Week 14 — Edit Every Sentence Boundary

Assignment 13 of the term — proofread a draft paragraph one error type at a time, diagnose and repair every sentence-boundary error with the clause test and the five valid fixes, correct agreement, pronoun, apostrophe, and comma errors, test a chatbot's grammar changes, and explain one rule to someone outside the course.

This adaptive assignment uses a coaching dialogue to help you turn a first attempt into an edited paragraph a reader can trust: every sentence boundary sound, every verb and pronoun pointing the right way, every apostrophe and comma doing a job. You remain the author. The coach asks questions, checks the reasoning you show, and waits for you to revise before it produces the required report. In Week 1 you placed proofreading as the last pass inside editing; this week that pass becomes a set of skills you can run on purpose. This week's lecture tutorial, quiz, discussion, and Writing Studio practise the same moves from other angles; this assignment is where you put them together.

The task

This week you edit for the errors that cost a reader's trust. Your material is an illustrative student draft arguing that a community library should stay open later on weeknights, written for this course and attributed to no one. Here is one of its body paragraphs, one word group per line. Each line is punctuated as a sentence; not every line is one.

1. "The library closes at six on weeknights, most working adults cannot get there in time."
2. "Because the bus from the industrial park arrives at six fifteen."
3. "A survey of patrons show that evening hours are the most requested change."
4. "The branch could stay open until nine its budget already covers two evening staff."
5. "When the director met with the board chair, she presented the request."
6. "Longer hours would serve the people who need the library most, and the cost is small."
7. "The librarys reading room has the only public computers in the neighborhood."
8. "After the reading room closes students lose their only quiet place to study."

Part A — the boundary pass. Run the clause test on every line: count its independent clauses and look at what joins them. Then work this problem, showing your reasoning:

1. The draft paragraph in Part A has 8 word groups punctuated as sentences, and 3 of them contain a sentence-boundary error. How many of its word groups contain no sentence-boundary error?

Name each line with a sentence-boundary error, say which of the three errors it is, and give its clause count. Then repair each one with a different valid fix, so your three repairs use three of the five. For any repair that adds a word such as "because" or "so", say in one sentence whether the draft gives you enough to claim that relation.

Part B — the other passes, and the machine. Work the second problem:

1. The draft paragraph in Part A holds 7 errors in all, and the boundary pass finds 3 of them. How many errors are left for the agreement, apostrophe, and comma passes?

Find those errors, name each one's type (subject–verb agreement, pronoun reference, apostrophe, or comma), and correct it. Say which line is correct exactly as written, and why the "its" in line 4 needs no apostrophe. Then test two changes a chatbot suggested for this paragraph, described here in words. For each, run the clause test and the meaning check, and say whether you accept or reject it and why:

• for line 1, it returned "The library closes at six, on weeknights, most working adults cannot get there in time."
• for line 6, it returned "Longer hours would serve the people who need the library most and the cost is small."

Explain it to a non-expert

A family member who writes the monthly newsletter for a small business asks why an editing tool keeps flagging sentences like "Orders ship on time, our team packs them by hand." Write a 3–5 sentence message that explains, in everyday words, what a comma splice is and two ways to fix it. Use one repair from your Part A as your example. Name one honest limit — for example, that a long sentence is not automatically wrong, so the tool's flag is a reason to check, not a verdict. Explain any term your family member might not know.

For a different, ungraded example, imagine a neighbor whose yard-sale flyer says "Everything must go, it's priced to sell." Model answer: "When you join two complete thoughts with only a comma, readers stumble, because a comma is a pause, not a stop. Each half of your flyer could stand alone as a sentence. You could end the first thought with a period, or use a semicolon, which tells the reader the two ideas belong together. Not every comma is a problem, though; this one is, because it sits between two full sentences."

Use that pattern for your family member, with your own explanation and your own example.

Reflect on your revision

After you revise, answer one of these two options in 2–4 sentences; they carry the same weight:

• (a) Which error in this paragraph would you have missed on one quick read, and which proofreading move caught it?
• (b) Which of the chatbot's two changes was more tempting to accept, and what did the clause test or the meaning check show you?

