Revise a conclusion without hiding the change
Responsible communication distinguishes what is established, uncertain and corrected.
Your task: Write a correction memo preserving the original claim, explaining the exclusion and stating what evidence would permit a new estimate.
Open the investigation · Read the explanation · Your saved evidence
Starting recap and vocabulary
Read a record as a claim with a creator, purpose and scope. A count, interpretation and value judgment are different kinds of evidence. Percent means per hundred; percentage points compare two percentages.
- provenance
- The origin and subsequent transmission of a record or claim
- teaching record
- An authored fictional example designed to make reasoning inspectable; not evidence that an event occurred.
What is the defensible corrected claim?
Write a correction memo preserving the original claim, explaining the exclusion and stating what evidence would permit a new estimate.
Use the supplied records to complete this task: Write a correction memo preserving the original claim, explaining the exclusion and stating what evidence would permit a new estimate. Explain your reasoning in a short paragraph or an annotated diagram with a text equivalent.
Make a provisional claim
Loading saved evidence…
Inspect evidence and test an explanation
The investigation controls appear when your local evidence is ready.
Connect evidence and mechanism
Revise a conclusion without hiding the change
Responsible communication distinguishes what is established, uncertain and corrected. A correction should identify the original error, its effect and the revised claim. Changing a headline silently makes the evidence trail harder to inspect.
Inspect the case
Teaching draft says a program cut waiting times by half, using averages of 20 and 10 minutes.
Teaching audit finds the second sample excluded appointments over 30 minutes while the first included them. Raw records for excluded cases are unavailable.
Both records are original fictional teaching materials. They are not historical primary sources and do not report real participants.
Work through the mechanism
The numerical comparison is not like-for-like. Without the excluded records, a corrected overall average cannot honestly be computed.
Your investigation
Write a correction memo preserving the original claim, explaining the exclusion and stating what evidence would permit a new estimate.
Make a two-column evidence table: what the record directly supports, and what you infer. Use record IDs when connecting observations to claims. Preserve your first interpretation before revising it.
Practice and transfer
What is the defensible corrected claim?
Try the practice check in the activity, explain your choice, then apply the idea to this unfamiliar case: Recovered records show a comparable average of 18 minutes. What changes?
Limits and responsible interpretation
A transparent correction is evidence of revision, not automatic proof that the new claim is correct.
A successful answer identifies a mechanism and its boundaries. A saved response records practice; it does not automatically establish independent understanding.
Further reading
Two counts on one page: audit the timestamps
A fictional audit note says: “This newspaper gives different totals for the same moment, so one count must be fabricated.” Test the premise using the actual dated records. Preserve your first explanation before drafting a correction.
Inspect the primary page and selected records
Eugene Register-Guard, city edition, 3 November 1948, page 1, University of Oregon's historical newspaper archive. The large table is marked 1:30 p.m. EST; an update box beside the main story is marked 1:05 p.m. PST. Both attribute their selected counts to Associated Press. These times describe returns, not two separate publication dates.
The four candidate totals and voting-unit counts below were checked against the scan. Other news on the page is outside this numerical extract. The newspaper reports partial returns; the packet supplies neither a final certified count nor all individual unit records. Its printed aggregate equals the sum of these four listed candidates. Do not treat it as all eventual ballots or confuse it with the page's separate electoral-vote projections.
ERG-TABLE: 1:30 p.m. EST. ERG-UPDATE: 1:05 p.m. PST. Both refer to 3 November 1948. The printed standard-time labels make PST three hours behind EST; 13:05 PST becomes 16:05 EST.
Elapsed interval: 155 minutes. Comparing 1:05 with 1:30 while dropping the zones reverses the order. This conversion uses the explicitly printed standard-time labels, not a guessed modern daylight-saving rule.
| Quantity | ERG-TABLE: 13:30 EST | ERG-UPDATE: 16:05 EST | Later minus earlier |
|---|---|---|---|
| Dewey, listed popular votes | 18,770,581 | 19,508,937 | 738,356 |
| Truman, listed popular votes | 20,308,873 | 21,085,553 | 776,680 |
| Wallace, listed popular votes | 953,674 | 985,528 | 31,854 |
| Thurmond, listed popular votes | 750,718 | 757,337 | 6,619 |
| Printed aggregate of the four listed candidates | 40,783,846 | 42,337,355 | 1,553,509 |
| Voting units reporting | 111,807 | 113,309 | 1,502 |
| Stated voting-unit universe | 135,858 | 135,858 | Unchanged |
The later printed aggregate is larger by 1,553,509, and 1,502 more units are listed as reporting. These differences alone do not show that all additional votes came from those newly reporting units: earlier units could also have updated. A remaining-vote forecast is not computed.
There are two selected snapshots but one identified wire attribution and one newspaper issue. This documents a reported update sequence, not independent corroboration by two reporting organizations. State what original dispatches or unit-level records would be needed to test a claimed error.
Unfamiliar denominator case: three invented reporting units
In this separate fictional case, final ballot quantities are supplied: A has 100, B has 100 and C has 800. Every selected unit reports all of its ballots. These sizes and completeness assumptions are not known for the historical newspaper.
