Generated 2026-11-02 · State: KS · Standards scope: Grades 6–8 mathematics + Grade 6 English Language Arts · Rebuilt from the audited crosswalk data on each revision.
Grades 6–8 went live in AcornSmarts on 15 September 2026. The other bands for Kansas are published separately: K–5 · grades 9–12 · all reports.
| Code | Verdict | Evidence (generator · forms) | Notes |
|---|---|---|---|
| 6.RP.1 | PARTIAL two-auditor sweep 03 Oct 2026 |
ratiosSaplingQ L5 kind 0 MC: 'In a group of N, there are a X and b Y. What is the ratio of X to Y?' Answer a:b; distractors a:total (part-to-whole) and b:a (order flip); why names part-to-part vs part-to-whole explicitly. ▶ Play a live item from this skill |
Distinguishing part-to-part from part-to-whole is exercised at L5 (the a:total distractor forces it). But at L5-7 the child selects notation; no item there has them produce or select ratio LANGUAGE ('for every 2 wings there was 1 beak'-style sentences describing a ratio relationship) — the L7 recipe item prints 'for every' in its prompt only. 'For every' appears in no answer choice. The nearest are unit rates on a graph, not ratios between two counted groups: Ratios L8 keys one ticket's price (for example 'buying 1 ticket costs $2', beside '$1 buys 2 tickets'), and Functions L5 (a Grade 8 skill) keys the unit rate (for example 'the unit rate: y grows by 3 when x grows by 1'). In these graph items 'for every' appears only in an explanation shown after the answer ('the line climbs 3 for every step of 1'). No ratio item has the child produce a sentence. Describe is a production verb under the house standard, so PARTIAL. 26 Sep 2026 (round 2): with Writing Coach on — a paid plan or a live school sponsorship, district switch on — the child writes two ratio-language sentences (part-to-part and part-to-whole) for one class, graded against a rubric requiring both; one fixed prompt, live rounds only. Without Writing Coach this row is served by selection only, so the published verdict stays PARTIAL: no row whose clause is met only through Writing Coach is marked FULL. Blocking clause: Use ratio language to describe a relationship between two quantities |
| 6.RP.2 | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
ratiosSaplingQ L6 kinds 0-2 MC: price per notebook ('per' = divide, halves allowed, $1.50-style); train unit rate answered in 'miles per hour' language; better-buy with per-ounce comparison: the options name only the packages ('package A', 'package B') beside 'they cost the same per ounce', so the child works out both unit prices; the two cost the same per ounce in about a third of deals. ▶ Play a live item from this skill |
Unit rates are found and compared at Ratios L6: '5 notebooks cost $12.50. What is the price per notebook?', 'A train travels 120 miles in 2 hours at a steady speed. What is its unit rate, in miles per hour?' (typed on a number pad in about six train items in ten), and a better-buy item ('Which is the better buy? A: 12 oz for $3.00. B: 16 oz for $4.80.') whose options name only the packages ('package A', 'package B') beside 'they cost the same per ounce'; the per-ounce prices appear only in the explanation, so the child divides to compare. Either package can be the better buy (the bigger one wins about half the time), and in about a third of deals the two packages cost the same per ounce, which is then the answer. Since 20 Sep 2026 the same level also deals at least two complex-fraction rates per round, typed in a/b notation with the / key (a Grade-7 family). The child reads and picks unit-rate language but never uses it. The only exception is Writing Coach's explain prompt ('A store sells 3 notebooks for $6. Explain what the unit rate is here and how you found it.'), which is not counted. No other item has the child say or write 'per', 'for each' or 'for every'. PARTIAL on the language clause. Blocking clause: Use unit rate language ("for each one", "for every one" and "per") |
| 6.RP.3 | FULL two-auditor sweep 03 Oct 2026 |
ratiosSaplingQ L5-L7 Solving real problems by table reasoning (L5 kind 1 missing-value table), calculation (L5 kind 2 equivalent ratios, L7 kind 2 recipe scaling), and unit-rate calculation (L6 all kinds). ▶ Play a live item: level 5 · level 6 · level 7 |
The verb is 'solve' and solving happens across three levels (L5-L7). The listed representations are an 'e.g.' list; at these levels only tables and calculations appear. Equations for a rate relationship are served elsewhere: Equations L6 selects d = 5t from a table and uses the table's relationship to find d for a new t, and Ratios L8 selects t = 8n for a priced-per-item story. No visual in the app is a tape diagram or a double number line, and neither is named in any item. Not blocking (illustrative list). |
| 6.RP.3a | PARTIAL two-auditor sweep 03 Oct 2026 |
ratiosSaplingQ L5 kind 1; L6 kinds 0-2 Find missing value in a stated equivalent-ratio table (lemons→oz); unit-pricing problems; constant-speed unit rate; better-buy comparison of unit rates. ▶ Play a live item from this skill |
Find-the-missing-value (Ratios L5: 'A table of equivalent ratios shows: 2 lemons → 6 oz, 3 lemons → ? oz, 4 lemons → 12 oz'), unit pricing ('5 notebooks cost $12.50. What is the price per notebook?') and constant speed ('A train travels 120 miles in 2 hours...', typed on a number pad in about six train items in ten) are practised. Plotting is practised too. In live rounds about one Ratios L5 question in six prints a table of equivalent ratios ('lemons → ounces: 1 → 3, 2 → 6, 3 → 9'). The child taps the three pairs onto the grid and then picks, from three options, what the pattern shows. The child copies pairs from a printed table. When the rate is 4, the third pair (3, 12) falls off the 0-9 grid and the item cannot be completed (about a third of these items). The child never makes a table of equivalent ratios. No item uses tables to compare two ratios: the better-buy item compares two unit prices, and either package can be the cheaper one. Blocking clause: Make tables of equivalent ratios ... and plot the pairs of values on the coordinate plane |
| 6.RP.3b | FULL two-auditor sweep 03 Oct 2026 |
percentsSaplingQ L5-L6 L5: p% of n with p a multiple of 5 (incl. 5,15,...,95) and n a multiple of 20, one-place-shift distractor. L6 branch 1: 'X is p% of a number, what is the number?' (find the whole given part+percent). L6 branch 2: percent from part/whole. ▶ Play a live item: level 5 · level 6 |
L5: 'What is 35% of 120?', with p from 5, 15, 25, 35, 45, 55, 65, 85, 95 and n a multiple of 20 (20-180), whole answers. L6: '24 is 40% of a number. What is the number?' (find the whole) and 'In a class of 40, 30 students carry a backpack. What percent is that?'. A percent of a quantity is also the first step of the L8 tax and tip totals. Weakness: in about five of six L5 items the options include the key and ten times the key (the p/10 slip), so 'pick the smaller of that pair' finds the answer without the rate-per-100 work. The L6 find-the-whole items have no such pattern. |
| 6.RP.3c | FULL two-auditor sweep 03 Oct 2026 |
ratiosSaplingQ L7 kinds 0-1 Convert inches→feet (÷12, incl. half-feet like 90 in) and hours→minutes (×60, incl. the .5h→50min trap); recipe-scaling kind 2 is ratio reasoning with quantities. ▶ Play a live item from this skill |
Ratios L7 converts inches to feet (÷12, including half-feet such as 78 in = 6.5 ft) and hours to minutes (×60, including 2.5 h = 150 min and its '50 minutes' trap). Recipe scaling at the same level is ratio reasoning with quantities. Both directions (multiply and divide) appear, and the why checks that going from a small unit to a big one divides. Weaknesses: at this level only those two conversion pairs appear; metric, capacity and weight conversions live in the Grade 4-5 Unit Conversions skill, which is not credited here. In half-foot items the keyed answer is the only option ending in .5. Units carried through division appear at Ratios L6, where a rate is found in the unit the question names (miles per hour, acres per bag). |
| 6.NS.1 | PARTIAL two-auditor sweep 03 Oct 2026 |
fractionsSaplingQ L6-L7 L6: whole ÷ unit fraction (servings), unit fraction ÷ whole (share brownies), same-denominator counting a/d ÷ 1/d. L7: 1/b ÷ 1/(bk) with whole answer; a/b cut into 1/d-lb pieces word problem; structure check 'which expression equals a/b ÷ c/d' (invert-and-multiply). ▶ Play a live item: level 6 · level 7 |
Fraction ÷ fraction is interpreted and computed. Fractions L6 gives it meaning: how many 1/d-cup servings fit in w cups, 1/d of a pan shared among w friends, and how many times 1/d fits into a/d. Fractions L7 computes 1/b ÷ 1/(bk) and a/b ÷ 1/d (in a cut-the-chocolate word problem), and checks the invert-and-multiply structure of a/b ÷ c/d. These are all multiple choice, and every quotient computed in this skill has a unit-fraction or whole-number divisor; a non-unit divisor appears only in the structure check. (With the Writing Coach entitlement, a round may also deal one written explanation of a fraction quotient, judged by the writing grader.) General a/b ÷ c/d word problems are computed and typed at ratios L6, a level a Grade-6 child reaches, in at least two items of every six-item round: unit rates such as 3/4 of a mile in 2/3 of an hour. There the divisor is a non-unit fraction in about two draws in three, and the quotient is not a whole number in more than nine in ten. The row is held PARTIAL because no item at these levels shows a concrete or visual model. Every item is text, so the standard's 'connecting various concrete and abstract models' is not practised. Blocking clause: Interpret and compute quotients of fractions |
| 6.NS.2 | FULL two-auditor sweep 03 Oct 2026 |
divisionSaplingQ L6-L7 L6: bare multi-digit division nd ÷ dv with 2-digit divisors (12-29), dividends to ~4 digits, off-by-one distractors, inverse-multiplication check taught. L7: packing word problem, dividends to ~3500, estimation strategy taught. ▶ Play a live item: level 6 · level 7 |
The verb (divide) is exercised on multi-digit numbers with exact quotients. At L6 it is a bare 3- or 4-digit dividend (132 to 1,218) divided by a 2-digit divisor (12–29); at L7 it is a packing word problem with dividends up to about 3,500. About two items in five are answered by typing the quotient on a number pad; the rest are multiple choice. Adversary angle: the multiple-choice L6 items offer near-miss quotients, and checking which choice times the divisor ends in the dividend's last digit finds the key in most of them without dividing. Every division is exact (no remainders), so fluency is inferred from correct answers, not observed. |
| 6.NS.3 | PARTIAL two-auditor sweep 03 Oct 2026 |
decimalsSaplingQ L6-L7 L6: add and subtract money-like decimals (1.05-8.50, hundredths), place-alignment distractors. L7: multiply 0.x × 0.y (single tenths digits, place-count why); divide x.y ÷ 0.d (single-digit, multiply-both-by-10 why). ▶ Play a live item: level 6 · level 7 |
Addition and subtraction use multi-digit decimals (hundredths, 1.05 to 8.49), but the choices give the key away: the wrong choices are 0.10 and 1.00 above the key, or one 0.10 below and one 0.10 above it, so the key is the smallest choice when one choice sits about 1 away and the middle choice otherwise, in every deal. Multiplication is only ever tenth × tenth with one digit each (0.2 to 0.8). Division is only a tenths dividend of at most two digits (0.6 to 8.8) divided by a one-digit tenth, with a whole-number quotient. All items are multiple choice. Two of the four required operations sit well below the standard's multi-digit size, so the fluency claim cannot be FULL. Blocking clause: multiply, and divide multi-digit decimals |
| 6.NS.4 | PARTIAL two-auditor sweep 03 Oct 2026 |
factorsSaplingQ L5-L6 L5: GCF of two numbers, coprime-multiple construction, pool filtered to keep both ≤100. L6 branch 1: LCM of pairs all ≤12, with a·b-is-not-always-least distractor. L6 branch 2: rewrite a+b as g(m1+m2), distractors g(a+b) and a wrong-factor form. ▶ Play a live item: level 5 · level 6 |
Factors Level 5 asks for the GCF of two numbers up to 84 (typed on a number pad about two times in five in English). Level 6 asks for the LCM of a pair up to 12, and asks the child to pick the distributive rewrite of a sum, such as 18 + 24 = 6(3 + 4), from three choices; the rewrite is never written. In Spanish the GCF and LCM are never typed, and every LCM item puts the right answer between the two wrong ones (the GCF below it, a × b above), so it can be picked without finding the LCM. PARTIAL. Blocking clause: Use the distributive property to express a sum of two whole numbers 1–100 with a common factor as a multiple of a sum |
| 6.NS.5 | FULL two-auditor sweep 03 Oct 2026 |
integersQ L1 + L4 Below-zero temperature → −n; number-line side of −n; drone above / diver below sea level with 0 as reference; earned/spent as +/−. ▶ Play a live item: level 1 · level 4 |
'Understand ... quantities having opposite directions or values' assessed by recognition across temperature, elevation, and money — three of the standard's four example contexts (electric charge absent, but the list is an e.g.). |
| 6.NS.5a | PARTIAL two-auditor sweep 03 Oct 2026 |
integersQ L1 form 1; L4 form 0 Select the number representing 'n degrees BELOW zero' (−n vs n vs −(n+10)) and the number representing a diver 'down' m below sea level (−down vs down vs −up, the drone/diver mixup distractor). ▶ Play a live item from this skill |
The real-world signing is well targeted, in English and Spanish: the child picks the number for a temperature n degrees below zero, for a diver below sea level (in about two deals of three the drone's height is one of the wrong choices), and what the signs mean for money earned and spent. But 'represent' is a production verb, and here the child only picks the signed number from three choices; no item asks the child to write or type it. Contexts are limited to temperature, sea level and money. Blocking clause: Use positive and negative numbers to represent quantities in real-world contexts — the child only picks the signed number from three choices |
| 6.NS.5b | PARTIAL two-auditor sweep 03 Oct 2026 |
integersQ L4 form 1 'Writing earnings as +E and spending as −S, what does 0 represent here?' — answer 'neither earning nor spending — the starting point'; distractors 'running out of money' (balance-zero confusion) and the difference. The diver item's why also states 'Zero is sea level — the reference point'. ▶ Play a live item from this skill |
The zero-as-reference-point idea is directly targeted, including the balance-of-zero trap: in the earnings-and-spending story the child picks what 0 means ('neither earning nor spending — the starting point') or what the signs mean. But the child only picks among written sentences, and only for money — the diver item asks for the number, not for what 0 means. A written explanation ('Explain what 0 means in this situation and what the signs tell you') is dealt only through the metered Writing Coach grader, which this report does not count. Blocking clause: Explaining the meaning of 0 in each situation |
| 6.NS.6 | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
integersQ L1; coordinateSaplingQ L4 Locate −n as 'n units LEFT of zero' on the number line; identify the quadrant of (x, y) from the signs. ▶ Play a live item from this skill |
Number-as-point is assessed by recognition: the child picks where −n sits ('n units LEFT of zero', in words — no line is drawn), and since the number asked about is always negative, the right answer is always the LEFT one. Pair-as-location is assessed by naming the quadrant of a signed pair from its signs (coordinate-grid L4) and by plotting a rectangle's four vertices on a first-quadrant grid (coordinate-grid L5). Every number located at the Grade 6-8 levels is an integer; fractions are placed only in below-band levels (the 0-to-1 line in the Grade 3-5 Fractions levels, and marks on line plots in Graphs & Data), and no decimal or negative fraction is placed anywhere, so 'a rational number' is met only through its integer cases. Blocking clause: Understand a rational number as a point on the number line — only integers are ever located at the Grade 6-8 levels, and only in words |
