INVESTIGATION 4 OF 6

Explain what evidence licenses you to say

Build a paragraph that distinguishes observation, inference and value judgment, and tests the warrant connecting them.

Why learn this? Quotations and numbers need interpretation; their presence does not make a paragraph’s conclusion justified.

Loading saved English work…

Quick recap: A warrant connects evidence to a claim. An alternative explanation can weaken a causal inference even when the observation is accurate.

Choose your task depth

Choose either at any time. Your drafts stay with you.

1. Try — read before judging

Original fictional revision problem. The tutor log predates any policy trial. The passage response has not been implemented.

DRAFT: Participation rose from 8 speakers to 14, proving that requiring written preparation improves discussion.
LOG: The counts came from two weeks with different topics. They count speaking at least once, not the quality of discussion.
POLICY STATUS: No written-preparation requirement was in place in either week.
COURSE AIM: Students should use passages as evidence.
POSSIBLE REASON FOR A TRIAL: Writing a passage-based response creates an opportunity to practice that skill; practice alone does not guarantee improvement.

Go directly to the workbench

2. Predict — make a provisional choice

Which part of the draft is false attribution, and which part makes an unsupported leap about the outcome being measured? 0 / 16,000 characters · length is a storage limit, not a quality score

3. Experiment — test your reading

Annotate the draft and packet to distinguish direct observation, proposed mechanism and unsupported inference. Include one challenge annotation.
Write a paragraph with claim, evidence, interpretation and a limit. Identify the warrant in your reflection.
Preserve a baseline and revise the paragraph after considering a competing explanation. Do not invent new data to rescue the claim.

Your evidence board

Select a line, name its role and explain your reasoning. Labels record your interpretation; the site does not decide whether it is convincing.

Passage lineRole

DRAFT: Participation rose from 8 speakers to 14, proving that requiring written preparation improves discussion.

Annotation explanation

Working draft / artifact 0 / 16,000 characters · length is a storage limit, not a quality score

Comparison draft

Save your first version here before revising. It will remain unchanged while you edit the working draft.

Revision commentary: what changed, why, and what remains uncertain? 0 / 16,000 characters · length is a storage limit, not a quality score

4. Understand — connect choice and effect

Separate a reason for trying from proof of success

The draft attributes a rise in speakers to a requirement that had not started. This is more than a small-sample problem: the stated cause was absent. Even if it had been present, different topics, attendance or discussion formats could complicate the comparison.

Check the outcome as carefully as the cause. Speaking at least once is one operational measure of participation. It is not a direct measure of how well a student uses evidence, how much someone learns or whether a quiet student is thinking carefully. A number is only as useful as the relationship between the counted event and the question.

Build the paragraph in four moves. State a bounded claim. Introduce the relevant observation with attribution. Explain the warrant connecting it to the claim. Then test the inference against a limitation or alternative explanation. These moves need not always occupy four separate sentences, but the reasoning should be recoverable.

The course aim supplies a defensible rationale for a trial: a passage response offers practice in selecting and interpreting textual evidence. That is a mechanism you propose, not an observed benefit. To evaluate benefit, you would need suitable evidence of students’ reasoning and a comparison designed for the question.

Notice where values enter. You might prefer the method that offers more opportunities to practice, even before strong outcome evidence exists. Explain that preference and its cost. Do not disguise it as a proven causal result. A carefully limited recommendation can still be useful to a decision-maker.

Causal inference: A conclusion that one factor helped produce another, requiring more than their association.
Operational measure: The concrete observation used to represent an outcome, such as counting speakers to represent one aspect of participation.

Worked example — a possible revision, not the only answer

Before

The increase proves written preparation works.

After

The log cannot evaluate a requirement that had not begun. A stronger reason for a trial comes from alignment with the course aim: a passage response asks students to practice textual evidence. Whether that practice improves discussion remains an empirical question.

The revision removes an impossible causal attribution and separates a curricular rationale from an outcome claim.

Return to your workbench and reconsider a choice

5. Apply — make something and try a new case

Your main artifact

Make three explained annotations and write a 180–250-word paragraph. Underline the limit you added.

Keep your artifact in the working draft above. Suggested lengths help you plan; they do not measure quality.

Unfamiliar passage

A club’s newsletter readership increased during the month before its new design launched. An editor attributes the increase to the redesign.

Repair the causal claim and propose a comparison that would actually address the design’s effect. State a remaining limitation. 0 / 16,000 characters · length is a storage limit, not a quality score

0 saved attempt(s).

6. Continue — record work without claiming mastery

Completion means you worked through this investigation. Your interpretation and writing still need judgment using the project rubric; no automatic proficiency score is awarded.

Continue: Change the genre without changing the facts

Choose another lesson · Saved English work and export