Rubrics show up everywhere from second-grade book reports to college research papers, and the word itself confuses plenty of students and parents. This guide explains what a rubric is, the parts every rubric has, the main types with sample rubrics you can copy, how teachers turn a rubric score into a grade, how to use a rubric to raise your own grade, and how to write one if you teach.
What is a rubric?
Carnegie Mellon University’s Eberly Center for teaching defines a rubric as “a scoring tool that explicitly describes the instructor’s performance expectations for an assignment or piece of work.” In other words, it is the answer key for work that does not have one right answer. A math quiz can be graded right or wrong; an essay, a lab report, a speech or an art project needs a rubric to say what “good” means.
A rubric does two jobs at once. Before you start, it tells you what your teacher is looking for. After grading, it explains where your points went. That is why teachers often hand out the rubric with the assignment instead of keeping it to themselves.
Where does the word “rubric” come from?
According to Merriam-Webster, rubric comes from the Latin rubrica, red ocher, from ruber, “red,” and was first used in English in the 14th century. Scribes copying manuscripts wrote headings and special instructions in red ink to set them apart from the black text, and those red headings came to be called rubrics. Over time the word also came to mean an authoritative rule or a category, and educators borrowed it for a set of grading rules. Outside school you may still see “under the rubric of” used to mean “under the heading of.”
What does a rubric look like?
Most rubrics are tables. Here is a sample analytic rubric for a five-paragraph persuasive essay, the kind of rubric you are most likely to get in middle school, high school or a first-year college writing class. We wrote it for this guide; teachers are welcome to adapt it.
| Criterion | 4: Exceeds | 3: Meets | 2: Approaching | 1: Beginning |
|---|---|---|---|---|
| Thesis | Clear, specific claim that takes a position and previews the reasons | Clear claim that takes a position | Claim is present but vague or hard to find | No identifiable claim |
| Evidence | Three or more relevant, accurate sources, each explained | Two or more relevant sources, mostly explained | One source, or sources listed without explanation | No supporting evidence |
| Organization | Each paragraph has one job; transitions guide the reader | Logical order with a few rough transitions | Some paragraphs out of order or mixing ideas | No clear order |
| Counterargument | Names a fair opposing view and answers it with evidence | Names an opposing view and responds | Mentions an opposing view without responding | No opposing view |
| Mechanics | No more than two errors; citations formatted correctly | A few errors that do not distract; citations mostly correct | Frequent errors that sometimes distract | Errors make the essay hard to read |
Read it row by row. Each row is one criterion. The cells to the right describe what work at each level looks like for that criterion. A student whose thesis is clear but vague about reasons would land in the “3: Meets” box for that row, and so on down the table.
The four parts of a rubric
Carnegie Mellon’s guide breaks a rubric into three parts, and most classroom rubrics add a fourth for points:
- Criteria. The aspects of the work being graded, such as argument, evidence, clarity, teamwork or accuracy. Good rubrics keep these to a handful, usually three to six.
- Performance levels. The rating scale, written as numbers (4, 3, 2, 1), labels (Exceeds, Meets, Approaching, Beginning), letters, or a mix.
- Descriptors. The sentences inside each box that say what work at that level looks like. Descriptors are what make a rubric useful; a rubric with only numbers is just a score sheet.
- Points or weights. How much each criterion counts. Some rubrics give every row the same value; others double the weight of the thesis or the data analysis because it matters more.
Why are rubrics important?
The Eberly Center lists benefits on both sides of the desk. For teachers, rubrics cut grading time because a descriptor replaces a long written comment, keep grades consistent across a stack of papers and across different graders, show which skills a whole class is struggling with, and head off complaints about grades. For students, rubrics make expectations clear up front, make feedback easier to use, let you check your progress as you work, and show you exactly which skills are strong and which need work.
Rubrics also make grading fairer, though not perfectly fair. Two teachers using the same rubric can still disagree on whether an argument is “clear” or “mostly clear,” which is why careful rubrics use concrete descriptors (“two or more sources”) instead of vague ones (“good sources”).