| 6.NS.6a | PARTIAL two-auditor sweep 03 Oct 2026 |
integersQ L1 form 0 'What is the opposite of v?' where v may itself be negative — so opposite-of-−n = n is computed; the number-line item's why states 'Opposite signs mean opposite sides of zero'. ▶ Play a live item from this skill |
Integers L1 asks for the opposite of a positive or a negative integer ('What is the opposite of -12?'), with a rationale that places it on the other side of zero at the same distance, and asks where −n sits on the number line ('n units LEFT of zero'). Asking the opposite of a negative implicitly touches −(−n) = n. But the opposite of an opposite is never posed in −(−3) notation and is never asked as its own question. The only questions that print a '− (−n)' string are the Grade 7 subtraction items at integers L5 ('7 − (−4) = ?'), which subtract a negative rather than take a double opposite. No item asks for the opposite of 0 or states that 0 is its own opposite (in the opposite item the number is never 0: n ≥ 2). Blocking clause: recognize that the opposite of the opposite of a number is the number itself, (e.g. −(−3) = 3,) and that 0 is its own opposite |
| 6.NS.6b | PARTIAL two-auditor sweep 03 Oct 2026 |
coordinateSaplingQ L4 Quadrant of (x, y) from signs (all four quadrants reachable, nz() draws both signs); reflection of (x, y) across the x-axis with (−x, y) and (−x, −y) as the one-axis/both-axes distractors, why contrasts flipping x vs y vs both. ▶ Play a live item from this skill |
Coordinate-grid L4 asks which quadrant a point is in; both signs are drawn at random, so points land in all four quadrants. The two wrong choices are two of the other three quadrants, picked at random each time, so every set of three quadrants is offered and no quadrant is offered only when it is the answer (until 24 Oct 2026 the wrongs were always the first two other quadrants, so IV was offered only when it was the answer). L4 also asks for the reflection of a point across the x-axis, but the point is always in quadrant I and the axis is always x, so the answer is always the only choice with a positive first coordinate. Coordinate-grid L6 asks which transformation carries (x, y) to (−x, y) or to (x, −y), with the x-axis and the y-axis each the answer about half the time - about one question in four at that level, in English and Spanish. No Grade 6-8 item asks about two points that differ in both signs (a reflection across both axes); at these levels that case appears only as a wrong choice and in an explanation. Blocking clause: recognize that when two ordered pairs differ only by signs, the locations of the points are related by reflections across one or both axes (no item ever asks about the both-axes case; the L4 reflection item gives its answer away by form) |
| 6.NS.6c | PARTIAL two-auditor sweep 03 Oct 2026 |
integersQ L1 form 2; coordinateSaplingQ L4 'On the number line, where is −n?' answered in words ('n units LEFT of zero'); quadrant identification for integer pairs. ▶ Play a live item from this skill |
The child never positions a value on a number line in the way the standard asks. Integers L1 describes a location in words ('4 units LEFT of zero'). Two number lines have tappable numbers: the Grade 7 frog hop at integers L5, where the line runs into the negatives and the child taps where a sum lands, and the inequality graph at Inequalities L3-4, where the child taps the boundary of x > c — always a positive whole number, though the line runs to −1 or −2 when the boundary is 2 or 3. On both, every number is printed on its own button, so the child finds a labelled number rather than positioning one. On the plane, coordinate-grid L4 classifies points by quadrant and reflects them with no diagram; coordinate-grid L5 has the child plot a rectangle's four vertices, but that grid is first-quadrant only (0 to 9), so no negative coordinate is positioned. Fractions are placed only as positive values in below-band levels not credited here (the 0-to-1 line in the Grade 3-5 Fractions levels; line plots in Graphs & Data). Two gaps remain: no negative value is ever positioned by the child on a line or plane, and no non-integer rational is positioned in band. Blocking clause: Find and position integers and other rational numbers on a horizontal or vertical number line diagram |
| 6.NS.7 | PARTIAL two-auditor sweep 03 Oct 2026 |
integersQ L2 + L3 Order two negatives ('which is LESS', number-line why); debt comparison −$a vs −$b; |v| computation; the order-vs-magnitude combined statement item. ▶ Play a live item: level 2 · level 3 |
Ordering is practised with integers at integers L2: the child picks which of two negatives is less and types a true comparison such as −13 < −3; the who-has-less-money item with two negative balances is not counted, because its right answer is always Beto. Absolute value is computed at L3 ('|−16| = ?'), always of an integer. No negative fraction or decimal is ever ordered or given an absolute value, so for negative numbers 'rational numbers' is met only through integers; non-integers are compared only as positive values (Decimals L4-5, e.g. 'Compare: 0.8 ? 0.73 — pick >, =, or <.'). The L3 order-versus-magnitude item is not counted: its right answer is always the only choice joined by 'but' ('pero'), so it can be picked without reading the numbers. Blocking clause: ordering and absolute value of rational numbers — at the Grade 6-8 levels only integers are ordered or taken absolute value of |
| 6.NS.7a | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
integersQ L2 form 0; L3 form 1 'Which is LESS: −a or −b?' with the why locating both on the number line ('−lo sits farther left'); 'Which statement is true about −a and −b?' where the correct option is itself an inequality chain −a < −b. ▶ Play a live item from this skill |
The comparison items (which of two negatives is less; type a true comparison of two negatives) carry a number-line reading in their feedback ('−9 sits farther left than −7'), but they can also be answered with the sign rule alone, and no number line is drawn. No item hands the child a written statement like −3 > −7 and asks what it says about position, so the interpret-as-position clause is met only in feedback. Blocking clause: Interpret statements of inequality as statements about the relative position of two numbers on a number line diagram |
| 6.NS.7b | PARTIAL two-auditor sweep 03 Oct 2026 |
integersQ L2 form 1 'Ana's account reads −$d and Beto's reads −$d2. Who has LESS money?' — real-world order interpretation with the owing-more-is-having-less why. ▶ Play a live item from this skill |
Order is interpreted in a real-world context by one item (who has less money with two negative balances), but its right answer is always Beto — the name with the bigger debt — so it can be picked without comparing the numbers. The child does type an order statement at integers L2 ('Write a TRUE comparison of −3 and −13 using < or >', about one L2 question in five), but always for two bare negative integers, never for a real-world quantity like −3°C > −7°C. A written explanation ('Explain why −3 is less than −1…') is asked only through the metered Writing Coach grader, which this report does not count. Blocking clause: Write, interpret, and explain statements of order for rational numbers in real-world contexts |
| 6.NS.7c | PARTIAL two-auditor sweep 03 Oct 2026 |
integersQ L3 forms 0-1 |v| = ? computed with the distance-to-zero why; debt magnitude appears in the L3 form-1 why ('A $a debt is less money AND more debt than a $b one'). ▶ Play a live item from this skill |
Computing |v| is asked ('|−16| = ?'), with feedback that frames it as distance to zero. No question puts absolute value in a real-world situation — the debt reading appears only in the feedback of the order-versus-magnitude item — and the child never writes |−30| = 30 for a debt. A written explanation ('Explain in your own words what |−7| means … Give a real-life example') is dealt only through the metered Writing Coach grader, which this report does not count. Blocking clause: Explain the absolute value of a rational number as its distance from 0 on the number line |
| 6.NS.7d | FULL two-auditor sweep 03 Oct 2026 |
integersQ L3 form 1 'Which statement is true about −a and −b?' — correct: '−a < −b, but |−a| > |−b|'; distractors assert the wrong order or the wrong magnitude; why: a $a debt is less money AND more debt. ▶ Play a live item from this skill |
This is the standard's own example (account balance less than −30 = debt greater than 30) rebuilt as a distinguish-by-recognition item. Distinguish is a compatible verb. Clean FULL. |
| 6.NS.8 | PARTIAL two-auditor sweep 03 Oct 2026 |
coordinateSaplingQ L5 Distance between (x, y1) and (x, y2) — same x, opposite-sign y's so the segment crosses the axis; why teaches |y1 − y2| as adding both distances to zero. ▶ Play a live item from this skill |
Distance is served in two pieces at coordinate-grid L5: the bare two-point item is same-first-coordinate only (x is fixed; the y's differ and always straddle the x-axis), and the same-second-coordinate (horizontal) case is served only as the follow-up to the plot-a-rectangle item, in the first quadrant. Graphing exists - half of coordinate-grid L5 questions have the child plot a rectangle's four vertices on a tap grid - but that grid is first-quadrant only, so the child never plots a point with a negative coordinate on it. Absolute-value notation appears only in the explanations (|y1 − y2| for the vertical pair, |x2 − x1| for the rectangle side); neither prompt mentions absolute value. Neither item carries a real-world context. Blocking clause: by graphing points in all four quadrants of the coordinate plane |
| 6.EE.1 | PARTIAL adversarial: NOTE-ONLY · two-auditor sweep 03 Oct 2026 |
expressionsQ L2 branch 0 b^e evaluated (b 2-6, results kept small), with b×e and b+e as the counts-factors-not-addends distractors. ▶ Play a live item from this skill |
Evaluating exponent expressions is practised: b^e with b from 2 to 5 and e from 2 to 4, about half of Level 2, in English and Spanish. It is multiple choice; b × e and b + e are the planted wrong answers, though in about half the items one of them is swapped for a nearby number. 'Write' is a production verb with nothing behind it: no item has the child write an exponential expression. Choosing one is asked only outside Grade 6, always picked from three: 'How do you write 1,000 as a power of 10?' in Place Value, and the Grade 8 scientific-notation and product-of-powers items. Blocking clause: Write and evaluate numerical expressions involving whole-number exponents |
| 6.EE.2 | PARTIAL two-auditor sweep 03 Oct 2026 |
expressionsQ L2 branch 1; L3 Evaluate ax + b at x (with the digit-gluing '3(5)=35' distractor); select the expression for a verbal phrase and for a real-world cost. ▶ Play a live item from this skill |
Write, read and evaluate are each practised. The child types expressions such as 2n − 5 and 4b + 3 from words (about half of Level 3, judged by equivalence). The child reads expressions to match them to word phrases and names the coefficient (multiple choice). The child types the value of ax + b at a given x (about a quarter of Level 2). This standard is graded with its parts. 6.EE.2a is FULL, but 6.EE.2b is PARTIAL (only 'coefficient' is asked outside Writing Coach) and 6.EE.2c is PARTIAL (no Grade 6 item puts an exponent inside an order-of-operations expression), so the umbrella stays PARTIAL. Blocking clause: Write, read, and evaluate expressions in which letters stand for numbers |
| 6.EE.2a | FULL two-auditor sweep 03 Oct 2026 |
expressionsQ L3 'Which expression says "k less than twice a number"?' (2n − k vs k − 2n vs 2(n − k)); ticket-cost-plus-flat-shipping expression selection (per·b + k vs the shipping-per-ticket error). ▶ Play a live item from this skill |
26 Sep 2026 (round 1, two auditors agreed FULL): expressionsQ L3 types 2n − k and per·b + k from words; the polynomial judge refuses k − 2n and 2(n − k) and accepts commuted forms; ~half of live L3 deals, EN+ES; typedKeyTruth re-derives every key from its prompt. |
| 6.EE.2b | PARTIAL two-auditor sweep 03 Oct 2026 |
expressionsQ L1 branch 1 'In the expression cx + k, what is the COEFFICIENT?' — c vs k vs c+k; why names the constant term. ▶ Play a live item from this skill |
26 Sep 2026 (round 2): with Writing Coach on, the child describes 4(x + 7) in writing as a product of two factors, one of which, (x + 7), is a sum — viewing the sum as one entity; without it only 'coefficient' is asked (multiple choice). Still PARTIAL: quotient is never asked, and no row whose clause is met only through Writing Coach is marked FULL. Blocking clause: view one or more parts of an expression as a single entity |
| 6.EE.2c | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
expressionsQ L2 branch 1 + volumeSaplingQ L4-L5 Evaluate ax + b at a value (multiply-before-add taught, digit-glue and wrong-grouping distractors); evaluate the real-world formulas V = lwh and V = Bh at given values including fractional (x.5) heights. ▶ Play a live item from this skill |
Practised: evaluation at specific values (ax + b at a given x, typed on the number pad in about a quarter of Level 2), formulas from real-world problems (V = l × w × h and V = Bh at half-unit heights, in the volume levels), and multiplication before addition. Not practised at Grade 6: an evaluated expression that mixes a whole-number exponent with other operations. In the Grade 6 items the exponent only ever appears alone (b^e = ?). Exponents meet other operations only in Grade 8 items: scientific notation such as '6 × 10^4 = ?', and the Pythagorean items, where the child squares and adds to find a side or a distance (picked from three) or to answer the converse yes or no. Blocking clause: Perform arithmetic operations, including those involving whole-number exponents, in the conventional order when there are no parentheses |
| 6.EE.3 | FULL two-auditor sweep 03 Oct 2026 |
expressionsQ L4 (typed exprwrite:combine / exprwrite:dist) Type ax + bx with like terms combined (one term; the given form refused) and a(x + b) without parentheses (brackets refused). ▶ Play a live item from this skill |
26 Sep 2026 (round 2): the child types equivalent expressions in about half of live Level 4 items, in English and Spanish. One kind is a(x + b) with the brackets gone (any bracket refused). The other is ax + bx as one term (a second term refused). The given form is refused in both. Two leniencies remain: a bracketed sum such as (3 + 4)x is still accepted as 'combined', and the expanded form need not be simplified (5x + 5 × 5, or x + x + x + x + x + 25, is accepted for 5(x + 5)). |
| 6.EE.4 | PARTIAL two-auditor sweep 03 Oct 2026 |
equationsSaplingQ L6 branch 0; inequalitiesQ L2 'Which value from the set {x−1, x, x+2} makes ax + b = rhs true?' with the why stating substitution IS the method; 'Which value from the set {...} makes x + a > c true?' with the boundary value excluded by strict inequality. ▶ Play a live item from this skill |
Equations Level 6 asks which value from a three-number set makes an equation such as 3x + 4 = 19 true, about three Level 6 questions in ten, in English and Spanish. The right value's place in the set varies, and the explanation substitutes the right value and says testing each value is the method. Inequalities Level 2 asks the same for x + a > c and x + a < c, but there the right value is always the largest number in the set for > and the smallest for <, so it can be found without substituting. Every set has exactly one value that works, so the idea that several values, or none, may make it true is never practised. PARTIAL. Blocking clause: which values from a specified set, if any, make the equation or inequality true |