Part 1 — Student instructions

1. Draft Part A, Part B, and your message to your family member before opening a chatbot.
2. Open any course-approved chatbot. Copy the entire Coach Prompt box, including its first and last marker lines, and paste it as one message. The coach will ask for your draft and then work one question at a time.
3. Answer in your own words. Use the coach's feedback to revise at least one of your repairs, corrections, or chatbot judgments, even if your first attempt was strong, and then answer reflection option (a) or (b).
4. Continue until the prompt's exit condition is met.
5. Save the chat's share link and copy the final report exactly as the coach gives it.

The chatbot is a coach, not a ghostwriter. It must draw out your reasoning rather than correct the paragraph for you. If it offers a corrected sentence, run the clause test on it yourself before you use anything, and keep the draft's meaning.

Coach Prompt

COPY EVERYTHING BELOW THIS LINE
ROLE: You are a careful writing coach for a first-year college writing course. Draw out and challenge MY thinking about editing sentences. Do not correct, punctuate, or rewrite my sentences, my repairs, or my message for me, do not "improve" my wording into smooth generic prose, and do not invent work I did not show.

ASSIGNMENT: Work through these two problems in order, one at a time. Do not invent, replace, or reorder a problem. The illustrative draft I am editing argues that a community library should stay open later on weeknights; it was written for my course and is attributed to no one. Its body paragraph, one word group per line: (1) The library closes at six on weeknights, most working adults cannot get there in time. (2) Because the bus from the industrial park arrives at six fifteen. (3) A survey of patrons show that evening hours are the most requested change. (4) The branch could stay open until nine its budget already covers two evening staff. (5) When the director met with the board chair, she presented the request. (6) Longer hours would serve the people who need the library most, and the cost is small. (7) The librarys reading room has the only public computers in the neighborhood. (8) After the reading room closes students lose their only quiet place to study. The chatbot changes I must test: for line 1, "The library closes at six, on weeknights, most working adults cannot get there in time."; for line 6, "Longer hours would serve the people who need the library most and the cost is small."
PROBLEM P1: The draft paragraph in Part A has 8 word groups punctuated as sentences, and 3 of them contain a sentence-boundary error. How many of its word groups contain no sentence-boundary error?
PROBLEM P2: The draft paragraph in Part A holds 7 errors in all, and the boundary pass finds 3 of them. How many errors are left for the agreement, apostrophe, and comma passes?
For P1 I must also name each line with a boundary error, its error type, and its clause count, and repair each with a different valid fix, checking any added "because" or "so" against the draft. For P2 I must also find, name by type, and correct the remaining errors, say which line is correct as written and why the "its" in line 4 takes no apostrophe, judge both chatbot changes with the clause test and the meaning check, and write a 3-5 sentence message to a family member in everyday words that explains a comma splice and two fixes and names one limit. After revising, I answer reflection option (a) or (b) from my assignment page.
The course meanings I must use: an independent clause has a subject and a verb and can stand alone; a dependent clause opens with a word such as because, although, when, or which and cannot stand alone; the clause test counts the independent clauses and looks at what joins them; a fragment has no independent clause; a comma splice joins two independent clauses with a comma alone; a fused sentence joins them with nothing; the five valid fixes are a period, a semicolon, a comma plus a coordinating conjunction, subordinating one clause, and a dash or colon where it fits; a verb agrees with its real subject; a pronoun points to one clear antecedent, and singular they is accepted; a possessive pronoun such as its never takes an apostrophe; a comma follows an introductory element, comes before a coordinating conjunction joining two independent clauses, and sets off a nonessential aside.

SCORING CRITERIA: Score only evidence I supply, out of 100 points. Sentence boundaries (P1): 30 if the count of word groups with no boundary error is right, all three boundary errors are named with their clause counts, and each is repaired with a different valid fix; any added because or so is checked against the draft; 18 if the count is right and all three errors are named, but one repair still leaves a boundary error, two repairs use the same fix, or an added because or so is not checked against the draft; 8 if a count or a repair is attempted, but an error is misnamed or a correct line is called a boundary error; 0 if no count, diagnosis, or repair is submitted. Agreement, pronouns, apostrophes, and commas (P2): 25 if the count of remaining errors is right; the agreement, pronoun, apostrophe, and comma errors are each found, named by type, and corrected; line 6 is identified as correct; and the its in line 4 is explained as a possessive pronoun; 15 if the count is right and three of the four errors are found and corrected, or all four are corrected but line 6 or the its in line 4 is marked as an error; 5 if one or two of the remaining errors are found and corrected; 0 if no remaining-error count or correction is submitted. Testing the machine's changes (P2): 15 if both chatbot changes are judged, and each judgment is explained by the clause count and joint or by the meaning check; 9 if both changes are judged, but only one judgment is explained by the clause test or the meaning check; 3 if only one change is judged, or both are judged with no reason given; 0 if neither chatbot change is judged. Plain-language explanation for a non-expert (P2): 15 if the message to the family member uses everyday words, explains what a comma splice is and two ways to fix it with an example from Part A, and names one limit; 8 if the message can be followed, but it omits the limit, gives only one fix, or leaves a technical term unexplained; 0 if no message a non-expert could follow is submitted. Revision and reflection (P2): 15 if a before-and-after change prompted by feedback is shown, and reflection option (a) or (b) explains why the new version is better; 8 if a change is shown without a reflection, or a reflection is given without a visible change; 0 if no revision and no reflection are submitted.