Types of rubrics, with examples
DePaul University’s Teaching Commons describes four main types: analytic, holistic, checklists and developmental rubrics. Many teachers also use a fifth, the single-point rubric. Here is how each one works.
Analytic rubric
An analytic rubric scores each criterion separately, like the essay rubric above. DePaul describes it as a grid with criteria in the left column and levels across the top row. Its strength is feedback: you can see that your evidence earned a 4 and your organization a 2. It also lets teachers weight criteria differently. The trade-offs are that it takes longer to write and to use, and graders can still disagree unless each level is clearly defined.
Holistic rubric
A holistic rubric gives one overall score. The grader reads the whole piece and matches it to the single description that fits best, usually on a 1 to 4 or 1 to 6 scale. Here is a sample holistic rubric for a five-minute class presentation:
| Score | What the presentation looks like |
|---|---|
| 4 | The main point is clear in the first minute, every slide supports it, the speaker talks to the audience instead of reading, stays within the time limit and answers questions with specifics. |
| 3 | The main point is clear, most slides support it, the speaker reads occasionally, runs slightly over or under time and answers most questions. |
| 2 | The main point takes effort to find, several slides drift off topic, the speaker reads much of the time and has trouble with questions. |
| 1 | No clear main point, slides are disorganized or mostly text, and the speaker reads throughout or stops well short of the time. |
Holistic rubrics are fast, which is why they are common for large-scale scoring and short responses. The downside, as both DePaul and Cult of Pedagogy point out, is that one number tells you little about what to fix, and the criteria cannot be weighted. They are also hard to use when a piece is strong in one area and weak in another.
Single-point rubric
A single-point rubric describes only what meeting the standard looks like, with open columns on either side for the teacher to write what is missing and where the work went beyond it. Here is a sample for a middle school lab report:
| Concerns (what is missing) | Criteria (the standard) | Evidence of exceeding |
|---|---|---|
| Question: The lab states a testable question and a prediction with a reason. | ||
| Method: Steps are listed in enough detail that a classmate could repeat them. | ||
| Data: Results appear in a labeled table or graph with units. | ||
| Conclusion: The conclusion answers the question using the data, and names one source of error. |
Teacher and writer Jennifer Gonzalez, who has championed the format on Cult of Pedagogy, traces the name to a 2000 conference presentation by Mary Dietz and notes that researcher Jarene Fluckiger published a study of it in 2010. Its appeal is that students read it: there is far less text than in a full analytic rubric. The cost is that the teacher writes more comments by hand.
Checklist
A checklist is the simplest rubric. Every item is either there or not. DePaul calls checklists “a distinct type of rubric” with only two performance levels, such as yes/no or pass/fail. Checklists work well for requirements that are not a matter of quality:
| Requirement | Yes | No |
|---|---|---|
| Title page with name, class and date | ||
| At least 750 words | ||
| Three or more sources cited | ||
| Works cited page in the required format | ||
| Submitted as a PDF by the deadline |
Many teachers combine the two: a checklist for the basic requirements and an analytic rubric for the quality of the work itself.
Developmental rubric
A developmental rubric tracks growth in a skill over time rather than grading one finished product. DePaul describes it as a kind of analytic rubric built on a theory of how a skill develops. You are more likely to meet one in a college program review or a teacher evaluation than on a homework assignment.
Holistic vs analytic rubric (and the other types) compared
| Type | How it scores | Best for | Main drawback |
|---|---|---|---|
| Analytic | A score for each criterion, added up | Essays, projects, lab reports; any work where specific feedback matters | Slower to write and to grade |
| Holistic | One overall score | Short responses, quick checks, large-scale scoring | Little feedback; criteria cannot be weighted |
| Single-point | Meets the standard or not, plus written comments | Drafts and feedback-heavy assignments | More handwritten comments for the teacher |
| Checklist | Yes or no for each item | Formatting and required components | Cannot measure quality |
| Developmental | A level of development on each dimension | Tracking growth over a program | Hard to design well |
How teachers grade with a rubric
With an analytic rubric, a teacher reads the work, finds the descriptor in each row that best matches it, circles or highlights that box, and adds up the points. Suppose your essay is scored on the sample rubric above:
| Criterion | Level earned | Points |
|---|---|---|
| Thesis | 3: Meets | 3 |
| Evidence | 4: Exceeds | 4 |
| Organization | 3: Meets | 3 |
| Counterargument | 2: Approaching | 2 |
| Mechanics | 3: Meets | 3 |
| Total | 15 out of 20 |
Fifteen out of 20 is 75%. If the teacher counts the thesis and evidence rows double, the maximum becomes 28 points, you earn 3×2 + 4×2 + 3 + 2 + 3 = 22, and your score is about 79%.
