A marking rubric is the marker's reading of your paper, written before the paper exists. That makes it the most direct guide to self-assessment you have: read your draft as the marker will, one criterion at a time.
EdCitation publishes this guide to checking a paper against a marking rubric, with the research and worked examples on real rubrics from Carnegie Mellon, the London School of Economics (LSE) and MIT. EdCitation's free Check your paper reads a brief and rubric into a checklist, each rule beside its sentence, and Theoretics QA, part of Max, checks the paper against them, quoting it for every finding. Pages cited were read on 23 September 2026.
What is a marking rubric, and how do analytic and holistic rubrics differ?
A marking rubric sets out what counts in an assignment and describes the quality of work at each level, as the University of Edinburgh's Academic Development (2026) guidance defines it. Carnegie Mellon's Eberly Center (n.d.) adds that it divides the work into parts, each described at varying levels of mastery. The same document is also called marking criteria, assessment criteria, grade descriptors or, in the LSE's Department of Sociology, a mark frame.
| Kind | What it looks like | What it tells you | A real example |
|---|---|---|---|
| Analytic | A grid: criteria down the side, levels across the top, a description in each cell | Where the paper stands on each criterion separately | Carnegie Mellon's philosophy paper rubric: four levels, Excellent to Unacceptable |
| Holistic | One scale, each level describing the whole paper | Which band the paper fits best overall | LSE Sociology's undergraduate mark frame: First (70 to 100%) down to Bad Fail (0 to 19%) |
| Single-point | One standard for each criterion | Whether each criterion reaches the standard | Edinburgh's example: a pass or not-yet-pass rubric |
DePaul University's Teaching Commons (n.d.) sets out the trade-off: an analytic rubric shows strengths and weaknesses and can weight its criteria, while a holistic one gives no specific feedback and is hard to apply when a paper sits at different levels on different criteria. Edinburgh agrees, and adds that holistic rubrics are not in common use there.
What the shape changes when you check
With an analytic rubric, check each criterion on its own, in proportion to its marks. With a holistic rubric or grade bands, the marker matches the whole paper to the band it fits best, so split each band into its clauses and check each clause as a criterion. Reading a rubric before you write is covered in how to read an assignment brief and rubric; this guide is the check after.
Does checking your own paper against a rubric help?
It tends to, on one condition: the benefit shows up when students use the rubric to judge their own work, not when they are merely handed it.
Panadero and Jonsson (2013) reviewed 21 studies of rubrics used to support learning rather than to grade. The firmest evidence of better performance came when rubrics were paired with self-assessment or a similar activity; where a rubric was simply handed out, with no training in using it, there was typically no significant effect.
A later meta-analysis by Panadero et al. (2023) pooled 54 effect sizes from 21 studies and found a moderate positive effect on academic performance: g = 0.45 after correcting for publication bias. Effects on self-regulated learning (g = 0.23) and self-efficacy (g = 0.18) were smaller, from five and three studies, with confidence intervals that include zero. The number of criteria or levels made no significant difference. The studies pooled for performance all combined the rubric with self- or peer assessment, so the rubric's own effect cannot be separated from theirs.
How students actually read rubrics
Andrade and Du (2005) held focus groups with 14 undergraduate teacher-education students at one American university. They used rubrics to plan, check work in progress and read feedback, but many did not read one in full: some read only the top level, or the top two. Fourteen students on one course is an illustration, not a law, but it points to the step most students skip: reading the band below the top one.
What a rubric cannot promise
Edinburgh's guidance warns that rubrics can suggest marking is completely objective, when it is often somewhat subjective. A self-check shows whether your paper does what the rubric describes; the mark stays the marker's.
How do I check my paper against the marking rubric?
Turn the rubric into questions, answer each with a passage from your paper, and fix whatever has none. DePaul notes that most rubrics convert fairly directly into a checklist; steps 4 to 6 are that conversion.
- Find the rubric that applies. A course's own criteria come first; the LSE's Sociology department, for one, posts them on the course's Moodle page.
- Paste the brief and the rubric into Check your paper. Free, with no account, it lifts out the countable rules, each beside its sentence.
- Note the weights, and spend your time to match. Students may assume every criterion counts equally, Edinburgh's guidance notes; MIT's 24.908 rubric gives content and use of evidence 30 of 100 points, and mechanics 10.
- Turn each criterion's top band into questions, one for each clause. Some rubrics come ready-made: MIT's 24.908 rubric already asks "Are counterarguments considered?"
- Read the band below and mark the words that change. They are what the top band asks for beyond it.
- Find the passage for each question. Note the paragraph and its first few words. If you cannot point to one, the answer is no.
- Mark each question met, partly met or missing, and fix first where the most marks sit furthest from the top band.
- Hand the rest to tools. Verify references checks every reference against the publisher's record, free; with Max, Theoretics QA reads the paper against the same brief and rubric.