| 6.EE.5 | PARTIAL two-auditor sweep 03 Oct 2026 |
expressionsQ L3 branch 1; equationsSaplingQ L5 Real-world expression selection (cost of b tickets = per·b + k); one-step equations solved for the unknown. ▶ Play a live item from this skill |
Variables-as-unknowns are used throughout the equations levels, and the any-number-in-a-set reading is carried by the 6.EE.4 items. But 'write expressions' is production — building the expression is always selection — and no item assesses the variable-can-represent-any-number idea as its own target. Blocking clause: write expressions when solving a real-world or mathematical problem |
| 6.EE.6 | PARTIAL two-auditor sweep 03 Oct 2026 |
equationsSaplingQ L5 One-step equations in all four operations: x + a, x − a, ax, x ÷ a — each solved by the inverse operation, with the same-operation-again distractors and balance language in the whys. ▶ Play a live item from this skill |
At Equations Level 5 the child types one-step equations from short stories, using whole numbers and the story's own numbers: Ivy had s stickers, gave 7 away and has 12 left (s − 7 = 12), and 4 equal boxes hold 72 crayons (4c = 72). That is about one Level 5 question in four, and the child also types one-step solutions for the same two shapes. Addition and decimal equations are written only by choosing from three (Sam earns or spends money; juice boxes at a decimal price), and no item asks the child to write a division equation. Decimal and unit-fraction one-steps (x + 1.5 = 7.3, (1/3)x = 4) are solved by choosing. A typed one-step addition equation (m + a = 180 or 90) appears only in Shapes & Geometry Level 9 angle items. PARTIAL. Blocking clause: Write and solve one-step equations involving non-negative rational numbers |
| 6.EE.7 | FULL two-auditor sweep 03 Oct 2026 |
inequalitiesQ L1 + L3 Select h > c for a more-than-c height rule and d < m for less-than-$m money (real-world constraints); 'what does the solution set of x > c look like?' asked for x > c, x < c, x ≥ c or x ≤ c and answered as an open or a filled circle at c, shaded right or left, going on forever, beside two of the other three circle-and-direction pictures (each named in the explanation); 'how many solutions does x > c have?' → 'infinitely many — any value above c is fine', beside 'exactly one — just the integer after c does' and 'none — x cannot possibly get greater than c' (one length and one word count, each wrong answer named). ▶ Play a live item: level 1 · level 3 |
26 Sep 2026 (round 1, two auditors agreed FULL): Write typed at L1 (~half of deals), 'infinitely many' asked directly at L3, and RayPick has the child BUILD the boundary, open/closed circle and direction (about a quarter of L3 — 27% measured 03 Oct 2026); checkAlgebra refuses h ≥ 40 against h > 40 on direction. |
| 6.EE.8 | PARTIAL two-auditor sweep 03 Oct 2026 |
equationsSaplingQ L6 branch 1 Real-world two-quantity contexts (cost/pounds at $2 per lb, distance/hours at 60 mph, earnings/lawns at $15 each) asking which is the DEPENDENT variable. ▶ Play a live item from this skill |
At Equations Level 6 the dependent-variable items describe a rate in words (apples at $2 a pound, a car at 60 mph, $15 a lawn) and ask which quantity is the dependent variable; no letters are used. Letters appear in the table and graph items: the child types d = kt from a table of distance d against hours t (about one Level 6 question in eleven), finds d for a time beyond the table, or picks the equation for a shown graph of three or four points. The letters d and t and what they stand for are always given, and distance and hours is the only relationship. PARTIAL. Blocking clause: Use variables to represent two quantities in a real-world problem that change in relationship to one another |
| 6.EE.8a | PARTIAL two-auditor sweep 03 Oct 2026 |
equationsSaplingQ L6 branch 1 'A car drives at a steady 60 mph. Which is the DEPENDENT variable?' — correct: the distance traveled; distractors: the independent variable and 'neither — they do not depend on each other'; why gives the changes-in-response test. ▶ Play a live item from this skill |
Equations Level 6 asks 'Which is the DEPENDENT variable?' for three fixed situations (apples at $2 a pound, a car at a steady 60 mph, $15 per lawn), about one Level 6 question in five, in English and Spanish. The independent variable is only ever a wrong choice; no item asks the child to pick it out. PARTIAL. Blocking clause: Identify the independent ... variable |
| 6.EE.8b | FULL adversarial: UPGRADE · two-auditor sweep 03 Oct 2026 |
equationsSaplingQ L6 (twoVarRelationQ form 0, typed d = kt or d = t + k) From a table of distance d and hours t, type d in terms of t (d = …) — the standard's d = 65t pattern, or d = t + k from an additive table. ▶ Play a live item from this skill |
26 Sep 2026 (round 2): the child types d = kt from a table of distances and hours — the standard's own d = 65t pattern — with the dependent variable d required alone on one side; the ordered pairs are given, not listed or graphed by the child. 10 Oct 2026: the table is proportional (d = kt) or additive (d = t + k) in equal shares, and the hint asks whether d is t times something or t plus something — both shapes named, neither given away. |
| 6.EE.8c | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
equationsSaplingQ L6 (twoVarRelationQ: table → equation, table → value beyond the table, shown graph → equation) 'The graph shows distance d (y-axis) for hours t (x-axis). Which equation shows the relationship?' over three plotted points at t = 1, 2, 3 — e.g. (1, 2), (2, 4), (3, 6) or (1, 3), (2, 4), (3, 5); the four options use one adjacent pair of numbers, {1, 2} (d = t, d = 2t, d = t + 1, d = t + 2) or {2, 3} (d = 2t, d = 3t, d = t + 2, d = t + 3), and any of the four can be the key; the table item offers the same four-equation set for an adjacent pair from 2 to 8 (e.g. d = 4t, d = 5t, d = t + 4, d = t + 5) beside a t | d table; 'Using the relationship, what is d when t = 6?' from the table. ▶ Play a live item from this skill |
Equations Level 6 shows a table of distance d against hours t: the child types the equation from it (about one Level 6 question in eleven) or uses the relationship to find d for a time beyond the table (about one in five), in English and Spanish. Since 10 Oct 2026 the relationship is proportional (d = kt) or additive (d = t + k), each in half the deals, and the shown-graph item (three plotted points, at t = 1, 2, 3 in every deal) and the table item each offer FOUR equations — the key, the same form with the wrong number, the other form with the key's number and the other form with the wrong number. The two numbers are an adjacent pair drawn first ({1, 2} or {2, 3} on the graph, so d = t is a real key; any adjacent pair from 2 to 8 on the table) and the key takes either number and either form, so every option has one twin in form and one in number and every equation shown is the key in a quarter of the deals that show it — none is shown that is never the key. Measured over 600+ deals per language, 'the option with a twin in form and a twin in number' scores 25% on the four options (chance) where it scored 75% on the old three (chance 33%); 'pick a product', 'pick a sum', the smaller or larger number, the shortest or longest option and 'strike any option never seen as the key' all score 25% as well. Every wrong is named in the explanation with the point (or row) it does and does not fit, and the key is re-derived from the plotted points in saplingGrade6Gaps.test.ts, which locks these rates at the 95% bound. Choosing the equation now takes reading the graph or the table, so each is related to the equation; the graph is never related to the table, no value is read off the graph, and the analysis stops at choosing the equation. Verdict held at PARTIAL for the second re-audit. Blocking clause: Analyze the relationship ... using graphs ... and relate these to the equation |
| 6.G.1 | PARTIAL adversarial: NOTE-ONLY · two-auditor sweep 03 Oct 2026 |
geometrySaplingQ L6-L7 L6: triangle area (half-parallelogram why), parallelogram area (perimeter distractor), trapezoid area (missing-÷2 distractor). L7: L-shaped rectilinear figure by decompose-and-subtract; kite area via diagonals (half-rectangle why). ▶ Play a live item: level 6 · level 7 |
Geometry L6-7 asks the area of a triangle (from base and height), a parallelogram, a trapezoid, a kite (from its diagonals) and an L-shaped rectilinear figure given as a rectangle with a corner missing, which the child finds by subtracting the corner; the explanations derive each formula by composing or decomposing (a triangle is half a parallelogram, a kite fills half a rectangle). English and Spanish; the answer is chosen from three options. Not practised: none of these items has a real-world setting — each is a bare shape labelled cm, m, 'units' or nothing — so 'apply these techniques in the context of solving real-world problems' is not met. No figure is drawn, and the L-shape item states its decomposition in the prompt. Blocking clause: apply these techniques in the context of solving real-world ... problems |
| 6.G.2 | PARTIAL two-auditor sweep 03 Oct 2026 |
volumeSaplingQ L4-L5 L4: V = lwh with a fractional (x.5) height, the half-layer meaning taught, rounded-edge distractors. L5: V = Bh given base area B and fractional height, with the B-already-is-the-area why; the volume is typed on the number pad in interactive rounds ('What is the volume, in cubic centimeters?') and picked from three on printed sheets, the placement check and Duel. ▶ Play a live item: level 4 · level 5 |
Volume L4 asks for the volume of a right rectangular prism with one edge of n.5 units by V = lwh; the explanation reads the half unit as half a layer. Volume L5 gives the base area B (any whole number from 6 to 24, so the volume itself ends in .5 about half the time) and an n.5 height and says 'Use V = Bh'. Since 24 Oct 2026 (founder decision) interactive rounds have the child type the volume on the number pad ('What is the volume, in cubic centimeters?') with no choices on screen; the pad shows a decimal-point key only when the volume needs one, which is exactly when B is odd, and a typed slip (the half dropped, B + h added, or the height rounded up) gets its own explanation. Printed sheets, the placement check and Duel cannot take typing and keep three choices: the two wrong ones are a pair drawn from seven named slips (the half dropped, the height rounded up, B + h added, both bases added on, one layer too many, one layer short, the half layer worked out as B ÷ ½), so the answer is the smallest, middle or largest choice about a third of the time each, no choice is ever double (or another whole-number multiple of) another, and no single .5, doubling or 'two closest choices' rule picks it more than about 44% of the time (chance 33%). But there the choices still give the answer away to a determined guesser: how many choices end in .5, read together with which two choices sit closer (or whether all three are evenly spaced), finds the answer in about 81% of deals (chance 33%). Both are bare mathematical prisms: no item at volume L4-L5 puts a prism volume in a real-world setting. The fractional edge is always exactly one half, always on the height; length and width stay whole. English and Spanish. Blocking clause: in the context of solving real-world and mathematical problems (no real-world item) |
| 6.G.3 | PARTIAL two-auditor sweep 03 Oct 2026 |
coordinateSaplingQ L5 Distance between two points sharing an x-coordinate (crossing the axis), which is the side-length skill for vertical polygon sides. ▶ Play a live item from this skill |
Drawing exists in a limited form: half of coordinate-grid L5 deals have the child plot a rectangle's four given vertices on a tap grid, then read the length of its horizontal side from the two same-second-coordinate vertices ('how long is the side from (2, 1) to (4, 1)?'). Limits that remain in this generator: the polygon is always a rectangle in the first quadrant (the tap grid has no negative axes), the child taps the vertices rather than drawing the sides, only the horizontal bottom side is ever asked, the separate two-point distance item is between bare same-first-coordinate points and is not framed as a polygon's side, and neither item carries a real-world context. Blocking clause: Draw polygons whose edges meet at right angles (rectilinear figure polygons) in the coordinate plane given coordinates for the vertices |
| 6.G.4 | PARTIAL two-auditor sweep 03 Oct 2026 |
geometrySaplingQ L8 Cube: 'a net is 6 equal a × a squares' → total surface area (volume a³ as distractor). Rectangular prism: net described as three rectangle pairs → SA = 2(lw + lh + wh), with the forgot-the-2 and volume distractors. ▶ Play a live item from this skill |
Geometry L8 computes surface area from a net described in words: 'A cube's net is 6 equal 3 × 3 squares. What is the total surface area?' and 'A rectangular prism's net has 2×4, 2×3, and 4×3 rectangles — two of each.' About one geometry L8 question in ten on a first round at that level (about one in twenty when the level is replayed, since the item is held back for two rounds once shown) shows three drawn nets — a cross of six squares, a T of five squares and a 2 × 3 block — and the child picks the one that folds into a cube. Not practised: the child never represents a figure with a net (the picker is a choice among the same three drawings, and its answer is always the cross); no net ever contains a triangle; the surface-area nets are words, not drawings; and the one real-world surface-area item, a soup can, is a cylinder worked without a net. Blocking clause: Represent three-dimensional figures using nets made up of rectangles and triangles |
| 6.SP.1 | PARTIAL two-auditor sweep 03 Oct 2026 |
statisticsQ L1 'Which is a STATISTICAL question?' and 'Which is NOT?' over a 7-entry bank of paired questions (class homework minutes vs YOUR minutes, etc.), why: statistical questions anticipate variety. ▶ Play a live item from this skill |
Recognition is covered in both directions with well-matched pairs echoing the standard's own example. 'Generate' is a production verb with no item behind it — the child never writes a statistical question. Blocking clause: Recognize and generate a statistical question |
| 6.SP.2 | FULL two-auditor sweep 03 Oct 2026 |
statisticsQ L2-L4 Compute mean, median (sort-first taught; typed on the number pad in every interactive deal), mode, range from small data sets; compare spread of two groups by range; identify the outlier in a clustered set; identify which measure an outlier's removal changes most. ▶ Play a live item: level 2 · level 3 · level 4 |