Not every teacher converts a rubric score straight into a percentage. On a four-level scale, straight math makes a 2 (“Approaching”) worth only 50%, a failing grade in many schools, even though the work is partly there. So some teachers and schools use a conversion chart instead, for example treating a 3 as a B and a 2 as a C. Check your syllabus or ask how rubric scores turn into grades in your class; our guide to the syllabus explains where grading policies usually live.
How to use a rubric to get a better grade
A rubric is the closest thing to seeing the test before you take it, so use it at every stage, not just after the grade comes back.
- Before you start: read the top level of every row. That column describes the work your teacher hopes to see. Turn each descriptor into a mini to-do list.
- While you draft: look for criteria you might forget, such as a counterargument or a source-error discussion. Rows that are easy to overlook are where easy points slip away.
- Before you submit: score your own work honestly against each row. If you would give yourself a 2 anywhere, fix that row first; it is usually the cheapest improvement.
- After grading: compare your scores with your self-scores. Where they differ, read the descriptor again and ask your teacher what moved you down a level.
If a grade seems wrong, the rubric is also the best way to raise it politely: point to the row and the descriptor rather than to the total. Our guide to whether teachers give bad grades on purpose walks through how to question a grade, and how to email a professor covers the message itself.
How to make a rubric
If you teach, or you have been asked to build a rubric for a group project, these steps keep it short and usable:
- 1. Start with the goal. Write down the two or three skills the assignment is meant to show.
- 2. Pick three to six criteria. Each criterion should be something you can see in the work, such as “uses evidence” rather than “tries hard.”
- 3. Choose the levels. Four levels avoid a safe middle score; three are quicker to use.
- 4. Write the top level first. Describe excellent work for each criterion, then the acceptable level, then work backward.
- 5. Make descriptors concrete. Use counts and observable features (“two or more sources”) instead of adjectives (“good sources”).
- 6. Decide the points. Weight the criteria that matter most and decide how scores convert to grades before you grade anything.
- 7. Test it. Score two or three real samples, or have a colleague score the same piece. If you disagree, tighten the wording.
- 8. Share it early. Hand it out with the assignment so students can use it while they work.
What makes a good rubric?
A good rubric is short enough to read, specific enough that two graders reach the same score, and focused on the skills the assignment is really about. Signs of a weak rubric: descriptors that only swap adjectives (“excellent,” “good,” “fair”), criteria that overlap so one mistake costs points twice, and rows for things students were never taught.
Rubric vs checklist vs grading scale vs criteria
| Term | What it is | Example |
|---|---|---|
| Rubric | Criteria plus descriptions of quality at each level | The essay rubric above |
| Checklist | Required items marked present or missing | “Works cited page included: yes/no” |
| Grading scale | How percentages convert to letter grades | 90-100 = A, 80-89 = B |
| Criteria | The qualities being judged; the rows of a rubric | Thesis, evidence, organization |
| Matrix | Any grid; a rubric is one kind | A rubric laid out as rows and columns |
A rubric tells you how your work earns points; a grading scale tells you what letter those points become. If you are trying to figure out whether a score is passing, see our guide to what counts as a good GPA, which includes the letter-grade conversion.
