Design and defend an investigation
Use two mechanisms to explain how a result could mislead you, repair the investigation and defend a bounded conclusion.
For example, combine calibration with model discrimination, or group composition with sampling. Choose at least two distinct settings of one mechanism and evidence from a second. The required report fields and captures establish submission completeness, not reasoning quality.
Open the additional calibration investigation and retained review workflow
Select evidence
Select up to twelve saved captures. Capture recipes replay their results; repeated identical recipes are not independent evidence.
No captures yet. Open a lesson to collect evidence.
Draft your argument
Editing your draft keeps the previous submitted snapshot until you submit again.
Review criteria
- The question and predicted outcomes are explicit.
- The procedure can be reconstructed with units, settings and controls.
- The numerical evidence supports the stated comparison.
- Uncertainty and alternative explanations are addressed.
- The conclusion states what would reverse it and does not claim physical evidence from a simulation.
Calibrate an instrument and defend the correction
This additional inquiry investigation combines competing models, repeated raw records, a changed-range check and a human-review handoff. It preserves your existing project above and adds no new requirement to previously earned completion. The readings are original synthetic evidence, not physical measurements.
Try → predict → experiment → understand → apply → continue: define the discrimination question, lock a prediction, inspect evidence, derive a correction, test a changed range and revise your argument. A reviewer can supply a separate fully specified assignment before locking the plan. The public changed case remains supported practice.
1. Predict and lock the assignment
An exact 10 cm reference gives a mean reading of 13 cm. Two proposed responses are y = x + 3 and y = 1.1x + 2 (x and y in cm). Choose a second reference before inspecting its recorded readings. Predict its result under each model. Also plan how to test whether the chosen correction applies beyond the calibration range.
Uncommitted form text is temporary. Locking retains an immutable copy; later corrections belong in the report.
2. Inspect, calculate and test the range
The plan must be retained before this reveal. The source is public and inspectable; this is a workflow aid, not a secure test.
3. Apply the correction and write your report
Retain raw readings and calculations, a graph with units, a residual comparison and the meaning of each uncertainty interval. Explain how an excluded or missing observation would change the conclusion. Distinguish repeated readings of one instrument from independent instruments. Answer the locked assignment in the transfer field. Describe an alternative explanation and a prospective test.
Each field permits 1,200 characters. Completeness checks only presence. Retain longer plots, calculations and code externally and cite them here. A different text does not automatically constitute a scientifically substantive revision.
4. Compare retained reports
No report submitted yet.
5. Retain an actual human review
Pending review-form text is temporary until saved. Reviewer/date/help are local assertions, not authenticated certification. Every judgment needs a specific comment and, when evidence is insufficient, a targeted retry. The review binds to the selected report and its locked assignment; it cannot describe a later editable draft. If no reviewer is available, leave review pending or explicitly identify self-review.
The TXT includes raw packet, source references, locked assignment, submitted reports and bound reviews. It excludes editable drafts. Use the science JSON backup for restoration. Neither download sends work to another person.
Return to calibration · Choose discriminating evidence · Continue to the existing transfer checks
Save a science backup or import one
Science work is stored locally in this browser, separately from mathematics and AI. It does not sync automatically. Use this science-only backup; this checkout's older all-subject export does not include science.
A legacy file is available only when the complete snapshot fits the 36 MB limit. For larger snapshots, prepare a complete file set below; never discard notebooks to make a smaller file.
Matching imported notebook drafts replace their current drafts. Omitted lessons and known first attempts remain. Export your current work first if you want to retain both drafts.
Complete file sets and course backups
A file set can hold more work than one science file. Keep its manifest and every part together. Each file is limited to 36 MB. This format needs this version of LearnAI or a compatible newer version; older versions can use only the legacy single-file backup above.
The set includes every saved science course and global continuation at this moment. Later edits need a new export. A conflicting tab exports its own visible work, which may differ from saved work in another tab.
0 file(s) selected. File order and filenames do not determine identity. Missing, changed or conflicting parts block the entire restore; no good part is applied by itself.
Science storage: loading. No account or remote computation is required.