Center (mean and mode, plus a median item), spread (range, IQR and MAD) and shape-via-outlier are all analyzed at L2-L4. The standard's own lists are and/or lists, so range and outlier alone satisfy the letter. About a quarter of L4 deals have the child type a range, a MAD or an IQR; the IQR comes from seven ordered values. The IQR is also read off a box plot at L5. Weaknesses an adversary will use: no item in Data & Statistics asks the child to identify a cluster, a peak, a gap or symmetry, and none names skew — 'clusters' appears only in the L4 outlier explanation. The nearest asks are the L5 dot-plot mode, the L5 histogram question 'Which interval holds the MOST students?' (the tallest bar stands over any of the three intervals), and one L6 two-samples form. That form describes a lopsided sample from its mean, mode and range without naming skew. The L2 median item prints its five values in a shuffled order, never already sorted, so the median can sit at any printed place (it is the middle printed value in about one deal in five), and each list has one far value whose mean is itself one of the five values. Since 24 Oct 2026 (founder decision) interactive rounds have the child type the median on the number pad in every deal, with no choices on screen; typing the list's mean or the middle of the list as printed gets that slip's own explanation. Printed sheets, the placement check and Duel cannot take typing and keep three choices, all values from the list, the mean among them in about five deals in eight. There, no rule on the choices' size order or spacing (alone or seen from the side the far value sits on) finds the answer more than about 42% of the time (chance 33%), and the choice nearest the mean, or nearest the midpoint of the extremes, about 37%; but the choices still give the answer away, stated openly: the mean is always the median's neighbour in the sorted list, so working out the mean exactly and taking the closest other choice finds the median every time; with no arithmetic, 'if a choice is the list's largest or smallest value, take the choice farthest from it, otherwise the middle choice' finds it about 75% of the time (it cannot fall much below two in three while every choice is a list value), about 80% read with the list's range, and up to about 89% with the choices' span as well; 'if the smallest choice is 6 or less, take the largest, otherwise the smallest' finds it about 58% of the time; and the choice second-closest to the midpoint of the extremes about 63%. The FULL rests on the and/or reading. |
| 6.SP.3 | FULL two-auditor sweep 03 Oct 2026 |
statisticsQ L2-L4 Range why: 'measures how SPREAD OUT the data is, not where it sits'; group-spread item: 'Spread asks how far apart the data sits — not which group has bigger numbers'; house-price item picks the center measure; outlier-removal item contrasts mean/median/mode sensitivity. ▶ Play a live item: level 2 · level 3 · level 4 |
The child recognizes across several items that a center summarizes and a variation describes spread, and the explanations say it in words. The range explanation says it 'measures how SPREAD OUT the data is, not where it sits'. The two-group spread item says 'Spread asks how far apart the data sits — not which group has bigger numbers'. In the house-price item, 'the range' must be rejected as a measure of center. At L6 one item prints two medians and two ranges: the medians are equal in half the deals and far apart in the other half, and the options cross a center claim ('a typical value is about the same' or 'higher for' one group) with a spread claim (which group varies 'far more'), so the center is read off the medians and the spread off the ranges. Two of the three wrong combinations are dealt — the narrower group named as the one that varies more, the wrong center claim (a bigger range read as higher typical values when the medians match, or the gap between the medians ignored), or both — and each is named in the explanation. Weakness: in the two-group spread item the key is always the option that pairs 'more spread out' with 'its range is larger', and it is also the shortest option in every deal, so it can be picked without reading the numbers. |
| 6.SP.4 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Statistics L5 (statisticsDisplayQ) serves reading items over drawn displays - a dot plot (total count, mode), a histogram with three interval bars (tallest interval, total), a box plot (median, range, IQR, share of data inside the box), a stem-and-leaf display (how many values, largest, smallest) and a circle graph. Example: "The box plot summarizes pages read. What is the interquartile range (IQR)?" In interactive rounds the child also builds a dot plot from listed data on the shipped plot builder ("Make the dot plot: one ✕ for each data point"). BLOCKING: the standard's verb is Display, and the dot plot is the only one of these displays the child makes. No construct widget builds a histogram, a stem-and-leaf display or a box plot - the plot builder makes dot/line plots, bar graphs and picture graphs only - so those three are read, never displayed by the child. Blocking clause: Display numerical data on histograms, stem-and-leaf plots, and box plots |
| 6.SP.5 | PARTIAL two-auditor sweep 03 Oct 2026 |
statisticsQ L2-L4 Data summarized in context (pets per family, quiz scores, house prices): measures computed, outliers named, measure choice justified. ▶ Play a live item: level 2 · level 3 · level 4 |
The umbrella is carried by its four parts. 5a and 5b are identification items at Level 5, about one L5 deal in five. Over short investigation descriptions they ask 'How many observations does this data set have?', 'What ATTRIBUTE was under investigation?' and 'What UNIT were the observations recorded in?'. The child also counts the ✕s on a dot plot and the leaves on a stem-and-leaf display. 5c's measures are typed at Levels 2 and 4, and 5d's choice of a measure of center is asked at Levels 3-4. The row is graded PARTIAL, like its parts, for three reasons. The attribute and the outlier are only picked from three choices; the child never describes the attribute, the overall pattern or the deviation. No item asks how the data was measured, although each description states it (with a stopwatch, a ruler, a scale). No item relates the choice of a measure of variability to the data. Blocking clause: describing the attribute (5b) and the overall pattern and deviations (5c), and relating the choice of a variability measure to the data (5d) |
| 6.SP.5a | PARTIAL two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Statistics L5 asks for the number of observations four ways: from a short investigation description ("How many observations does this data set have?" over "Ravi counted the raisins in 18 snack boxes..."), from a drawn dot plot ("How many families IN TOTAL does the dot plot show?"), from a drawn histogram (total of the three bars) and from a stem-and-leaf display ("How many values does it hold?"). Together these are about one statistics L5 question in four, in English and Spanish. BLOCKING: "reporting" as free production - every one of these items is answered by choosing a number from three options. The statistics skill does type some answers (mean, median, mode and range at L2; range, MAD and IQR at L4 and L6), but never the number of observations, and no item has the child report the count as part of a written summary of a data set. Blocking clause: Reporting the number of observations: every count item is chosen from three options |
| 6.SP.5b | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Statistics L5 (attributeUnitsQ) gives a one-sentence investigation and asks about it directly: "What ATTRIBUTE was under investigation?" and "What UNIT were the observations recorded in?" over descriptions such as "Ana measured the height of 10 sunflower seedlings with a ruler, recording centimeters." Six investigations, each with an English and a Spanish version. Limits: how the attribute was measured (ruler, stopwatch, thermometer, scale) appears only in the description - none of the generator's three forms asks about it; the attribute and the unit are picked from three options rather than described; and 'describing' as free production has no interaction mode in this skill. Blocking clause: Describing the attribute, including how it was measured |
| 6.SP.5c | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
statisticsQ L2 + L4 Mean, median, mode, and range each computed as its own question form; the outlier item names the striking deviation in context (pets per family) and the why describes the cluster near the common value. ▶ Play a live item: level 2 · level 4 |
Mean, median, mode and range are computed at Level 2: the median is typed on the number pad in every interactive deal (since 24 Oct 2026), the mean, mode and range in about half of interactive deals; printed sheets, the placement check and Duel keep choices. Range, MAD and IQR are typed at Level 4, in about a quarter of L4 deals. The outlier item's context is minutes to finish a quiz, and its explanation describes the cluster near the common value. The row stays PARTIAL because no item has the child describe the overall pattern and the striking deviations in words; the outlier is picked from three of the listed values. Blocking clause: describing any overall pattern and any striking deviations from the overall pattern with reference to the context |
| 6.SP.5d | PARTIAL two-auditor sweep 03 Oct 2026 |
statisticsQ L3 branch 0; L4 branch 1 'Which measure of center BEST describes a typical house?' — median, because the $200k+ outlier drags the mean (with 'the mean — it uses every value' as the trap); 'if the outlier is removed, which measure changes MOST?' — the mean, with the why stating that choosing the measure to fit the data's shape IS the skill. ▶ Play a live item from this skill |
The choice of a measure of CENTER is related to the data's distribution at Levels 3-4. One item asks 'Which measure of center BEST describes a typical house?': the median, because one $200k-$350k house drags the mean. Off a list of six values with one outlier, printed only as 'Data: …', items ask which measure best describes a typical time, or which changes most when the outlier is removed. The row is held PARTIAL because no item asks the child to choose a measure of VARIABILITY (range, IQR or MAD) to fit the data. Why the IQR resists an outlier appears only in the explanation after an IQR is computed. Blocking clause: Relating the choice of measures of variability to the distribution of the data |
| Code | Verdict | Evidence (generator · forms) | Notes |
|---|---|---|---|
| 7.RP.1 | FULL two-auditor sweep 03 Oct 2026 |
ratiosSaplingQ L6 kinds 0-2 Price per notebook (whole/half-dollar), train unit rate in mph, better-buy per-ounce comparison. ▶ Play a live item from this skill |
Generator: complexRateQ, dealt from Ratios L6 (about one L6 question in three, held to at least two per round). Forms: a typed complex-fraction unit rate, for example 'A hiker covers 3/4 of a mile in 2/3 of an hour. What is the unit rate, in miles per hour?', answered with the / key, in like-unit (gear turns, juice to water), area (acres per bag, acres per hour) and different-unit contexts. Note: 26 Sep 2026 (round 1, two auditors agreed FULL): complexRateQ types the unit rate (numpad, unconditional in live rounds); like-unit, area and different-unit contexts are all dealt; at least 2 per round. |
| 7.RP.2 | PARTIAL two-auditor sweep 03 Oct 2026 |
ratiosSaplingQ L8 Is-this-table-proportional (constant y/x, broken-last-pair case, does-not-need-(1,1) distractor); find k from a proportional table. ▶ Play a live item from this skill |
Represent is practised with a typed form. At Equations L6 (about one question in eleven, live rounds) the child reads a distance-time table such as 1 | 7, 2 | 14, 3 | 21, 4 | 28 and types d = 7t. The check requires d alone on one side, so 7t = d is accepted and t = d/7 is not. Proportional lines are also plotted from printed coordinates (Functions L5 'Plot the proportional line: (0, 0), (1, 2), (2, 4), (3, 6)'; Ratios L5 plots the pairs of an equivalent-ratio table). Ratios L8 has the child pick t = pn for a priced-per-item story and y = kx for a shown graph. Recognize is not credited. The one item that asks whether a relationship is proportional (Ratios L8 'Table: (3, 9), (5, 15), (8, 27). Is the relationship proportional?') gives its answer away through its options. 'no — the last pair breaks the constant ratio' appears only when it is the answer, and when it is absent the answer is always 'yes'. PARTIAL until those options are balanced. Blocking clause: Recognize and represent proportional relationships between quantities |
| 7.RP.2a | PARTIAL two-auditor sweep 03 Oct 2026 |
ratiosSaplingQ L8 kind 0 Three (x,y) pairs; decide proportional or not. Non-proportional case breaks the LAST pair's ratio; distractors include 'no — does not start at (1,1)' and 'cannot tell from three pairs'. ▶ Play a live item from this skill |
Ratios L8 asks 'Table: (3, 9), (5, 15), (8, 27). Is the relationship proportional?' in English and Spanish, but the options give the answer away. 'no — the last pair breaks the constant ratio' is offered only when it is the answer. When it is not offered, the answer is always 'yes — every y is k times its x'; the other options, 'no — the pairs do not start at (1,1)' and 'it cannot be determined from three pairs', are never right. Under the published rule an item whose options reveal the key does not count. The graph method appears only at Ratios L5, where the child plots the pairs of an equivalent-ratio table and picks 'the points line up on a straight line through (0, 0)'. That is the same answer every time, and the item never asks whether the relationship is proportional. PARTIAL. Blocking clause: Determine whether two quantities are in a proportional relationship |
| 7.RP.2b | PARTIAL two-auditor sweep 03 Oct 2026 |
ratiosSaplingQ L8 kind 1 Given a proportional table, find the constant of proportionality k; distractors k·x1 (the first y) and x1; second row confirms in the why. ▶ Play a live item from this skill |
Ratios L8 asks for the constant of proportionality from a two-row table labelled proportional ('In the proportional table (4, 32), (7, 56), what is the constant of proportionality k?', k = 2-8, picked from three options). A shown graph asks what the point (1, r) means (one ticket costs $r, in one of three wordings such as 'buying 1 ticket costs $2'; r = 2 or 3). One pair is enough to answer either: the table is already called proportional, and the 'second row confirms it' step is only in the explanation. The one item that makes a child check that every pair shares one rate, the Ratios L8 'Is the relationship proportional?' table, gives its answer away through its options (see 7.RP.2a). The items call the constant k or r, never m; Functions L5 (Grade 8) shows the same slope between any two points with one fixed answer sentence. PARTIAL. Blocking clause: recognize that, in a proportional relationship, every pair of numbers has the same unit rate |
| 7.RP.2c | FULL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Represented with a typed equation at Equations L6 (live rounds, about one question in eleven, English and Spanish): 'The table shows distance d for hours t. Write d in terms of t (d = …).' over a table such as 1 | 7, 2 | 14, 3 | 21, 4 | 28. The child types d = 7t. The check requires d alone on one side, so 7t = d is accepted and t = d/7 is not. Ratios L8 adds the standard's own t = pn form as a choice: 'Each pencil costs $8. Which equation gives the total cost t of n pencils?' (t = 8n against t = n + 8 and n = 8t; about one Ratios L8 question in twelve). It also has a shown-graph form ('Which equation gives the relationship?', y = 2x or 3x). The typed item uses one context (distance and time, rate 2-8); the cost form is chosen, not written. Equations L7's flat-fee rental ('a flat $9 plus $5 per hour') is not proportional and is not counted. |
| 7.RP.2d | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Ratios L8 deals proportionalGraphQ for about one question in three, and two of its four forms interpret the special points of a proportional graph: "The graph shows the cost y (dollars) of x tickets. What does the point (1, 2) mean?" (one ticket costs $2, in one of three wordings such as 'buying 1 ticket costs $2', against the reversed rate, for example '$1 buys 2 tickets', and, since 10 Oct 2026, a starting-cost reading, for example 'the price before buying a ticket is $2', in place of a second copy of the reversed rate, so the key is no longer the only option without '$1'; the explanation names the unit rate) and "The graph shows the cost y of x tickets. Why does the line pass through (0, 0)?" (zero tickets cost zero, for example 'for zero tickets you pay zero dollars'). English and Spanish; each is about one ratios L8 question in thirteen. The grid plots (0, 0) together with (1, r), (2, 2r), … and draws the line y = rx from the origin to the top of the grid (since 10 Oct 2026, commit 282dd18; before that only the points from (1, r) onward were plotted and no line was drawn), with the three answer options beneath it. Further limits in this generator: the only points interpreted are (0, 0) and (1, r) - no form asks what a general point (x, y) on the graph means in terms of the situation; r is only 2 or 3; one context (tickets and cost); and the meaning is offered as three options rather than explained by the child. Blocking clause: Explain what a point (x, y) on the graph means in terms of the situation |
