Claims, Evidence & Reasoning
Learn to write an arguable, specific claim, choose evidence that proves that exact claim, and write reasoning that links the two — with a full worked paragraph.
What you'll do in this lesson
A voice-first session with the Crimsora tutor on Claims, Evidence & Reasoning, then targeted practice and FRQs — with the tutor adapting to where you get stuck.
What this lesson covers
Almost every essay you write this year asks you to argue something. The trouble is that arguments fall apart in three predictable places: the claim is really just a topic, the evidence is only about the same subject rather than proof of the point, or the reasoning — the sentences that explain how the quote proves the claim — never gets written at all.
This lesson takes those three moves one at a time. You will learn a quick test for whether a claim is arguable, a fit test for whether a piece of evidence supports your particular claim, and a set of sentence starters for reasoning that pushes past "this quote shows." By the end you should be able to build a body paragraph where a reader can trace the logic from the quotation all the way back to your thesis without guessing.
This lesson takes those three moves one at a time. You will learn a quick test for whether a claim is arguable, a fit test for whether a piece of evidence supports your particular claim, and a set of sentence starters for reasoning that pushes past "this quote shows." By the end you should be able to build a body paragraph where a reader can trace the logic from the quotation all the way back to your thesis without guessing.
What Makes a Claim Arguable and Specific
A claim is a statement someone could reasonably disagree with, and that you could support with evidence. Two things disqualify a sentence from being a claim.
The first is that it is a fact. "Romeo and Juliet die at the end of the play" cannot be argued; it can only be checked. The second is that it is pure taste. "The play is boring" gives a reader nothing to argue against except your mood. Between those two poles sits the useful zone: a statement about meaning, cause, effect, or value that a thoughtful person could contest.
Specificity is a separate requirement. Compare these three sentences about the same text.
The third version is better because it is riskier. It tells the reader exactly what evidence would prove it and what evidence would refute it. That is the real test: if you cannot imagine a piece of evidence that would count against your claim, your claim is probably not a claim.
A useful revision habit is to add a because-clause and then delete it. "Curfews for teenagers are unfair because they punish an entire age group for the behavior of a few" — the because-clause forces the sentence to take a position. You can keep it or cut it, but writing it reveals whether you actually had a position to begin with.
The first is that it is a fact. "Romeo and Juliet die at the end of the play" cannot be argued; it can only be checked. The second is that it is pure taste. "The play is boring" gives a reader nothing to argue against except your mood. Between those two poles sits the useful zone: a statement about meaning, cause, effect, or value that a thoughtful person could contest.
Specificity is a separate requirement. Compare these three sentences about the same text.
| Sentence | Problem or strength |
|---|---|
| The novel is about identity. | Topic, not claim. Nobody disagrees. |
| The novel shows that identity is complicated. | Arguable but vague; "complicated" commits to nothing. |
| The novel argues that Junior's identity becomes stable only after he accepts that he belongs fully to neither community. | Arguable and specific; names who, what, and under what condition. |
A useful revision habit is to add a because-clause and then delete it. "Curfews for teenagers are unfair because they punish an entire age group for the behavior of a few" — the because-clause forces the sentence to take a position. You can keep it or cut it, but writing it reveals whether you actually had a position to begin with.
Evidence That Fits the Claim, Not Just the Topic
Evidence is the specific material — a quotation, a statistic, a detail from a text, a documented example — that a reader can verify. The most common failure in student arguments is not missing evidence. It is evidence that shares a topic with the claim but does not test it.
Suppose your claim is that later school start times improve student attendance. A study reporting that teenagers naturally fall asleep later is about the same topic, but it does not support that claim; it supports a different claim about biology. Evidence that actually fits would report attendance figures before and after a district changed its schedule.
Use a two-question fit test on every piece of evidence before you commit to it. First, does this evidence mention the specific thing my claim asserts — the same group, the same outcome, the same cause? Second, if this evidence turned out to be false, would my claim be weaker? If the answer to the second question is no, the evidence is decorative.
Three other habits matter. Keep quotations short and surgical; a four-line block quote usually means you have not decided which words matter. Introduce evidence with context so the reader knows who is speaking and when — "After Atticus loses the case, Jem says…" does more work than a bare quotation dropped into the paragraph. And when evidence is numerical, name the source and the population, because a number without a population is not verifiable.