A worked example: Carnegie Mellon's philosophy paper rubric
Harrell (n.d.) wrote this analytic rubric for papers across Carnegie Mellon's philosophy courses, and the Eberly Center publishes it: four levels, Excellent to Unacceptable, and 17 criteria from argument to organisation. Three show the method.
Three criteria turned into questions
| Criterion | Excellent asks for | Where Good falls short | The question for your paper |
|---|---|---|---|
| Introduction | A clear thesis "contained in the introduction", and a concise summary of the argument's steps | The route is only "generally clear" | Does my first paragraph state my conclusion and its steps, briefly? |
| Counter-arguments | "Both obvious and unobvious" objections, with "original and/or thoughtful responses" | Only the obvious objections, answered | Have I answered an objection a reader would not think of first, with a reason? |
| Conclusion | Loose ends tied: an unanswered objection, an implication, or further work | Some loose ends, mixed with a restatement of the introduction | Does my conclusion do one thing my introduction did not? |
An invented draft, traced
Invented for this example: 1,800 words arguing that the wealthy must give to effective charities. The thesis appears only at the end, the familiar objection that such a duty asks too much is dismissed in a sentence, and the conclusion repeats the introduction.
| Question | The draft | Level it matches | The fix, yours to write |
|---|---|---|---|
| Conclusion and route up front | Thesis at the end | Needs Improvement: thesis "not contained in the introduction" | State the conclusion and its steps at the start |
| An unobvious objection, answered | One obvious objection, waved away | Needs Improvement: "mere claims of refutation" | Give the reply a reason; answer a less obvious objection too |
| A conclusion that adds something | A restatement | Needs Improvement: "merely a restatement" | Add an implication, or an objection left open |
None of this is a mark: the marker weighs all 17 criteria.
Reading grade bands: the LSE mark frame
Grade bands are read the same way. The LSE's Department of Sociology (n.d.) mark frame gives one paragraph to each class. The Upper Second (60 to 69%) "attempts a systematic analysis"; the First (70 to 100%) shows "clarity of analysis", "engages directly with the question", offers an "independent and critical interpretation" and shows "an outstanding breadth and depth of reading".
Each difference is a question. Does every section answer the question as set? Is the interpretation my own? Does my reading go beyond the lecture material? Correct citations and bibliography appear in both upper bands: a floor, not a step up.
What does Check your paper read from a brief and rubric?
EdCitation's Check your paper reads the rules a paper can be measured against, from word limits for each part to sources, style, font, spacing and margins, each beside its sentence; anything misread can be unticked. What no rule can test is listed separately, for a reader to judge. It is free, with no account.
What came back from a real MIT brief and rubric
We pasted three things from MIT's 24.908, Creole Languages and Caribbean Identities: the essay guidelines and the prompt for Essay Assignment 2, on Edwidge Danticat's The Farming of Bones (DeGraff, 2017a), and the course's grading rubric (DeGraff, 2017b). Seven rules came back.
| What it asks | Asked for | Read from |
|---|---|---|
| Font size | 12 pt | "use a normal 12-point font" |
| Line spacing | 2× | "Be sure to double-space" |
| Margins | 1 to 1.25 in | "1-1.25 inch margins" |
| Page numbers | Required | "Number your pages" |
| Word count | About 1,000 words | "Length of essay: 1,000 words TOTAL, including title and bibliography" |
| Citation style | MLA | "Consult the MLA Handbook for Writers of Research Papers" |
| Reference list | Required | The word-count sentence, which names the bibliography |
Read it before trusting it: each rule shows the sentence it came from, so a misreading is easy to catch and untick. Here the margins came back as the range the brief allows, and page numbers from "Number your pages", not from the sentence asking for page numbers in citations.
What it did not read as a rule
Ten sentences were set aside as needing judgement: nine from the brief, from putting your name and the date on the first page to choosing at most three passages and supporting each assertion with cited readings, and one from the rubric, on background in the introduction. The rubric's other questions and its weights did not appear. Pasted alone, Carnegie Mellon's rubric gave no rules and 19 set-aside sentences from all four levels, none on counter-arguments; the LSE mark frame gave one rule, a reference list, and three sentences, all from its fail bands.
For a rule reader that is right: a rule invented from a description of quality would be checked as though the brief had set it. But the set-aside list is not your top-band checklist. Write those questions yourself, and give the judgement to a reader.
What does Theoretics QA check against a rubric?
Theoretics QA checks your paper against the instructions you give it, and a rubric is instructions. Give it the brief, the rubric or both, confirm the checklist, and upload the paper in your dashboard; it comes back marked rule by rule, with a reader's judgement above.