| 7.RP.3 | FULL adversarial: NOTE-ONLY · re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
percentsSaplingQ L8; ratiosSaplingQ L5-L7 Tax on a purchase → total (two-step, with the ×1.15-style one-step in the why); percent change from old to new price; recipe scaling, missing-value ratio tables, unit-rate problems. ▶ Play a live item from this skill |
Multistep percent and ratio solving both happen. Percents L8 serves tax then total ('A $120 purchase carries 15% tax. What is the total?'), percent increase and percent decrease ('A price falls from $140 to $70. What is the percent decrease?'). In about one question in three it serves simple interest, tip then total, or commission. Ratios L5-L7 add missing-value tables, unit rates and recipe scaling. Not present: percent error, markups named as such, or a fee charged as a percent (fees appear only as flat dollar charges in equation items). Percent-change answers are drawn from {10, 20, 25, 50}. Prices are multiples of $20 (interest principal multiples of $100, sales multiples of $500). The commission item always shows ten times the key among its options. |
| 7.NS.1 | PARTIAL two-auditor sweep 03 Oct 2026 |
integersG7Q L5 (−a) + b computed; a − (−b) computed; distance between p and q asked numerically. ▶ Play a live item from this skill |
At integers L5 the child taps where a frog lands for a sum starting below zero or a difference that crosses zero. Still PARTIAL: the child only taps the landing number — the app marks the start and lights the jump after the answer, so the child never represents the operation; every number is printed on its own button, and the buttons wrap onto two or more rows rather than forming one line; and adding or subtracting a negative is never shown on the line. Blocking clause: Represent addition and subtraction on a horizontal or vertical number line diagram |
| 7.NS.1a | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Integers L5-7 serve the additive-inverse idea three ways: the bare zero pair ('7 + (−7) = ?', answer 0), the temperature that rises and falls by the same amount ('What was the total change?', answer '0° — the opposite changes cancel'), and the same morning rise asked the other way ('What afternoon change would bring it back to where it started?', answer the exact opposite). In sampled six-question rounds these were about 8% of questions at L5, 5% at L6 and 6% at L7. Limits: the temperature story is the only situation; it is always given, and the child never describes or supplies a situation; the child picks among three choices, so nothing is shown or modelled (no zero-pair counters); and the items share a bank with the integer-operation items and with forms credited to other standards. A written explanation ('Explain why 8 + (−8) = 0…') is asked only through the metered Writing Coach grader, which this report does not count. Blocking clause: Describe situations in which opposite quantities combine to make 0. Show that a number and its opposite have a sum of 0 |
| 7.NS.1b | PARTIAL two-auditor sweep 03 Oct 2026 |
integersG7Q L5 form 0 (−a) + b = ? with sign-error distractors; why narrates starting below zero and climbing. ▶ Play a live item from this skill |
The sum is usually just computed and picked (with sign-error traps). About one L5 question in twenty is the addition frog hop, where the child taps where a jump of q from p lands on a row of numbers; but q is always positive there, so p + q with a negative q is never shown — the only leftward jump on the line is subtracting a positive number (the subtraction hop, about one L5 question in twenty-two). Every number is printed on its own button, and the app lights the jump only after the answer. A child can still pass the computed items on sign rules alone. Blocking clause: Show p + q as the number located a distance |q| from p, in the positive or negative direction |
| 7.NS.1c | PARTIAL adversarial: DOWNGRADE · re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
integersG7Q L5 form 1 a − (−b) = ? with distractors a − b (ignoring the double sign) and −(a + b); why states p − q = p + (−q) via the remove-a-debt story. ▶ Play a live item from this skill |
Subtracting a negative is computed with integers ('7 − (−4) = ?') and with signed decimals ending in .5 ('2.5 − (−8.5) = ?'), where the a − b trap catches anyone who does not add the opposite; one form asks the child to pick the rewrite of p − q as addition ('Rewrite 9 − 12 as addition' → 9 + (−12)). Still PARTIAL: the child picks the rewrite and never writes it (the wrong rewrites are two of p + q, −p + q, q + (−p) and −p + (−q), so brackets do not mark the key), the first number is always positive, and no signed fraction appears. Blocking clause: Model subtraction of rational numbers as adding the additive inverse, p − q = p + (−q) |
| 7.NS.1d | PARTIAL adversarial: DOWNGRADE · re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
integersG7Q L5 form 2 What is the distance between p and q on the number line? (p negative, q positive); distractors |q + p| and |p|. ▶ Play a live item from this skill |
Distance-as-absolute-difference is directly the question. The pair either straddles zero (p negative, q positive, with a |q + p| distractor that punishes sign confusion) or is two negatives on the same side of zero (with an |a| + |b| distractor). Weaknesses: two positive points never generate, the distance items use integers only, and no line is rendered. Blocking clause: Model subtraction as the distance between two rational numbers on the number line where the distance is the absolute value of their difference |
| 7.NS.1e | PARTIAL two-auditor sweep 03 Oct 2026 |
integersG7Q L5 Integer addition/subtraction with negatives across three forms. ▶ Play a live item from this skill |
Adding and subtracting signed INTEGERS is exercised, and two L5 forms put signed decimals on the operands ('(-8.5) + 2.5 = ?', '4.5 − (−1.5) = ?') — but every decimal ends in .5 and the answers are whole numbers. Signed fractions are never added or subtracted in any Grade 6-8 item; a negative fraction appears only in the reciprocal item ('What number do you multiply −3/4 by to get 1?') and inside the rationale of the quotient-of-integers item. No integers L5-7 item asks the child to use a property of operations as a strategy for adding or subtracting signed numbers. Blocking clause: add and subtract rational numbers |
| 7.NS.2 | PARTIAL two-auditor sweep 03 Oct 2026 |
integersG7Q L6 (−a)(−b), (−a)(b), and (−n)÷d computed with sign-rule distractors. ▶ Play a live item from this skill |
Sign rules for integer products and quotients are exercised at integers L6 (parity-of-negatives whys). One L6 form uses signed decimals ('(−3.5) × 8 = ?', a negative integer ÷ 1.5) and one L7 context uses a .5 rate, always with whole-number results. But 'all rational numbers' includes signed fractions, which are never multiplied or divided in any Grade 6-8 item — a negative fraction appears only in the reciprocal item ('What number do you multiply −3/4 by to get 1?') and inside the rationale of the quotient-of-integers item. Blocking clause: multiply and divide all rational numbers |
| 7.NS.2a | PARTIAL two-auditor sweep 03 Oct 2026 |
integersG7Q L6 forms 0-1 (−a) × (−b) with −ans distractor; single-negative product; whys give flipping-across-zero and count-the-signs arguments. ▶ Play a live item from this skill |
The child computes signed products with the rules ('(−4) × (−4) = ?'), and the feedback explains them as flipping across zero and counting the negative signs. Nothing in the app — question or feedback — derives the rules from the distributive property or shows (−1)(−1) = 1. 'Describe' is a production verb; a written explanation ('Explain why (−3) × 4 = −12, and then why (−3) × (−4) = 12…') is dealt only through the metered Writing Coach grader, which this report does not count. Blocking clause: Describe how multiplication is extended from positive rational numbers to all rational numbers |
| 7.NS.2b | PARTIAL two-auditor sweep 03 Oct 2026 |
integersG7Q L6 form 2 (−nd) ÷ d = ? with sign distractors; why verifies by multiplying back. ▶ Play a live item from this skill |
Signed integer division is computed ('(−24) ÷ 4 = ?'). Two choice items at integers L5-7 touch the rationality idea, but neither can count: 'What is ALWAYS true of the quotient of two integers, like −5 ÷ 3?' has the same right sentence in every deal ('it is a rational number (as long as the divisor is not 0)'), and in 'What kind of number is −12 ÷ 4?' the right answer is the only choice with a number in it. The child never explains either idea; the only written prompt nearby is a metered Writing Coach item about the sign rule for division ('Explain why −12 ÷ 3 = −4 but −12 ÷ (−3) = 4…'), which this report does not count and which does not ask about rational numbers. The identity −(p/q) = (−p)/q = p/(−q) is never assessed: in the integers skill a negative fraction appears only as a wrong choice and in feedback. Blocking clause: Explain that integers can be divided ... −(p/q) = (−p)/q = p/(−q) |
| 7.NS.2c | PARTIAL two-auditor sweep 03 Oct 2026 |
integersG7Q L6 Integer sign-rule products and quotients, three forms. ▶ Play a live item from this skill |
Signed products and quotients are computed with integers and, in one L6 form, with signed decimals ending in .5 ('(−1.5) × 6 = ?', a negative integer ÷ 1.5), always with whole-number results. No signed fraction is ever an operand, and no integers item asks the child to use a property of operations (commutative, associative, distributive) as a strategy for multiplying or dividing. Blocking clause: multiply and divide rational numbers |
| 7.NS.2d | PARTIAL two-auditor sweep 03 Oct 2026 |
longDivisionQ (integers L6); percentsSaplingQ L7 (SAPLING_EQUIV bank) Long division at integers L6: supply the next quotient digit from a quotient-so-far and a remainder, read a returning remainder (repeats) or a remainder of 0 (terminates), decide terminate-or-repeat from a lowest-terms denominator, convert generated fractions (denominators such as 7, 12, 13, 16, 24) to decimals. Also at percents L7: Which decimal equals p%? Which fraction equals 0.xx? — memorized halves/quarters/fifths/tenths equivalences. ▶ Play a live item from this skill |
Long division serves as an algorithm at integers L6: the child supplies the NEXT quotient digit from a quotient-so-far and a remainder, reads what a returning remainder (the digits repeat) or a remainder of 0 (the decimal terminates) means, decides terminate-or-repeat from a lowest-terms denominator, and converts generated fractions (denominators such as 7, 12, 13, 16, 24) to decimals; every full round at this level carries at least two such items. Held PARTIAL because each step is selected rather than carried out unaided — 'using long division' names a method the app cannot watch a child perform — and every fraction is positive and proper (no negative rationals, no improper fractions). Blocking clause: convert a rational number in the form of a fraction to its decimal equivalent using long division; know that the decimal form ... terminates in 0s or eventually repeats |
| 7.NS.3 | PARTIAL two-auditor sweep 03 Oct 2026 |
integersG7Q L7 Temperature drops r° per hour for h hours (signed product in context); friends split a −$ debt (signed quotient in context). ▶ Play a live item from this skill |
At integers L7 the real-world items are multiplicative: a temperature that drops r° an hour for h hours (r from 2 to 5, or a .5 rate from 1.5 to 3.5, always with a whole-number total) and friends splitting a debt. The only additive situations are the rise-and-fall temperature stories, whose total is always 0. No problem adds, subtracts, multiplies or divides fractions; at L7 fractions appear only in the reciprocal and repeating-decimal items. In the debt item the wrong shares are drawn on both sides of the answer (the share with its sign dropped, the whole debt, a dollar or two off), so the answer's rank varies and the division is needed. Complex fractions do serve elsewhere (unit rates at ratios L6) but not as four-operations problems here. Blocking clause: problems involving the four operations with rational numbers ... extend the rules for manipulating fractions to complex fractions |
| 7.EE.1 | PARTIAL adversarial: DOWNGRADE · re-audited ×1 · re-audited ×2 (CONFIRMED-AFTER-FIX) · two-auditor sweep 03 Oct 2026 |
expressionsG7Q L5 Factor −ax − ab → −a(x + b) with sign-trap distractor; expand −(x − b) with the leave-the-second-term-alone trap; combine ax − cx to a negative coefficient. ▶ Play a live item from this skill |
All four verbs appear in expressionsG7Q (expressions Level 5), all as multiple choice: factor −ax − ab, expand −(x − b), expand (1/2)(…) or 0.5(…), combine ax − cx, and add (m.5x + b) + (k.5x − d). The two expand items and the add item genuinely require the work, and only the expand-(1/2) and add items' options pass the answer-giveaway guard (everyAskGuard). Since 11 Oct 2026 the add item deals four options: the key and three named slips (the coefficient slipped — the halves dropped, or each half counted as a whole — the constant subtracted the wrong way round, and both), so every option is one edit from two others and no reading of the options finds the key: the option sharing the most with the others, the odd one out and the middle-length option are each the key in 25% of deals over 900 deals per language, where chance is 25%. On the combine and factor items the right answer is the option that shares the most with the other two, counting signs as well as numbers. On combine (3x − 5x) the key −2x shares its minus sign with one wrong answer (−2x²) and its 2x with the other (2x), and it is also the middle-length option, in every deal. On factor it shares its outside factor with one and its bracket with the other in every deal. On expand −(x − b) both wrong answers keep the − before b, so the key is the only option with a +. Fraction and decimal coefficients appear only in the expand-(1/2) and add items. The subtraction item (ax − cx) uses whole numbers, and no item subtracts one expression from another. Blocking clause: add, subtract, and factor linear expressions with rational coefficients — the combine, factor and expand −(x − b) items have options that give the key away |