One more distinction: an example is not the same as evidence from a source. Hypothetical examples ("imagine a student who…") can illustrate reasoning, but they cannot prove anything, because you invented them. Save them for clarifying a point you have already supported.
Suppose your claim is that later school start times improve student attendance. A study reporting that teenagers naturally fall asleep later is about the same topic, but it does not support that claim; it supports a different claim about biology. Evidence that actually fits would report attendance figures before and after a district changed its schedule.
Use a two-question fit test on every piece of evidence before you commit to it. First, does this evidence mention the specific thing my claim asserts — the same group, the same outcome, the same cause? Second, if this evidence turned out to be false, would my claim be weaker? If the answer to the second question is no, the evidence is decorative.
Three other habits matter. Keep quotations short and surgical; a four-line block quote usually means you have not decided which words matter. Introduce evidence with context so the reader knows who is speaking and when — "After Atticus loses the case, Jem says…" does more work than a bare quotation dropped into the paragraph. And when evidence is numerical, name the source and the population, because a number without a population is not verifiable.
One more distinction: an example is not the same as evidence from a source. Hypothetical examples ("imagine a student who…") can illustrate reasoning, but they cannot prove anything, because you invented them. Save them for clarifying a point you have already supported.
Reasoning: The Sentences That Connect the Two
Reasoning — sometimes called the warrant — is the explanation of why this evidence proves this claim. It is the part readers most often find missing, because writers can see the connection in their own heads and assume it is on the page.
A quick diagnostic: cover your claim with your hand and read only the evidence and the sentences after it. Could a stranger reconstruct the claim from what remains? If not, your reasoning is doing too little.
Weak reasoning restates the quotation in different words. "This shows that Jem is upset" after a quotation in which Jem is visibly upset adds nothing. Strong reasoning does at least one of these jobs: it defines a key word from the evidence, it explains why the detail is surprising or significant, it identifies the assumption that makes the evidence count, or it rules out a competing interpretation.
Aim for at least two sentences of reasoning for every one sentence of evidence. That ratio feels strange at first, because evidence looks like the substantial part of a paragraph. In practice, the evidence is the raw material and the reasoning is the argument.
A quick diagnostic: cover your claim with your hand and read only the evidence and the sentences after it. Could a stranger reconstruct the claim from what remains? If not, your reasoning is doing too little.
Weak reasoning restates the quotation in different words. "This shows that Jem is upset" after a quotation in which Jem is visibly upset adds nothing. Strong reasoning does at least one of these jobs: it defines a key word from the evidence, it explains why the detail is surprising or significant, it identifies the assumption that makes the evidence count, or it rules out a competing interpretation.
| Move | Sample opener |
|---|---|
| Interpret a word choice | The word "surrender" matters here because it frames the decision as a defeat rather than a compromise. |
| Name the assumption | This figure only supports the claim if the two schools serve similar populations, which the report confirms. |
| Rule out an alternative | One might read this silence as indifference, but the earlier scene rules that out. |
| Connect to the claim | Because the character repeats this phrase only around adults, the novel presents his politeness as performance rather than belief. |
Assembling and Troubleshooting a Body Paragraph
A reliable body paragraph runs claim, context, evidence, reasoning, and then a link back to the larger argument. The link matters: it is where you remind the reader what this paragraph contributed to the essay as a whole.
When a paragraph is not working, the problem is almost always locatable. Use this table to diagnose it.
One more troubleshooting move: if you cannot find evidence that fits your claim, that is information, not failure. It usually means the claim is too broad or slightly wrong. Narrowing a claim to match the evidence you actually have — from "the author condemns technology" to "the author condemns the way notification design interrupts sustained attention" — makes the paragraph both easier to write and more convincing.
Finally, sequence your paragraphs so each claim builds on the one before it. Readers judge an argument by whether the parts add up, not by whether each part is individually defensible.
