that hurt inter-rater reliability, and a workflow for calibrating with colleagues.
One holistic score, six rows of attributes
- The Academic Paper is scored ONCE on a 1 to 5 holistic scale, not per-row. Each score level specifies performance across all six attributes (topic, perspective, method, conclusion, communication, citation) simultaneously.
- To earn a score of 5, the paper must meet the 5-level criteria on all six rows of attributes. If the paper meets the 5-level on most attributes but only the 4-level on one, it scores 4 overall.
- Readers use a best-fit standard. The paper is assigned the score level that best characterizes the paper as a whole.
The 3-to-4 transition is the highest-leverage cutpoint
- The transition from 3 (Ineffectual Argument) to 4 (Well-Supported Argument) is the most consequential cutpoint on the rubric. A 3 paper conveys a new understanding but with underdeveloped reasoning or insufficient evidence. A 4 paper supports the new understanding through logical reasoning AND sufficient evidence.
- The transition from 4 to 5 is about the explicit treatment of limitations and implications. A 4 paper may oversimplify the limitations and implications. A 5 paper justifies the new understanding through inquiry choices AND explains the limitations AND explains the implications to the community of practice.
- The transition from 2 to 3 is about whether the paper makes a new argument at all. A 2 paper summarizes existing knowledge. A 3 paper attempts a new argument but with weak support.
Common pitfalls to avoid
- Awarding 5 to a paper that has rich analysis but inconsistent citation style, the 5-level requires few to no citation errors AND consistent discipline-specific style.
- Awarding 4 to a paper that does not explicitly address a gap, the 4-level requires explicitly explaining how the topic of inquiry addresses a gap.
- Awarding 3 to a paper that reports existing knowledge, papers that summarize rather than argue cap at 2.
- Awarding 2 to a paper with an overly broad topic, papers with overly broad topics cap at 1.
Tips for AP norming
- Anchor your norming session with the College Board's released sample AP Research papers, scored and annotated by AP Readers.
- Norming AP Research papers is harder than norming most AP rubrics because the holistic scoring requires evaluating six attributes simultaneously. Use the released exemplars to calibrate the 3-to-4 and 4-to-5 transitions in particular.
- Read each paper twice before scoring. The discipline of the holistic score makes first-impression scoring unreliable.
Notes for the AP Research Academic Paper Rubric
The Academic Paper is the central artifact of AP Research. It is a 4,000 to 5,000 word original research paper produced over the course of the year. Students develop their own research question, design and execute a research method, analyze results, and situate their findings in the scholarly literature.
Scoring is holistic on a 1 to 5 scale. Each score level requires meeting all six rows of attributes simultaneously. The College Board explicitly identifies the rows of attributes as topic of inquiry, perspective from scholarly works, research method, conclusion/new understanding, communication, and citation.
The 3-to-4 transition is the highest-leverage cutpoint and the most common debate in norming. A 3 paper attempts a new argument with weak support. A 4 paper supports the new argument through both logical reasoning AND sufficient evidence.
The Academic Paper is submitted by the AP submission deadline in late April and is externally scored by trained AP readers. Teachers may use the rubric formatively throughout the year (EnlightenAI can support this), but the official paper score comes from College Board readers.
See this rubric in action.
EnlightenAI scores student writing on this exact rubric, with per-criterion feedback that mirrors how you grade by hand. The sample response below shows how the rubric applies to a real piece of student writing, scored against every criterion.
Academic Paper excerpts on algorithmic course recommendation and post-secondary outcomes
Abstract (verbatim excerpt)
This study examines whether the introduction of an algorithmic course-recommendation system in a single mid-sized US public school district narrowed or widened the variance of post-secondary outcomes for students from low-income backgrounds. Using a difference-in-differences design comparing recommendation-eligible students (treated, n = 412) and non-eligible students (control, n = 389) over a 5-year period, supplemented by 24 semi-structured interviews with students, parents, and counselors, the study finds a small but statistically significant reduction in outcome variance for the treated group (p = .03), with the strongest effects concentrated among students with prior course choices that were misaligned with their stated career interests. The mixed-methods evidence suggests algorithmic recommendation can serve as a corrective for information asymmetries that historically disadvantage low-income students, while also revealing that algorithm trust varies substantially across student subgroups.