Unit coverage: 2/3 = 66.67%. Ballot coverage under the invented assumptions: 200/1,000 = 20%. Explain how the same unit coverage can accompany different ballot coverage, and why this does not estimate the historical missing votes.
Compare the audit and write a transparent correction
The premise of the fictional audit note is wrong: the selected snapshots are 155 minutes apart. A change between partial counts is compatible with ongoing reporting; it does not itself prove fabrication, perfect accuracy or a particular cause of every increment. Keep the original claim visible and correct only what the evidence establishes.
The changed denominator case keeps two of three units but moves ballot coverage from 20% to 90%. Units are not interchangeable with people or votes. The newspaper's source line also means these two snippets cannot automatically count as independent confirmation.
Write a private proposed correction with four parts: identify the original error, give the normalized dates/times and source IDs, explain its effect on the interpretation, and state unresolved evidence needs. Do not claim to have located a historical publisher's correction or to have certified the election from this extract.
Copy your count-verification report
Feedback opened this visit: no. Reset clears writing and restores printed clocks and A/B; exposure remains until reload. Temporary work can be copied before leaving. No correction is published and no mastery is certified.
Preserve a claim, then audit its vintage
Original classroom headline to investigate: “Net international migration more than doubled from the2023 period to the2024 period.” This is our teaching claim, not a quotation attributed to a news outlet. The authentic records below let you check what comparison could produce it.
Capture preserves the text you wrote. It does not prove you had not viewed evidence or guidance. Reset clears this temporary capture.
Period2023 means July1,2022–June30,2023; period2024 means July1,2023–June30,2024. Vintage identifies the estimate release series, not the year in which every person moved. These historical records are not current migration estimates.
Difference 1,647,130 people; relative difference 144.613337%, using the left value as denominator. Different periods and vintages: arithmetic mixes time change with revision. The number alone does not decide whether the original headline is defensible.
Open the source trail and unresolved record differences
- Census Vintage2023 CSV: national row, INTERNATIONALMIG2023.
- Census Vintage2024 CSV: national row, INTERNATIONALMIG2023 and2024.
- Census methodology explanation, December19,2024, Table1: a published comparison with slightly different numerical entries.
All were inspected on October29,2026. The CSVs carry historical vintage labels; their downloaded bytes are not proof of the exact files available on original release day. The blog explains that updated administrative inputs and methods revised earlier estimates. It cautions against mixing vintages for a trend, while permitting cross-vintage revision analysis.
| Vintage | Period | CSV | Blog Table1 | CSV minus blog |
|---|---|---|---|---|
| 2023 | 2023 | 1,138,989 | 1,135,220 | 3,769 |
| 2024 | 2023 | 2,294,299 | 2,289,938 | 4,361 |
| 2024 | 2024 | 2,786,119 | 2,785,517 | 602 |
The packet does not establish why these records differ. Do not label the gap rounding, an identified error or a specific excluded group without further evidence. Both forms support a substantial same-period revision, but exact calculations must name which form they use. They share one institutional producer; two URLs do not constitute independent corroboration. The unresolved discrepancy should remain in a transparent memo.
Revision task: compare all three modes using one source form, then transfer to the other form. Preserve your first claim and write a dated replacement stating reference interval, vintage, source form, corrected comparison and unresolved discrepancy. Distinguish your classroom headline correction from the agency's legitimate revision of estimates; this packet does not show that Census issued a false headline.
Compare your audit reasoning
CSV mixed-vintage arithmetic is about144.61%, but the consistently vintaged2024 trend is about21.44%. The same2023-period estimate was revised by about101.43%. These answer different questions. Switching to blog Table1 changes exact values, so retain its separate provenance rather than replacing numbers silently. No comparison here establishes causal explanations, individual legal status, or a statistical significance test.
Copy your revision dossier
Temporary practice: reset clears draft/capture/revision and restores CSV/mixed comparison, retaining guide/report exposure during this visit. Reload clears work and exposure. No independent verification or mastery is awarded.
Pew Research Center: writing survey questions provides disciplinary context. The teaching records above were authored for this lesson and are not quotations from that source.
Choose your depth
High school: Use the supplied records to complete this task: Write a correction memo preserving the original claim, explaining the exclusion and stating what evidence would permit a new estimate. Explain your reasoning in a short paragraph or an annotated diagram with a text equivalent.
AP connection: Use this as optional interdisciplinary evidence practice; there is no standalone AP Media and civic inquiry course. Compare two explanations using the supplied records and state a limitation.
Introductory college: Frame a researchable version of the case, document definitions and source selection, and compare the supplied mechanism with a rival explanation. Use the linked disciplinary reading to find one relevant scholarly or institutional source; distinguish that real source from these fictional records. Submit a revision log and methods note.
Worked example
The numerical comparison is not like-for-like. Without the excluded records, a corrected overall average cannot honestly be computed.
Check the reasoning, then transfer it
Loading response history…
Keep your evidence and choose a next step
Loading saved lesson status…