When a paragraph is not working, the problem is almost always locatable. Use this table to diagnose it.
| Symptom | Likely cause | Fix |
|---|---|---|
| Reader says "so what?" | Claim is a fact or a topic | Add a because-clause and take a position |
| Paragraph reads like plot summary | Evidence is present, reasoning is missing | Add two sentences explaining why the detail matters |
| Quotation feels dropped in | No context sentence | Name the speaker, moment, or source before the quote |
| Evidence and claim feel unrelated | Shared topic, not genuine support | Apply the fit test; find different evidence or narrow the claim |
| Paragraph wanders into a new idea | Two claims in one paragraph | Split into two paragraphs, each with one claim |
Finally, sequence your paragraphs so each claim builds on the one before it. Readers judge an argument by whether the parts add up, not by whether each part is individually defensible.
Key terms
- Claim.
- A debatable statement that the writer will support with evidence; a reasonable person could disagree with it and evidence could count against it.
- Arguable.
- Describes a statement that is neither a checkable fact nor a matter of pure personal taste, so it genuinely requires support.
- Evidence.
- Verifiable material — quotations, textual details, data, documented examples — offered to support a claim.
- Reasoning (warrant).
- The explanation of why a specific piece of evidence supports a specific claim; the logical bridge between the two.
- Fit test.
- Checking whether evidence addresses the exact assertion of the claim rather than merely sharing its subject matter.
- Context sentence.
- A brief lead-in identifying the speaker, moment, or source of a quotation so the reader can interpret it.
- Qualifier.
- A word or phrase such as "often," "in most cases," or "for first-year students" that limits a claim to what the evidence can actually support.
- Topic.
- The subject an argument is about; unlike a claim, it makes no assertion and cannot be agreed or disagreed with.
Worked example
You are writing an argument essay responding to this prompt: Should your school replace its printed student handbook with an app? Your research notes contain three items. Item A: a district survey reporting that 88 percent of students own a smartphone. Item B: a neighboring district's report that discipline referrals for handbook violations fell by about one third in the year after it launched a searchable handbook app, with no other policy changes. Item C: a quotation from a school administrator saying, "Students never read the printed handbook anyway." Write a claim, choose the best evidence, and write the reasoning.
Start with the claim. "Our school should think about a handbook app" is not arguable — nobody objects to thinking about things. "The handbook app is good" is arguable but vague. A specific, arguable version: "Our school should replace the printed handbook with a searchable app, because students break rules they cannot easily look up." Notice the because-clause commits to a cause, which tells you exactly what evidence you need: something linking ease of access to rule-following.
Now apply the fit test to each item. Item A shares the topic but tests nothing about rule-following; even if the figure were wrong, the claim about access and violations would stand. It is useful background for feasibility, not proof. Item C is an opinion, not evidence — it asserts the very thing you need to demonstrate, and "never" is unsupported. Item B is the fit: it reports the specific outcome the claim predicts, in a comparable setting, with the confounding variables addressed.
Draft the paragraph. Claim: Our school should replace the printed handbook with a searchable app, because students break rules they cannot easily look up. Context and evidence: In the year after the neighboring district launched its searchable handbook, discipline referrals for handbook violations dropped by roughly one third, and the district reported no other policy changes during that period. Reasoning: That drop matters because the rules themselves did not change — only how quickly students could find them, which points to access rather than attitude as the cause of many violations. If students were simply choosing to break rules they already knew, a change in format would not move the numbers. Link: A format that students can actually search therefore addresses the cause of the problem rather than its symptom.
Finally, add a qualifier if the evidence is thinner than the claim. "Many violations" is safer than "violations," because one district's results cannot prove a universal cause.
Now apply the fit test to each item. Item A shares the topic but tests nothing about rule-following; even if the figure were wrong, the claim about access and violations would stand. It is useful background for feasibility, not proof. Item C is an opinion, not evidence — it asserts the very thing you need to demonstrate, and "never" is unsupported. Item B is the fit: it reports the specific outcome the claim predicts, in a comparable setting, with the confounding variables addressed.
Draft the paragraph. Claim: Our school should replace the printed handbook with a searchable app, because students break rules they cannot easily look up. Context and evidence: In the year after the neighboring district launched its searchable handbook, discipline referrals for handbook violations dropped by roughly one third, and the district reported no other policy changes during that period. Reasoning: That drop matters because the rules themselves did not change — only how quickly students could find them, which points to access rather than attitude as the cause of many violations. If students were simply choosing to break rules they already knew, a change in format would not move the numbers. Link: A format that students can actually search therefore addresses the cause of the problem rather than its symptom.