It is the best tool we know of for this check, and the method's last step is why. A rubric check ends in finding the passage that meets each criterion, and that is Theoretics QA's evidence: for each thing your own instructions ask, it says whether the paper does it, quotes the passage that shows it, and says what is missing. It writes nothing: no sentence of the paper is rewritten or paraphrased, and no grade is given. A chatbot asked to lift a paragraph to the top band writes it for you, which may breach your course's rules on AI.
The same report checks formatting, comments and tracked changes left in the file, numbers and language; matches each citation to the list; and looks each reference up in the publisher's record, flagging invented and retracted ones. Reading the instructions is free; Theoretics QA is part of Max, $24 a month or $216 a year, free for the first 14 days (pricing). We could not run it for this guide, so no finding on the invented draft is shown, and it predicts no mark.
Which tool checks which line of a rubric
| A line in a rubric or brief | EdCitation tool | Cost |
|---|---|---|
| Length, style, font, spacing, margins | Check your paper | Free, no account |
| Argument, counter-arguments, "engages directly with the question" | Theoretics QA | Max, $24 a month |
| "Are sources properly cited?": each reference real, none retracted | Verify references | Free, no account |
| Each entry in the named style | Cite a source | Free, no account |
| Every citation matched to the list, from the file | References from a file | Pro, $8 a month; three checks a month with a free account |
| Left-over comments and tracked changes, language slips | Mechanics QA | Pro, $8 a month; three checks a month with a free account |
Mechanics QA measures formatting against the APA 7 student paper, so for an MLA essay use it for the leftovers and the language. For the order of every check, see the complete pre-submission checklist; for the question itself, does my paper answer the assignment question; for markers, checking submissions against the assignment brief.
Quick questions
What is the difference between an analytic and a holistic rubric?
An analytic rubric describes each criterion separately at several levels, showing where a paper is strong and weak. A holistic rubric describes the whole paper at each level, and the marker picks the best fit.
How do I use a marking rubric to self-assess my essay?
Turn each criterion's top band into questions, find the passage in your essay that answers each, and read the band below for what the top one adds. EdCitation's free Check your paper lists the brief's countable rules first.
Should I read every band of a rubric, or only the top one?
Read at least the top band and the one below it. Some students in Andrade and Du's focus groups read only the top level, but the words that change between bands are what the higher one asks for.
Can a tool tell me what mark my paper will get against the rubric?
No. EdCitation's Theoretics QA, part of Max, reads your paper against the brief and rubric and quotes the passages behind each finding, but it gives no grade. The mark stays the marker's judgement.
References
- Andrade, H., & Du, Y. (2005). Student perspectives on rubric-referenced assessment. Practical Assessment, Research & Evaluation, 10(3). https://doi.org/10.7275/g367-ye94
- Carnegie Mellon University, Eberly Center. (n.d.). Grading and performance rubrics. https://www.cmu.edu/teaching/designteach/teach/rubrics.html
- DeGraff, M. (2017a). Assignments [Course page]. 24.908 Creole Languages and Caribbean Identities, MIT OpenCourseWare. https://ocw.mit.edu/courses/24-908-creole-languages-and-caribbean-identities-spring-2017/pages/assignments/
- DeGraff, M. (2017b). Grading rubric [Rubric]. 24.908 Creole Languages and Caribbean Identities, MIT OpenCourseWare. https://ocw.mit.edu/courses/24-908-creole-languages-and-caribbean-identities-spring-2017/613472855deeb6a17ec81c0fea910f04_MIT24_908s17_assn_rubric.pdf
- DePaul University Teaching Commons. (n.d.). Types of rubrics. https://resources.depaul.edu/teaching-commons/teaching-guides/feedback-grading/rubrics/Pages/types-of-rubrics.aspx
- Harrell, M. (n.d.). Philosophy paper grading rubric [Rubric]. Eberly Center, Carnegie Mellon University. https://www.cmu.edu/teaching/resources/Teaching/CourseDesign/Assessment-Grading/Rubrics/PhilosophyPaperRubric.doc
- London School of Economics and Political Science, Department of Sociology. (n.d.). Undergraduate mark frame [PDF]. https://www.lse.ac.uk/sociology/assets/documents/study/Assessment-and-Feedback/BSc-Mark-Frame.pdf
- Panadero, E., & Jonsson, A. (2013). The use of scoring rubrics for formative assessment purposes revisited: A review. Educational Research Review, 9, 129-144. https://doi.org/10.1016/j.edurev.2013.01.002
- Panadero, E., Jonsson, A., Pinedo, L., & Fernández-Castilla, B. (2023). Effects of rubrics on academic performance, self-regulated learning, and self-efficacy: A meta-analytic review. Educational Psychology Review, 35(4), Article 113. https://doi.org/10.1007/s10648-023-09823-4
- University of Edinburgh, Academic Development. (2026, July 24). Assessment rubrics. https://registryservices.ed.ac.uk/academic-development/teaching/assessment-feedback/rubrics