| 7.EE.2 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Equations L7 serves expressionContextQ: in sampled six-question rounds it supplied a little over one question in four (26-28%), and its two rewriting forms together about one question in six (17-19%). Those two forms are this standard: "A price a rises 5%. The expression a + 0.05a equals 1.05a. What does that form reveal?" (answer: raising by 5% is the same as multiplying by 1.05) and "A store offers 25% off a price p. Which expression gives what you PAY?" (0.75p, against two of 0.25p, 1.25p and p − 25, drawn at random). The third form is a reasonableness check, not rewriting. Limits: only percent increase and percent discount contexts, four percent values each; in the increase form the equivalence is stated for the child, and since the 10 Oct 2026 repair all three choices open the same way ('raising by 5% …') and name the percent and a factor; since 31 Oct 2026 the two wrong ones are drawn from five readings of the same length — multiplying by 0.05 (the increase alone) or by 1.50 (5% read as 0.50), adding 1.05 or 1.50 dollars, or dividing by 1.05 (which would lower the price) — so neither the length nor the opening words point to the right one (before the repair it was the only choice with a percent sign). Matching still helps somewhat: picking the choice that names the 1.05 printed in the prompt finds the right one about half the time in English and a little over two times in five in Spanish (one in three by chance), and in English the right choice is also, about half the time, the one sharing the most words with the others or the only one with its operation, the most any surface rule reaches; a child who knows only that 1.05a is a multiplication finds it in about 85% of deals; and neither form asks the child to judge whether two given forms are equivalent; the child chooses among three options and never rewrites an expression; and the rest of the equations L7 bank is equation items. The same idea also appears as an explanation in a percentsSaplingQ L8 why ("adding a percent is multiplying by 1.05"). Blocking clause: rewriting an expression in different forms in a problem context |
| 7.EE.3 | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
percentsSaplingQ L8; equationsSaplingQ L7; integersG7Q L7 Tax-then-total; percent change; flat-fee-plus-rate rental equations; signed real-world products/quotients. ▶ Play a live item from this skill |
Real-life problems with rational numbers are solved at Percents L8 (tax then total, percent change, interest, tips, commissions), Equations L7 (flat fee plus hourly rate, including half-dollar rates such as '$2.50 per hour' in about half of the write-or-choose-the-equation rental items) and Integers L7 (signed rates and changes, including 'The temperature drops 2.5° each hour for 4 hours.'). 'Assess the reasonableness of answers' appears in one family at Equations L7, about 1 question in 10: 'You buy 16 notebooks at $6 each. Your classmate estimates $250 total. Is that reasonable?' (prices $2 to $9, 11 to 29 notebooks). In about half of the deals the estimate is a rounding within a quarter of the product and in the rest it is another product from the same range, 1.7 to 12 times too big or too small, so yes and no are both keys, every estimate is a multiple of 10 whether fair or off, and fair and off estimates share one range of sizes (repaired 10 Oct 2026: before that a fair estimate always fell between $30 and $110 and an off one at $330 or more or $12 or less, so its size alone decided the item). The yes and no options are parallel and both print the product ('16 × $6 is around $96, close to / far from $250'), which still does the mental computation for the child; the explanation names how far off the estimate is ('about 3 times too big'). 'Convert between forms as appropriate' is assessed as a separate skill (Percents L7; Equations L7 '25% off a price p' → 0.75p), not as a step chosen inside a multi-step problem. PARTIAL. Blocking clause: assess the reasonableness of answers using mental computation and estimation strategies; convert between forms as appropriate |
| 7.EE.4 | PARTIAL two-auditor sweep 03 Oct 2026 |
equationsSaplingQ L7; inequalitiesG7Q L4 Choose the equation ph + fee = total for a rental scenario; solve two-step equations and inequalities. ▶ Play a live item from this skill |
The child types two-step equations from rental stories (px + q = r, half-dollar rates included; about one Equations Level 7 question in ten) and, more rarely, p(x + q) = r equations from perimeter and ticket stories. The child also types a budget inequality (pm + q ≤ total) in about three Inequalities Level 4 questions in ten. Word problems are solved by choosing from three. No item has the child construct an equation or inequality and then solve it, no inequality word problem is ever solved, and the letters are always given. PARTIAL. Blocking clause: construct two-step equations and inequalities to solve problems |
| 7.EE.4a | PARTIAL two-auditor sweep 03 Oct 2026 |
equationsSaplingQ L7 Solve px + q = r (negative solutions allowed); model a rental as ph + fee = total. ▶ Play a live item from this skill |
Both forms serve at Equations Level 7, bare and in context: px + q = r (typed and by choice) and p(x ± q) = r (by choice), with the standard's rectangle-perimeter problem, a tickets-plus-fee problem and a skate-rental problem solved by choosing from three. Every full Level 7 round carries at least two. The word problems use whole numbers only; non-integer coefficients appear in rental equations the child writes ($1.50 an hour) and in Level 8 bare equations (0.5x + 3 = 7, (1/4)(x + 2) = 3). One compare item sets Maya's arithmetic beside Leo's algebra, but whether her last number matches his x decides the answer, so the order of operations need not be read. PARTIAL. Blocking clause: equations of the form px + q = r, and p(x + q) = r |
| 7.EE.4b | PARTIAL two-auditor sweep 03 Oct 2026 |
inequalitiesG7Q L4 Solve px + q > r and px < c, with a negative coefficient in about 3 deals in 5; four choices cross the inequality direction (kept right, or flipped wrongly or not flipped) with the boundary's sign (right or lost), and the why names each step and whether dividing flips the sign. ▶ Play a live item from this skill |
The child types the solution of px + q > r (and of −ax < −b, with the flip), and types a gym-budget inequality from words (8m + 20 ≤ 68). Still PARTIAL: no inequality word problem is solved by the child. The gym item asks only for the inequality, and interpretation in context is asked only through Writing Coach. The solve-and-graph number line is always centred on the answer and is always an open circle shaded right, so it can be built without solving. Since 11 Oct 2026 the multiple-choice solve items (px + q > r and px < c) deal four options — the key, the direction turned the wrong way, the boundary's sign lost, and both slips — each wrong answer named in the explanation. The coefficient is negative in three deals in five and the boundary is negative in half, so the child must decide when to flip, and no reading of the options finds the key (each 25% over 900 deals per language, where chance is 25%). Every typed or graphed number is an integer. Blocking clause: Graph the solution set of the inequality and interpret it in the context of the problem |
| 7.G.1 | PARTIAL two-auditor sweep 03 Oct 2026 |
geometrySaplingQ L9 kind 2 Map scale 1 cm : k km, towns c cm apart, find real distance; add-instead-of-multiply distractor. ▶ Play a live item from this skill |
Scale-factor length computation is exercised in one form. Areas under scale (the k² trap — the heart of Grade-7 scale) and reproducing at a different scale never appear, and no drawing is ever shown. Blocking clause: computing actual lengths and areas from a scale drawing and reproducing a scale drawing at a different scale |
| 7.G.2 | FULL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Volume L6 asks which solid a full turn generates: a rectangle spun about one of its sides (a cylinder) and a right triangle spun about one of its legs (a cone), plus a semicircle about its straight edge (a sphere), in English and Spanish, each with an explanation of what each edge sweeps. The child identifies the solid from three choices, which is the act the standard names. Limits: one fixed item per shape, no figure is drawn, and the rotation items are about one volume L6 question in twelve (the rectangle and triangle together about one in eighteen). |
| 7.G.3 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Volume L6 (crossSectionQ, SLICES branch) asks for the cross-section directly: "You slice a cylinder with a plane straight down through its center (perpendicular to the base). What shape is the cross-section?" (a rectangle). Five cases in English and Spanish: rectangular prism parallel to the base; cylinder parallel and perpendicular; square pyramid parallel (a smaller square) and through the apex (a triangle). Limits: the prism is sliced only parallel to its base - no perpendicular or slanted slice of a prism; five fixed cases; no figure is drawn; 'describe' is served by choosing a shape name from three options; and the items share the volume L6 bank, where the crossSectionQ family (slices and rotations together) is about a quarter of questions and the slice items about one question in six. Blocking clause: Describe the two-dimensional figures that result from slicing (describe is a production verb; served by picking a shape name) |
| 7.G.4 | PARTIAL two-auditor sweep 03 Oct 2026 |
geometrySaplingQ L9 kind 0 Circumference 2πr and area πr² of a circle, answers exact in π, with the formula-swap distractor guarded at r=2. ▶ Play a live item from this skill |
Using both formulas is squarely exercised and the rim-vs-inside whys are strong. The derivation clause is a production verb and is never assessed; also no problem ever combines the formulas (e.g. given C, find A). Blocking clause: give an informal derivation of the relationship between the circumference and area of a circle |
| 7.G.5 | PARTIAL two-auditor sweep 03 Oct 2026 |
volumeG8Q kind 0; geometrySaplingQ L8; volumeSaplingQ L5 Cylinder volume πr²h; prism surface area from a described net; V = Bh at the Grade-6 levels. ▶ Play a live item from this skill |
Stem code; graded on its sub-codes. Computation with the formulas exists; the investigate/generalize layer does not — see 7.G.5a-b. Blocking clause: Investigate the relationship between three-dimensional geometric shapes |
| 7.G.5a | PARTIAL two-auditor sweep 03 Oct 2026 |
volumeG8Q kind 0; volumeSaplingQ L5 Cylinder volume in terms of π with the circumference-confusion distractor; prism V = Bh at Grade-6 levels. ▶ Play a live item from this skill |
Volumes are COMPUTED; the generalization — that the same V = Bh spans prisms and cylinders — lives only in the why ('a disk of area r²π stacked h high'). Generalize is a production verb carried by the explanation, so graded down. Blocking clause: Generalize the volume formula for prisms and cylinders (V = Bh) |
| 7.G.5b | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
geometrySaplingQ L8 Cube SA (6 faces) and rectangular prism SA (2(lw + lh + wh)) from described nets, volume-confusion distractors. ▶ Play a live item from this skill |
Prism surface area is computed at geometry L8 from a described net ('A cube's net is 6 equal 3 × 3 squares. What is the total surface area?'; 'A rectangular prism's net has … rectangles — two of each. What is the surface area?'), and cylinder surface area is computed in terms of π at geometry L8 ('A soup can has radius 5 and height 4. What is its surface area, in terms of π?') and L9 ('A cylinder has radius 4 and height 7. What is its surface area, in terms of π? (two lids + the side)'). No item states, asks for or uses the formula SA = 2B + Ph: the prism items list the faces, the L9 cylinder prompt names its parts, and the unrolled side as a 2πr-by-h rectangle is stated only in the L9 explanation. Blocking clause: in the case of a cylinder, perimeter is replaced by circumference |
| 7.G.6 | PARTIAL adversarial: DOWNGRADE · re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
geometrySaplingQ L6-L8; volumeG8Q Triangle/parallelogram/trapezoid/kite areas; L-shape decomposition; cube and prism surface area; cylinder (and cone/sphere) volume, π-exact. ▶ Play a live item: level 6 · level 7 · level 8 |
Area of 2-D figures (geometry L6-7: triangle, parallelogram, trapezoid, L-shape, kite; circle area at L9), volume (volume L4-6: right rectangular prisms, V = Bh prisms, cylinders, cones, spheres, pyramids) and surface area (geometry L8 prism nets and a soup can; geometry L9 cylinder, 'two lids + the side') are all computed, and none of these items requires a root. Limit: among these items the L8 soup can is the only real-world frame; the L9 cylinder, the prism nets, the geometry L6-7 and L9 area items and the volume L4-6 items are bare shapes. Blocking clause: Solve real-world ... problems involving area of two-dimensional objects and volume |
| 7.SP.1 | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
statisticsSamplingQ L6 Pick the most representative sample for a stated population (6 scenarios, cross-scenario bad distractor); explain why a named sample is biased. ▶ Play a live item from this skill |
Statistics L6 is built mostly on samples. Items ask which sample is MOST representative of a stated population, why a named convenience sample is biased, what five random-sample means from one school show together, and what two random samples' centers and spreads say about two populations. The level also deals some bare range, MAD and IQR items. The row is held PARTIAL with its parts: the child never designs a sample (7.SP.1a), and picks rather than states why a sample is biased (7.SP.1b). Blocking clause: generate a valid representative sample (7.SP.1a); justify your reasoning (7.SP.1b) |
| 7.SP.1a | PARTIAL two-auditor sweep 03 Oct 2026 |
statisticsSamplingQ L6 Choose the representative sampling method among biased alternatives; whys carry the equal-chance principle. ▶ Play a live item from this skill |
Knowing that generalization requires representativeness is exercised by selection. GENERATING a sample — designing one rather than recognizing one — is a production verb and never happens. Blocking clause: generate a valid representative sample of a population |
| 7.SP.1b | PARTIAL two-auditor sweep 03 Oct 2026 |
statisticsSamplingQ L6 (bias form) Why is sample X biased for population Y? Correct reason vs 'too large' and 'not paid' foils; six fixed scenario pairs. ▶ Play a live item from this skill |
Statistics L6 asks which sample is MOST representative of a stated population, over six fixed scenarios. The random option is always the key and is the only option that mentions randomness. It also asks why a named convenience sample is BIASED: the key states the scenario's selection flaw, against the same two wrong answers every time ('because the sample is too large to be a fair one', 'because the respondents were not paid for answering'). The row is held PARTIAL for three reasons. The justification is picked, never stated by the child — the rule these reports apply to every justification picked from choices. The key can be spotted without reading the scenario: both wrong answers always begin 'because' ('porque' in Spanish), the key never does, and the wrong answers never change. And the child never sees a random sample that fails to represent its population. Blocking clause: justify your reasoning |
| 7.SP.2 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Statistics L6 lists five random-sample means from one school (sleep hours) and asks what the five show together. The credited answer is that the estimates vary a little around a center near the true mean. Another form projects one random sample's pizza count onto the school, with 'about' credited over 'exactly'. Two-random-sample comparisons draw inferences from printed centers and spreads. The row is held PARTIAL. The child never generates samples or simulations: the five means arrive computed, their size is never stated, and the variation is judged in words only. Both multiple-samples forms always key the same wording ('the estimates vary a little…', 'about …'), so they can be answered without reading the numbers. Blocking clause: Generate multiple samples (or simulated samples) of the same size to informally gauge the variation in estimates or predictions |
| 7.SP.3 | PARTIAL two-auditor sweep 03 Oct 2026 |
statisticsSamplingQ L6 (comparison form) Team A mean vs Team B mean 'with similar spread' — pick the valid informal conclusion; over-claim and refuse-to-compare foils. ▶ Play a live item from this skill |