Gap in the research base (verbatim excerpt)
Existing research on K-12 algorithmic recommendation has focused predominantly on equity in college admissions algorithms (Friedler and Scheidegger 2021; Barocas et al. 2023). The empirical literature on within-K-12 course-recommendation systems is limited and divided. Selwyn (2022) and Williamson (2023) argue from a critical-technology perspective that algorithmic recommendation tends to reinforce existing inequalities; Holstein and Aleven (2023), arguing from a learning-engineering perspective, document cases where algorithmic recommendation has narrowed achievement gaps when paired with counselor oversight. The gap this study addresses is empirical: no published mixed-methods study has examined outcome VARIANCE rather than mean outcomes within a single district where a recommendation algorithm was newly introduced. This study tests whether the variance result holds in one district, and if so, examines the mechanisms through which variance reduction occurs.
Method (verbatim excerpt)
The research design combines a quasi-experimental difference-in-differences quantitative analysis with semi-structured qualitative interviews. The quantitative component uses district administrative data covering five graduating cohorts (n = 801 students total across treatment and control). The outcome variable is the standard deviation of a composite post-secondary outcome index combining matriculation type, first-year persistence, and major alignment with stated career interest. The qualitative component consists of 24 interviews (8 students, 8 parents, 8 counselors), coded inductively with two coders (Cohen's kappa = 0.81). The method is detailed enough for replication and aligned to the research question; the quantitative result is the primary test and the qualitative result is the mechanism check.
Limitations and implications (verbatim excerpt)
Three limitations bound this study. First, the sample is one district, and generalizability to other districts depends on similar counselor-algorithm-student dynamics. Second, the five-year window does not allow inference about long-term effects. Third, the qualitative sample is small. With those bounds, the study has implications for the community of practice that go beyond the immediate research base. For practitioners, the variance result suggests that algorithmic recommendation paired with counselor oversight can serve a corrective function in districts where information asymmetries about post-secondary pathways are large. For policymakers, the algorithm-trust variation across student subgroups raises the question of whether transparency requirements should accompany district-level adoption. For researchers, the variance-rather-than-mean framing opens a more fruitful inquiry pathway than the equity literature has typically pursued. Each implication is grounded in this study's evidence but is offered with appropriate epistemic humility given the limitations.
Rich analysis of a new understanding addressing a gap in the research base
Topic narrow (one district, variance vs mean); method (mixed-methods DID + interviews) replicable and aligned; perspectives from scholarly literature with explicit gap statement; conclusion justified with limitations AND community-of-practice implications. Earns a holistic 5.
Explicit limitations and implications differentiate the 5 from a 4
A 4 paper oversimplifies limitations and implications. This paper enumerates three specific limitations (district, window, sample) AND offers community-of-practice implications for three audiences (practitioners, policymakers, researchers). The 4-to-5 differentiator.
Discipline-specific style executed cleanly throughout
Mixed-methods reporting followed (quant result with effect size and p-value, qual result with inter-coder agreement). Citations consistent APA style with few errors. Communication enhances rather than distracts. Reinforces the holistic 5.
Score this rubric consistently, with the feedback students actually use
EnlightenAI is trained on your standards and your exemplars, then scores at the speed of your classroom.
Trained on your rubric
Upload this rubric, or any custom one, and the AI learns your exact criteria, descriptor language, and score level boundaries.
Per-criterion feedback
Students receive specific, actionable comments tied to each criterion, exactly the way you'd grade by hand.
Built for K–12 schools
Roster sync, FERPA-aligned data handling, and per-school configuration so every campus uses the same standards.
About the AP Research Academic Paper Rubric
What is the AP Research Academic Paper?
How is the AP Research Academic Paper scored?
Does every part of the AP Research paper need to meet the same level to earn that score?
What's the difference between a 3 and a 4 on the AP Research paper?
What's the difference between a 4 and a 5 on the AP Research paper?
Is this rubric the official version from College Board?
Where can I find the source document?
Can EnlightenAI score student writing using this rubric?
Use this rubric in EnlightenAI
Train EnlightenAI on the AP Research Academic Paper rubric and start scoring student paper drafts with consistent feedback aligned to the College Board's row-of-attributes scoring approach.