This packet is everything a parent or guide needs to assess Unit 06 at home — learning targets, the answers that count as correct, the mastery rubric, calibration examples, and a clipboard score sheet. No multiple-choice test: the student shows mastery by repeating a measurement, reporting its spread honestly, and telling random error from systematic error.
By the end of the Uncertainty, Error & Honesty unit, a student should be able to:
Six criteria, each judged Not yet / Approaching / Mastered (Page 3).
Time or measure the same thing five times; report the spread.
The student names random vs systematic error in their own setup (Page 4).
Every trial, the spread, and any outlier kept distinct.
You are making a decision, not adding up points. For each criterion, decide whether the work is Not yet, Approaching, or Mastered — the column language tells you which. A criterion counts as mastered only when the student can both report the spread honestly and tell random error from systematic error. A student carries three tokens per term; one token buys a re-do of one criterion on another day, so a single bad afternoon never sinks the unit.
Accept any answer in the synonyms column — they are pre-approved as equivalent. The third column flags the confusions that look close but are not yet, so you can coach precisely.
| Canonical answer | Accepted synonyms | Common confusion / discriminator |
|---|---|---|
| Error | ||
| Random error | scatter / noise | Scatters results both ways — a wobbly stopwatch thumb |
| Systematic error | bias / offset | Pushes every result one way — a ruler that starts at 1 cm |
| Uncertainty | give-or-take / margin | The doubt attached to a value, not the same as a mistake |
| Outlier | anomaly / odd point | A point far from the rest — flag it, don’t delete it |
| Honesty & spread | ||
| Spread (range) | low-to-high | “40 to 46 seconds” — how much the trials differ |
| Repeat trial | re-run / replicate | Doing it again to see how much the result varies |
| Intellectual honesty | reporting what really happened | Never changing a number to make it “come out right” |
| Disproven hypothesis | a real answer, not a failure | A guess shown wrong is a finding, not a broken experiment |
| Criterion | Not yet | Approaching | Mastered |
|---|---|---|---|
| Sources of error | Blames “human error” for everything, or cannot say where a measurement might go wrong. | Names a source of error but cannot tell whether it scatters results randomly or pushes them all one way. | Distinguishes random error (a wobbly stopwatch thumb) from systematic error (a ruler that starts at 1 cm), and names each. |
| Repeat trials & spread | Measures once and treats that single number as the truth. | Repeats the trial but ignores how much the results differ from each other. | Runs several trials and reports the spread: “the fizz tablet dissolved in 40 to 46 seconds across five tries.” |
| Reporting uncertainty | Reports a lone number as if it were exact. | Admits the measurement is not perfect but cannot attach a range to it. | States the measurement with its doubt: “about 12.5 cm, give or take half a centimeter, from how hard the ruler was to read.” |
| Anomalies & outliers | Quietly erases a point that does not fit, or never notices it. | Spots an odd point but drops it with no reason, or keeps it without a second look. | Flags the outlier honestly, keeps it in the record, and says what might have caused it — never deletes a “bad” point silently. |
| Intellectual honesty | Changes numbers or the hypothesis to make the results “come out right.” | Reports the real result but calls the experiment a failure because it disproved the guess. | Reports a result that disproves the hypothesis as a real finding: “the paper airplane did not fly farther — that is a true answer, not a mistake.” |
| Integration (cross-domain) | Treats honesty-with-data as isolated; makes no connection to the year’s anchor. | Mentions Semmelweis but cannot say what honesty had to do with his data. | Connects honest reporting to Semmelweis — who let uncomfortable death-rate data stand even when colleagues resented it — across History · Reading · Statistics, and defends why it mattered. |
The split between Approaching and Mastered is honest about the spread: one number is a guess, a range is a measurement. Ask “how much did it vary, and did you keep the odd point?” before accepting a result.
Read these before you grade. They show what Mastered and Not yet actually sound like, plus the edge cases where you should coach rather than decide on the spot.
Student: ______________________________________ Date: _______________ Guide: _________________________
| # | Criterion | Decision | Notes |
|---|---|---|---|
| 1 | Sources of error | NY / Appr / Mast | |
| 2 | Repeat trials & spread | NY / Appr / Mast | |
| 3 | Reporting uncertainty | NY / Appr / Mast | |
| 4 | Anomalies & outliers | NY / Appr / Mast | |
| 5 | Intellectual honesty | NY / Appr / Mast | |
| 6 | Integration (cross-domain) | NY / Appr / Mast |
☐ No ☐ Yes — for criterion: __________ Tokens remaining: ☐ 3 ☐ 2 ☐ 1 ☐ 0
NY = Not yet · Appr = Approaching · Mast = Mastered · Unsure between two levels? Circle the lower one and note what a re-do would need.