Finally, add a qualifier if the evidence is thinner than the claim. "Many violations" is safer than "violations," because one district's results cannot prove a universal cause.
Practice questions
Which of the following is the most arguable and specific claim about a novel your class has read?
- The novel takes place during the Great Depression and follows two migrant workers.
- The novel explores several important themes about friendship and loneliness.
- The novel presents loneliness as a product of economic conditions rather than personal failing, since every isolated character is isolated by work.
- The novel is one of the saddest books ever written.
Answer: The novel presents loneliness as a product of economic conditions rather than personal failing, since every isolated character is isolated by work.
The first option is a summary of verifiable facts, so there is nothing to argue. The second names topics but asserts nothing about them — a reader cannot disagree with "explores themes." The fourth is a matter of personal reaction with no evidence that could settle it. The third takes a contestable position (economic cause rather than personal failing), names the pattern that would prove it, and could be refuted by finding an isolated character whose isolation has nothing to do with work.
A student claims: "Requiring community service hours for graduation makes students more likely to volunteer as adults." As evidence, the student cites a survey showing that 72 percent of students at their school say community service is important. Explain why this evidence does not fit the claim, and describe what evidence would fit.
Answer: The survey measures current attitudes, while the claim asserts a long-term behavioral effect caused by a requirement; fitting evidence would compare adult volunteering rates between people who did and did not face the requirement.
Run the fit test. First question: does the evidence mention what the claim asserts? The claim is about adults volunteering and about a requirement causing that behavior. The survey is about teenagers' stated opinions — same topic, different assertion. Second question: if the survey were false, would the claim be weaker? Not really, because students could think service is unimportant and still volunteer later. Fitting evidence would track people over time, comparing adult volunteering rates for graduates of schools with and without the requirement, ideally controlling for differences between those populations. The gap here is the classic one: attitude data being used to support a causal claim about behavior.
Rewrite this weak reasoning so it does real work. Claim: The narrator's politeness toward his boss is a performance rather than genuine respect. Evidence: The narrator says, "Of course, sir, whatever you think is best." Weak reasoning: "This shows the narrator is being polite to his boss."
Answer: A sample strong revision: The phrase "whatever you think is best" surrenders judgment entirely, which is striking because the narrator has just spent a page explaining, to himself, exactly why the plan will fail. Holding a clear opinion and voicing its opposite is not respect but concealment, and the formal "sir" — a word he uses nowhere else in the chapter — marks the moment as a script he is performing.
The weak version restates the quotation, so a reader who covered the claim could not reconstruct it. The revision does three of the reasoning jobs at once: it interprets specific word choice ("whatever you think is best," "sir"), it points out what makes the detail surprising (the contrast with his private thoughts), and it rules out the competing reading (respect) by naming the evidence that defeats it. Any revision that performs at least two of those moves and reaches the word "performance" from the quotation is doing the work.
FAQ
- How do I know if my claim is arguable?
- Try to write a sentence that contradicts it. If the contradiction sounds silly ("The novel is not about identity"), your claim is a topic or a fact. If the contradiction sounds like something a classmate could defend in discussion, you have a claim. A second check: imagine a piece of evidence that would prove you wrong. If none exists, there is nothing to argue.
- How much evidence do I need in one paragraph?
- Usually one or two well-chosen pieces, not four. Paragraphs with a pile of quotations almost always have thin reasoning, because there is no room left to explain anything. A good target is one to two sentences of evidence followed by two or three sentences of reasoning, then a link back to your overall argument.
- What is the difference between evidence and reasoning?
- Evidence is material a reader could go check — a quotation, a number, a documented event. Reasoning is your explanation of why that material proves your point; it exists only in your sentences and cannot be verified elsewhere. If you deleted your reasoning, the evidence would still be true, but the argument would disappear.
- Can I use my own experience as evidence?
- Sometimes, and carefully. Personal experience can support a claim about what something feels like or about a specific situation you witnessed, and it works well as an opening illustration. It cannot support a claim about what is generally true, because one experience does not establish a pattern. If your claim uses words like "most," "usually," or "students in general," you need evidence about a group, not about yourself.
Learn this with a teacher, not a page
The Crimsora tutor teaches Claims, Evidence & Reasoning live — explaining on a whiteboard, asking you questions, and adapting to where you get stuck.