Both clauses have content at statistics L6. For visual overlap, two samples are drawn as stacked dot plots on matching number lines (both data sets also printed), and the child judges whether they barely overlap or pile together. The difference between the centers is expressed as a multiple of the MAD in three ways. It is typed from two team means and a stated MAD ('The difference between the means is how many MADs?'). It is picked from the same displayed samples, whose MAD is exactly 1. And it is read in words: what 1 MAD versus 2-4 MADs apart means. The row is held PARTIAL on the overlap clause. The drawn pairs are always 1 apart or 6-7 apart, so overlap is a two-level judgement with no middle case such as the standard's own 2-MAD example. Blocking clause: measuring the difference between the centers by expressing it as a multiple of a measure of variability (requires introduction of mean absolute deviation) |
| 7.SP.4 | FULL two-auditor sweep 03 Oct 2026 |
statisticsSamplingQ L6 (comparison form) Two means compared, spread stated qualitatively; conclusion about typical performance. ▶ Play a live item from this skill |
Statistics L6 compares two random samples using a center and a numeric spread printed for both (mean with MAD, or median with IQR), and asks whether one population generally has higher values. The gap between the centers (4, 6 or 8) never decides it; the printed spread does. A gap more than twice the spread keys a real difference, a spread twice the gap keys heavy overlap, and 'the bigger center settles it' is a live wrong answer. Another form prints both medians and both ranges: the medians are equal in half the deals and far apart in the other half, and the options cross a center claim (a typical value is about the same, or higher for one group) with a spread claim (which group varies far more), so the center is read off the medians and the spread off the ranges; a wrong option naming the narrow group as the more varied is on screen in every deal. A third holds the means equal and compares modes and ranges, crossing the mode's side of the mean with the group. In those two forms two of the three wrong combinations are dealt, each named in the explanation. The standard itself says students use the MAD rather than calculate it. There are five scenario families, and every full round carries at least two comparative items. Limit: the two spread outcomes sit at the extremes, with no near-threshold case. |
| 7.SP.5 | FULL adversarial: NOTE-ONLY · re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
probabilityG7Q L4 P(color) = favorable/total from a marble bag; the a/b part-to-part distractor is explicitly named as not-a-probability in the why. ▶ Play a live item from this skill |
Expressing probability as a 0-to-1 fraction is directly exercised (L4 marble bag, P(color) = favorable/total), with the part-to-part ratio a/b as a distractor. The likelihood scale is assessed by one of the three equally weighted L4 forms: given a probability, the child picks unlikely / likely / as likely as not. Because that form has only five prompts and a session does not repeat a prompt, it measured about 3 items in 10 in single six-item sessions and about 2 in 10 across back-to-back sessions. Weakness an adversary will name: that form draws from only five fixed probabilities (1/8, 1/10, 1/2, 7/8, 9/10), always handed over as a fraction, and it is separate from the computing items — in Chance & Probability no item asks the child to interpret the size of a probability they have just computed. |
| 7.SP.6 | PARTIAL two-auditor sweep 03 Oct 2026 |
probabilityG7Q L4-L5 Expected count = P × trials ('about' emphasized); experimental probability heads/flips vs the theoretical 1/2 foil. ▶ Play a live item: level 4 · level 5 |
Two parts are practised at probability L4-L5. The child predicts a count from a probability ('P(red) = 2/7. If you draw 42 times (replacing each time), about how many red marbles do you expect?'). The child also reads an experimental probability from given results ('You flip a coin 50 times and get 18 heads…'), with the theoretical 1/2 as a wrong choice. No item has the child collect data: the numbers of flips and heads are always given in the question. The idea that experimental results move toward the theoretical probability as the number of trials grows appears only in the explanation and is never asked. Held PARTIAL. Blocking clause: Collect data from a chance process (probability experiment) |
| 7.SP.7 | PARTIAL two-auditor sweep 03 Oct 2026 |
probabilityG7Q L5 Experimental vs theoretical coin probabilities; die model used for P(> k). ▶ Play a live item from this skill |
Models are used to find probabilities: a fair die ('What is P(rolling more than 2)?'), marble bags, and a coin with a die. Every model is given, and the child never builds one. The only item with observed data asks for the experimental probability of heads from given flips, with the theoretical 1/2 as a wrong choice. No item asks the child to compare the two values, judge whether they agree, or explain a gap; the difference is described only in the explanation. Held PARTIAL. Blocking clause: Develop a probability model and use it to find probabilities of events |
| 7.SP.7a | PARTIAL two-auditor sweep 03 Oct 2026 |
probabilityG7Q L5 (die form) P(rolling more than k) on a fair die; the k/6 wrong-side distractor; why enumerates favorable outcomes. ▶ Play a live item from this skill |
The uniform model is USED correctly (count favorable over 6 equally likely). Assigning the probabilities — building the model — is given, never asked. The use half is genuinely strong. Blocking clause: Develop a uniform probability model by assigning equal probability to all outcomes |
| 7.SP.7b | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
probabilityG7Q L5 (coin form) Flip n times, get h heads, state the experimental probability h/n; the theoretical 1/2 is the key distractor. ▶ Play a live item from this skill |
The only item that starts from observed data asks for the experimental probability of heads from a given number of flips and heads ('You flip a coin 50 times and get 18 heads…'), with the theoretical 1/2 as a wrong choice. It estimates one event's probability for a near-even coin. No item builds a model across all outcomes, and no non-uniform process (such as the standard's spinning penny or tossed cup) appears. Whether the outcomes look equally likely is never asked. Held PARTIAL. Blocking clause: Develop a probability model (which may not be uniform) by observing frequencies in data generated from a chance process |
| 7.SP.8 | PARTIAL adversarial: DOWNGRADE · re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
probabilityG7Q L6 P(two heads) = 1/4 with the 1/2 and equally-likely-counts 1/3 foils; spinner × die sample-space size m×n. ▶ Play a live item from this skill |
Two probability L6 forms ask for a compound-event probability. One uses a coin and a 6-, 8- or 10-sided die ('What is P(heads AND rolling a 5)?' = 1/12). The other is the fixed two-coin item ('What is P(two heads)?' = 1/4); because it is a single item, it is only about 5% of L6 questions across back-to-back rounds. In both, the event is a single outcome and the sample space appears only in the explanation. A third form shows the coin-and-die sample space as a 2 × 6 table, and the child taps every outcome in an event such as 'heads AND a prime number'; it does not ask for the probability. A fourth form asks only how many outcomes a spinner-and-die sample space has. No item uses a list, a tree diagram or a simulation to find a compound probability; simulation appears only as an L5 choice of which device models a stated simple chance. Held PARTIAL. Blocking clause: using organized lists, tables, tree diagrams, and simulation |
| 7.SP.8a | FULL adversarial: NOTE-ONLY · re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
probabilityG7Q L6 (two-coin item) Sample space HH/HT/TH/TT enumerated in the why; answer is the fraction of outcomes composing the event. ▶ Play a live item from this skill |
Two L6 forms ask for a compound-event probability as a fraction of the sample space. One uses a coin and a 6-, 8- or 10-sided die ('P(heads AND rolling a 5)' = 1/12). Its wrong choices are the forget-the-coin and forget-the-die slips, and in about half of draws one of them is swapped for a near-miss fraction such as 1/13. The other is the fixed two-coin item ('P(two heads)' = 1/4). Its wrong choices are 1/2 and the equally-likely-counts 1/3, and the 1/3 is swapped for a near-miss fraction in about half of draws. A third form shows the 2 × 6 coin-and-die sample space as a table, and the child taps every outcome in an event such as 'heads AND an even number' (2 or 3 cells). The probability for that event (for example 3/12) appears only in the explanation. Weakness: every probability the child is asked for is a single outcome of the sample space. |
| 7.SP.8b | PARTIAL two-auditor sweep 03 Oct 2026 |
probabilityG7Q L6 (spinner-die form) How many outcomes are in the sample space of (color, number) pairs? m+n and n foils. ▶ Play a live item from this skill |
Since 26 Sep 2026 the coin-and-die sample space appears on screen as a 2 × 6 table. The child taps every outcome in an event described in words: 'heads AND an even number', 'tails AND a number greater than 4' or 'heads AND a prime number'. This is about 14-19% of probability L6 questions, in English and Spanish, and covers identifying the outcomes that make up an event. Held PARTIAL because the table is drawn for the child: no item has the child build an organized list, table or tree diagram. Only three events and one coin-and-die setting appear, and another L6 form only counts a spinner-and-die sample space by multiplying. Blocking clause: Represent sample spaces for compound events using methods such as organized lists, tables and tree diagrams |
| 7.SP.8c | NOT COVERED re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— | Design-and-use-a-simulation is a conduct/simulate verb; nothing in the app runs or interprets a simulation. Known wall. |
| Code | Verdict | Evidence (generator · forms) | Notes |
|---|---|---|---|
| 8.NS.1 | PARTIAL two-auditor sweep 03 Oct 2026 |
irrationalsQ L1 Which number is irrational (√non-perfect vs √perfect vs terminating/repeating decimals and fractions); why is √perfect rational (with the every-root-is-rational foil). ▶ Play a live item from this skill |
Irrational-versus-rational (including the √perfect-square case) is well exercised, and irrationals L1 now also serves: converting pure repeats (0.444…, 0.3636…) AND delayed repeats (0.1666…, 0.8333… — the word 'eventually') to fractions, with every option shown in lowest terms and no size-rank pattern; the most digits a repeating block can have (the remainder count); and reading a long-division remainder trail to say which block repeats, where the lead digits that sit outside the block are a live wrong answer. Every full round at L1 carries at least two of these. Held PARTIAL because 'show that the decimal expansion repeats' is met by reading a division already run and selecting the block, not by the child producing the argument. Blocking clause: convert a decimal expansion which repeats eventually into a rational number |
| 8.NS.2 | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
irrationalsQ L2-L3 Between which consecutive integers does √n sit (boxing with perfect squares); which integer is √n closer to (midpoint-squared test in the why). ▶ Play a live item: level 2 · level 3 |
Rational approximation of square roots is the whole of irrationals L2-L3, and the method (perfect-square boxing, then the .5² refinement) is the standard's own example: "Between which two consecutive integers does √11 sit?" and "Which integer is √11 CLOSER to?". English and Spanish. Weaknesses: on the between-which-integers form the two wrong pairs are the neighbouring pairs on either side of the right one, with how many sit below it dealt at random, so the right pair is the middle one only by chance (for √2 and √3 nothing sits below it); comparisons are only root-vs-integer, never irrational-vs-irrational; no expression is estimated (π² style) and π never appears as an irrational; and nothing is located on a drawn number line. Blocking clause: compare the size of irrational numbers, locate them approximately on a number line diagram, and estimate the value of expressions (e.g. π²) |
| 8.EE.1 | PARTIAL two-auditor sweep 03 Oct 2026 |
rootsSciNotationQ (expressions L6); irrationalsQ L1 (adjacent) Evaluate √p (perfect squares 3² to 25²) and ∛p (perfect cubes); solve x² = p for a square's side and x³ = p for a cube's edge, typed in live rounds; write a non-perfect solution with the root symbol (√50 or 5√2, ∛30). Also √perfect evaluated inside the irrationals L1 rational-or-irrational sort. ▶ Play a live item from this skill |
KS's rewritten 8.EE.1 (roots, not exponent laws). Roots are served at expressions Level 6 in about 15% of questions. Square roots of perfect squares (√9 to √625) and cube roots of 8, 27, 64 and 125 are typed on the number pad in live rounds. Solutions of x² = p for a square's side and x³ = p for a cube's edge are typed in about half of live deals, multiple choice ('x = r') otherwise. √2-class irrationality is the target of the irrationals Level 1 sort ('Which of these numbers is IRRATIONAL?'). Writing a non-perfect solution with a root symbol is typed too (x² = 50 answered √50 or 5√2, x³ = 30 answered ∛30), but rarely: about 1 in 100 Level 6 questions for √ and 1 in 200 for ∛ (measured 03 Oct 2026). Every p is a whole number; none is a fraction or decimal. The expressionsG8Q product-of-powers form is NOT credited here or anywhere in KS (MO-only; KS deleted CCSS 8.EE.1). Blocking clause: cube root symbols to represent solutions to equations of the form x³ = p ... cube roots of whole number perfect cubes |
| 8.EE.2 | PARTIAL two-auditor sweep 03 Oct 2026 |
expressionsG8Q kinds 0-1 Single digit × 10^k written from and expanded to standard form. ▶ Play a live item from this skill |
The single-digit-times-power-of-10 form runs both directions, and a GENERATED how-many-times family now serves with unequal front numbers (the standard's own 7 × 10^9 vs 3 × 10^8 shape), including very small quantities with negative exponents and cases where the front quotient is below 1 and has to be regrouped, across six plausible contexts. Held PARTIAL because every quantity is handed over already in scientific notation: no item asks the child to ESTIMATE a messy quantity as a single digit times a power of ten. Blocking clause: express how many times as much one is than the other |
| 8.EE.3 | PARTIAL two-auditor sweep 03 Oct 2026 |
expressionsG8Q kinds 0-1 Write d×10^k in scientific notation (format-rule distractor d0×10^(k−1)); convert d×10^k to a decimal numeral (exponent-counts-places why). ▶ Play a live item from this skill |
Reading is practised for large and small numbers. The child expands scientific notation to a numeral (3 × 10^5, 1.5 × 10^8 km) and reads a calculator's E-notation with either sign (E5 and E−5; exponents 3 to 5). A choose-the-sensible-unit form covers seven contexts (two where the printed unit is the right one), and six contexts ask for the measurement re-expressed in the chosen unit. Held PARTIAL because the child never writes scientific notation: 'Write 4,500,000 in scientific notation' and 'Write 0.000077 in scientific notation' are choose-one-of-three. Also, the re-express family is six fixed scenarios, and the child never chooses a rate unit such as millimeters per year. The one rate shown, a river's 4 × 10^7 liters a day, is re-expressed as '40 megaliters' with the 'a day' dropped. Blocking clause: choose units of appropriate size for measurements of very large or very small quantities |
| 8.EE.4 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
ratiosSaplingQ L8 Constant of proportionality from a table; proportionality test on pairs. ▶ Play a live item from this skill |