Rubric examples by assignment
Criteria change with the assignment. These are common rows teachers use, which you can mix and match when building your own rubric or predicting what yours will look for:
| Assignment | Typical criteria |
|---|---|
| Persuasive or argumentative essay | Thesis, evidence, reasoning, counterargument, organization, mechanics, citations |
| Research paper | Research question, source quality, synthesis, analysis, citation format |
| Oral presentation | Main point, organization, visuals, delivery, time management, answering questions |
| Group project | Final product quality, individual contribution, collaboration, meeting deadlines |
| Science lab report | Question and hypothesis, method, data presentation, analysis, conclusion, sources of error |
| Elementary writing | Ideas, order (beginning, middle, end), word choice, capitals and punctuation, neatness |
| Art or design project | Use of the technique taught, composition, originality, craftsmanship, artist statement |
For younger students, teachers often replace numbers with pictures or plain words (“Wow!”, “Got it”, “Almost”, “Not yet”) and keep a rubric to three or four rows.
What is a rubric in AI?
The same idea is now used to grade artificial intelligence. Instead of a teacher scoring an essay, a rubric lists what a good answer from an AI model should include or avoid, and each answer is scored against it. OpenAI’s HealthBench, published May 12, 2025, is a well-documented example: 5,000 realistic health conversations, each with its own rubric written by physicians, for a total of 48,562 unique criteria. Each criterion carries a point value, and a grading model checks which criteria a response meets.
Rubrics also show up in hiring, where interviewers score candidates against the same set of criteria, and in grant and contest judging. The principle is the same everywhere: decide what good looks like before you start judging.
Frequently asked questions
What is a rubric in simple terms?
A rubric is a scoring guide for an assignment. It lists what the teacher is looking for (the criteria) and describes what weak, average and strong work looks like for each one, so you know before you start how your work will be graded.
What is a rubric used for?
Teachers use rubrics to grade consistently, to explain a grade without writing long comments, and to show students what good work looks like in advance. Students use them to plan an assignment, check a draft before turning it in, and understand the feedback afterward.
Is it rubric or rubrics?
Both. Rubric is the singular and rubrics is the plural: one rubric for an essay, three rubrics for three projects. Some teachers also say “grading rubric” or “scoring rubric,” which mean the same thing.
What is a rubric score?
It is the score you earn on each criterion of a rubric, or the total of those scores. On a four-level rubric with five criteria, for example, you might earn 3, 4, 3, 2 and 4 for a total of 16 out of 20. Your teacher then turns that total into a grade, usually a percentage or a letter.
What does a rubric measure?
A rubric measures the quality of specific parts of a piece of work, such as the argument, the evidence, the organization and the mechanics of an essay. It does not measure effort or time spent unless the teacher includes those as criteria.
What are the four parts of a rubric?
Most rubrics have criteria (what is being graded), performance levels (the rating scale, such as 1 to 4), descriptors (what work at each level looks like) and points or weights (how much each criterion counts). Holistic rubrics and checklists leave some of these out.
What is the difference between a holistic and an analytic rubric?
A holistic rubric gives one overall score based on a general description of each level. An analytic rubric scores each criterion separately and adds them up. Holistic is faster to grade; analytic gives more useful feedback and lets the teacher weight some criteria more than others.
Is a checklist a rubric?
A checklist is the simplest kind of rubric. Each item is just met or not met, with no levels in between. DePaul University’s teaching center describes checklists as a distinct type of rubric with only two performance levels.
Can a teacher grade outside the rubric?
A teacher should grade against the criteria they published. If a grade seems to depend on something the rubric never mentions, it is reasonable to ask politely which criterion it falls under. Some rubrics include a general criterion, such as following directions, that covers more ground than students expect.
Where does the word rubric come from?
From the Latin rubrica, meaning red ocher, from ruber, “red.” Medieval scribes wrote headings and instructions in red ink, so those red headings came to be called rubrics. The word later came to mean a rule or category, and teachers adopted it for a set of grading criteria.
Josh Hutcheson — Editor, PriorityLearn
Josh researches, writes, and updates the answers on PriorityLearn, checking each one against current tools, official sources, and real school policies — and flagging what varies by state or district. About PriorityLearn →