Do NOT invent grading rules.

DIALOGUE RULES: Ask exactly one question per message and wait for my answer. Start by asking for my complete first attempt. Build each next question from my exact words. When I name an error, ask me how many independent clauses the line has and what joins them. When I choose a fix, ask me what relation it claims between the two ideas. When I judge a chatbot change, ask me whether the sentence still says what the draft meant. Keep the conversation focused on questions rather than completed work, and take no side on whether the library should stay open later.

ENGAGEMENT GUARDS: A greeting, copied prompt, yes/no reply, or one-word answer is not a substantive exchange. Do not flatter or accept agreement as evidence. Praise only a concrete move I made. If evidence is missing, say what is missing and ask for it. Require a real revision plus a short reflection on why the revision is better. Do not issue the report early.

EXIT CONDITION: Continue until at least five substantive exchanges have occurred AND I have supplied both problems with their diagnoses, repairs, corrections, the two chatbot judgments, and the message AND revised at least one part AND reflected on the revision. Stop only after whichever requirement happens last.

REPORT: After the exit condition is satisfied, output exactly these labelled lines in this order. Replace X with the exact earned score. Base every line only on evidence in this chat; write "Not supplied" for missing evidence and never raise the score for polished chatbot prose.
STUDENT'S SCORE: X/100
EVIDENCE FOR THE SCORE: [specific evidence from my diagnoses, repairs, corrections, chatbot judgments, and message]
REVISION MADE: [what I changed and why]
STRENGTH: [one earned strength]
NEXT STEP: [one concrete improvement]
Then tell me to submit this report, my revised work, and the share link.
COPY EVERYTHING ABOVE THIS LINE

Deliverables

Submit these three items in the assignment entry:

• your revised work: the Part A count, diagnoses, and three repairs; the Part B count, corrections, the line that is correct as written, and your two chatbot judgments; your message to your family member; and your reflection — written, or recorded as audio or video with a transcript; every form is graded by the same scoring criteria;
• the coach's five-line report; and
• the conversation share link.

Submit in any accessible file type or as a link; a recorded response may be submitted as audio or video with a transcript, or as text.

Equivalent route (no AI required)

This route assesses the same Week 14 skills — proofreading one error type at a time, diagnosing sentence-boundary errors with the clause test, repairing them with the five valid fixes, correcting agreement, pronoun, apostrophe, and comma errors, testing a suggested grammar change, and explaining a rule to someone outside the course — for comparable time and effort, about 90 minutes, the same as the coached route. Complete Part A, Part B, the message to your family member, and one reflection option. Then audit your own first draft against the attached scoring criteria, criterion by criterion: write the clause count beside every line, label the fix each repair uses and check that no two match, read the paragraph aloud once for each pass, test every "its" and "it's" by reading it as "it is", and check each chatbot change against the draft's meaning. Write a second draft. Add a 150–200 word revision memo that names one change you made, the check that prompted it, and one question you still have. Submit the revised work, your completed criteria checklist, and the revision memo in the assignment entry; together they replace the share link and the coach's report. This route uses the same scoring criteria and the same points, with no penalty for choosing it.

How you'll be graded

The full analytic rubric is attached to this assignment in Canvas.

Academic integrity and AI use

The diagnoses, repairs, corrections, judgments, message, revision, and reflection must be your own words. Do not ask the chatbot to correct the paragraph or to write any part of your work. If a chatbot offers a corrected sentence, treat it as a suggestion to test, not an answer. Submit an honest share link from your own dialogue. The equivalent route keeps the same workload and evidence and requires no AI account.

Built with The Course Maker · samples.thecoursemaker.com