Graphing is practised: Functions L5 has the child plot 'the proportional line: (0, 0), (1, 3), (2, 6), (3, 9)' from printed coordinates and then pick what (1, 3) tells you (the unit rate, in one of three wordings such as 'the unit rate: y grows by 3 when x grows by 1'). The statement that this rate is the slope is in the explanation, not the answer, and the slope-as-unit-rate explain prompt runs only with Writing Coach on. The compare clause has no item. Functions L2 compares a table that passes through (0, 0) with y = mx + b where b is at least 1, and Functions L5 races a plotted line against y = mx + b with b at least 1; each pair includes a non-proportional function. The Ratios L6 better-buy compares two unit prices, and either package can be the cheaper one. PARTIAL on the compare clause. Blocking clause: Graph proportional relationships, interpreting its unit rate as the slope (m) of the graph |
| 8.EE.5 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
functionsQ L2 Rate of change computed from two table points (Δy/Δx in the why). ▶ Play a live item from this skill |
At functions L2 the explanation works out Function B's rate as a difference of y's over a difference of x's from two table rows. The answer choices print both rates, though, so the child never works it out. The similar-triangles idea appears by recognition only. At L5 the child plots (0, 0), (2, 2m), (4, 4m), compares rise÷run across the two segments and picks 'the slope is the same between any two points — the triangles are similar' (one of three wordings; since 10 Oct 2026 the two wrong options share their own vocabulary — 'the slope changes as x grows — the farther triangles are steeper', 'the slope depends on the points you pick' — so the key is no longer the option the other two most resemble); the item's explanation states the argument. Elsewhere at L5 the slope is read as the rise per step of 1 on points the child plots. At equations L6 the child types the proportional equation d = kt from a table (about one deal in ten). At ratios L8 the child picks y = kx for a shown proportional graph, against a non-proportional slip (y = x + k or y = kx + k) and a slope slip that since 10 Oct 2026 is dealt on either side of k — y = (k + 1)x or (k + 2)x, y = (k − 1)x, or y = x/k — so the smaller multiplier no longer marks the key. No Grade 6-8 item asks for the y = mx + b equation of a line. With Writing Coach on, the child also explains in writing why the slope of y = 3x is the same between two pairs of points. Without it, that clause is served by selection only. Blocking clause: Use similar triangles to explain why the slope (m) is the same between any two distinct points ... Generate the equation y = mx ... and the equation y = mx + b |
| 8.EE.6 | PARTIAL two-auditor sweep 03 Oct 2026 |
functionsQ L3-L4; ratiosSaplingQ L8 Linear-but-misses-(0,0) distractor rejected; b as the starting value in y = mx + b; proportional-does-not-need-(1,1) why. ▶ Play a live item: level 3 · level 4 |
The ingredients each appear in answer choices or explanations: a constant rate for every linear table, the origin required only for proportional relationships, and b as the starting value. Only one item asks about the y = mx ↔ y = mx + b relationship itself. With Writing Coach on (a paid plan or a live school sponsorship, with the district switch on), the child describes how y = 3x and y = 3x + 5 are alike and different and which one is proportional. Without Writing Coach nothing asks it, and the 'vertical translation' framing appears nowhere. No row whose clause is met only through Writing Coach is marked FULL. Blocking clause: Describe the relationship between the proportional relationship expressed in y = mx and the non-proportional linear relationship y = mx + b as a result of a vertical translation |
| 8.EE.7 | FULL two-auditor sweep 03 Oct 2026 |
equationsG8Q kind 0; equationsSaplingQ L5/L7; inequalitiesG7Q L4 One-step (all four operations), two-step px + q = r, variables both sides ax + b = cx + d, and two-step inequalities including the divide-by-a-negative flip. ▶ Play a live item from this skill |
One-step, two-step, both-sides, equations AND inequalities, with the same variable on both sides — the enumeration is met, and since 20 Sep 2026 equations that require distribution and decimal or unit-fraction coefficients serve as well. Weakness: 'fluently' is asserted, not measured. |
| 8.EE.7a | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
equationsG8Q kinds 1-2 ax + b = ax + d classified as no solution (b ≠ d) or infinitely many (b = d), with the opposite case and x = d as foils. ▶ Play a live item from this skill |
Equations Level 8 shows ax + b = ax + d and the child picks 'no solution' or 'infinitely many solutions', about one Level 8 question in four, in English and Spanish; the explanation subtracts ax and judges what is left. 'Exactly one solution' is offered but never right, and the equation shape never changes. The child never gives an example of each kind or shows the transforming steps. PARTIAL. Blocking clause: Give examples of linear equations ... Show which of these possibilities is the case by successively transforming the given equation |
| 8.EE.7b | FULL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
equationsG8Q kind 0; equationsSaplingQ L7 Both-sides equations solved by collecting like terms; two-step equations. ▶ Play a live item from this skill |
Equations Level 8 serves equations with rational coefficients (0.5x + 3 = 7, 1.5x + 2 = 11, (1/4)(x + 2) = 3) and equations that need distributing and collecting like terms (3(x − 2) + 2x = 19, also typed; 0.25(x − 4) + x = 9). Inequalities Level 4 serves −3(x + 2) ≥ 6 (the sign flips), 2(x − 3) + x > 9 and 0.5(x + 4) < 6, with answers in all four directions; no pattern in the choices gives the answer away. Every full round at both levels carries at least two, in English and Spanish. Most solving is by choosing from three. FULL. |
| 8.F.1 | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
functionsQ L1 Given three (input, output) pairs, is it a function? Both polarity variants; the legal two-inputs-share-an-output case is the key distractor. ▶ Play a live item from this skill |
At functions L1 the item lists three (input, output) pairs and asks 'Is this a function?'. Both cases generate. The options are a yes ('each input has one output', 'no input has two outputs' or 'every input gives exactly one output'), a no for one input with two outputs, a no for two inputs sharing an output, and 'three pairs are too few to tell', each in one of three wordings of different lengths; the yes reads the same whether or not it is the key, and no option names an input, so neither length nor a number marks the right choice. The child never explains the definition: the item is a choice, and its explanation states the rule. The question is posed from three-pair lists only, never from a graph, a mapping diagram or a verbal rule. At L5 the child builds graphs from listed ordered pairs, but none of those items asks whether the relation is a function. The functions skill uses no f(x) notation, which the KS text says is not required in Grade 8. Blocking clause: Explain that a function is a rule that assigns to each input exactly one output |
| 8.F.2 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
functionsQ L2 y = m1x + b1 (equation) vs a table through (0,0) — which grows faster? Rate read off the table; starts-higher foil separates rate from intercept. ▶ Play a live item from this skill |
Two templates compare two functions given in different representations, and growth rate is the only property compared. At functions L2 an equation is compared with a table. At L5 the child plots Function A's points and compares them with Function B's equation. Either function can be the faster one in both. The answer choices state each function's rate, though ('A — its rate is 6 per step', 'B — its rate is 3 per step'; at L5 'B — it climbs 3 per step; A only 2', where the wrong choice contradicts itself). So the child never has to read a rate from the table, graph or equation, only to know that 'grows faster' means the larger rate, not the higher start. The greater y-intercept and the intersection point are never compared, and verbal descriptions never serve as a representation. Blocking clause: Compare properties of two linear functions represented in a variety of ways (algebraically, graphically, numerically in tables, or by verbal descriptions) |
| 8.F.3 | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
functionsQ L3 Table of (x, y): linear (constant +m) vs nonlinear (x² + b, growing jumps); misses-(0,0) and only-a-graph-can-decide foils. ▶ Play a live item from this skill |
At functions L3 the child decides whether a three-row table is linear (a constant step of +m) or not (x² + b, growing jumps). The wrong options are 'it misses (0, 0)' when the table is linear, 'the values always rise' when it is not, and 'only a graph can decide' in both. The right verdict can be read from which choices are offered: 'yes — it climbs m each step' appears only when the table is linear, and 'no — the per-step change keeps growing' only when it is not. About one L3 item in ten shows a drawn graph and asks 'Is the function LINEAR or NONLINEAR?'. The graph is either the line from (0, 0) to (4, 8) or the curve through (0, 0), (1, 1), (2, 4), (3, 9). The nonlinear case is always x² (+ b), the standard's own A = s² example, from a table or that one curve. The child never gives an example of a nonlinear function. No item asks whether an equation y = mx + b is linear or what its graph looks like. Blocking clause: give examples of functions that are not linear |
| 8.F.4 | PARTIAL two-auditor sweep 03 Oct 2026 |
functionsQ L2/L4 Rate of change from a table; b interpreted as the starting cost in a rental context (with per-hour and max-hours foils). ▶ Play a live item: level 2 · level 4 |
The only typed constructions are two Grade 6 items. At expressions L3 (about one deal in four) the child writes the cost of b tickets from a per-ticket price and a flat shipping fee, for example 5b + 3; the rate and the starting value come from the description. At equations L6 (about one deal in ten) the child writes d = kt from a table, a proportional rule. From a graph, at functions L5, the child plots listed points and picks the rise per step and the y-axis crossing from three numbers. The table comparison at functions L2 prints both rates in its answer choices and is not counted. At functions L4 the child picks what either number of a kayak-rental equation y = mx + b means in context: the cost of each hour or the starting cost. In about one kayak item in ten the two numbers are equal (for example y = 10x + 10), so both meanings are right but only one is accepted; this is a defect. The ticket-cost expression is the only non-proportional rule the child writes (the skate-rental item at equations L7 has the child type an equation such as 3h + 5 = 23, which is solved for one unknown, not a function rule). From a graph the child only picks a proportional equation (y = kx at ratios L8, d = kt at equations L6); no item has the child write a rule from two points or from a graph. Blocking clause: Construct a function to model a linear relationship between two quantities |
| 8.F.5 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Generator: functionsQ L3 (qualitativeGraphQ). Forms: a rendered line graph with a story match, the flat (standing-still) interval, and linear vs nonlinear. Note: at functions L3, about one item in three shows a drawn line graph. The child picks the story a distance-from-home graph tells, the interval where the walker stood still, or whether the graph is linear or nonlinear. The child analyzes the graph but never describes it in words, because every answer is a choice. No item asks the child to sketch a graph from a verbal description. Blocking clause: Describe qualitatively the functional relationship ... Sketch a graph that exhibits the qualitative features of a function that has been described verbally |
| 8.G.1 | PARTIAL two-auditor sweep 03 Oct 2026 |
geometrySaplingQ L9; geometryG8Q; arcsQ Angle measures used fluently across supplements, vertical pairs, triangle sums, transversals, and arc fractions. ▶ Play a live item from this skill |
Degree measures are used in the angle computations at geometry L9-L11 (supplementary and vertical angles, triangle angle sums, transversals, exterior angles, arcs and sectors), and one geometry L10 item states the degree definition ('An angle turns through 72 one-degree angles. What is its measure?'), but its answer is always the number printed in the question, so it does not show that the child knows the definition. No Grade 6-8 item asks about an angle as two rays sharing an endpoint. Stem graded on its sub-codes: 1a PARTIAL, 1b NOT COVERED. Blocking clause: understand concepts of angle measurement |
| 8.G.1a | PARTIAL two-auditor sweep 03 Oct 2026 |
arcsQ L11 θ/360 as the fraction of the circle, driving arc length and sector area; whys name 90° = a quarter, 120° = a third, 180° = half. ▶ Play a live item from this skill |
arcsQ (geometry L11) is built on the angle-as-fraction-of-a-circle idea: a 90°, 120° or 180° angle is treated as θ/360 of the circle to get an arc length or a sector area. Within arcsQ the fraction reasoning always serves an arc or sector computation, and no figure is drawn, so no angle is shown against a circle there. The clause 'an angle that turns through 1/360 of a circle is called a one-degree angle' is not the question in any arcsQ form. The only item in the app that uses the term one-degree angle is in a different generator: geometry L10 (g8GeomKnowledgeQ) asks "An angle turns through 45 one-degree angles. What is its measure?", which asks the n-one-degree-angles clause (8.G.1b) and states the 360 split only in its explanation. Blocking clause: An angle that turns through 1/360 of a circle is called a "one-degree angle" |
| 8.G.1b | NOT COVERED re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— | Geometry L10 (g8GeomKnowledgeQ) asks 'An angle turns through 45 one-degree angles. What is its measure?' in English and Spanish, about one geometry L10 question in fifteen, for n = 30, 45, 60, 72 or 110. The answer is always n° — the number printed in the question — and it is the only option that repeats that number (the wrong options are built from 360 − n and n ÷ 2, one of them sometimes swapped for a nearby number such as 22° or 68°), so choosing the number you see always scores. The item therefore does not show the child applying the definition. No angle is drawn, and no other item in the app uses the term one-degree angle. Blocking clause: An angle that turns through n one-degree angles is said to have an angle measure of n degrees |
| 8.G.2 | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | No item at the Grade 6-8 levels has the child measure an angle with a protractor or draw an angle of a given measure. The app does have an on-screen protractor: at geometry levels 3-5 (the grade 3-5 levels) four fixed items have the child turn a ray to draw a 60° or 120° angle, or read a 45° or 150° angle off the protractor from three choices. Geometry's Sapling levels (6-11), where a Grade 8 student works, never deal them, so the standard is not covered at this grade. |
| 8.G.3 | PARTIAL two-auditor sweep 03 Oct 2026 |
geometrySaplingQ L9 form 1; geometryG8Q kind 1 Supplementary partner (180 − a, with the complementary-confusion foil); third angle of a triangle (180 − a − b, with the quadrilateral-360 foil). ▶ Play a live item from this skill |
Unknown-angle arithmetic is served as text: geometry L9 asks for a supplementary partner ('Two angles are supplementary. One measures 65°. What is the other?') and geometry L10 for a triangle's third angle ('Two angles of a triangle measure 45° and 35°. What is the third?'). In live rounds geometry L9 also has the child type the equation with a symbol for the unknown angle ('… Write an equation for the other angle m.' → m + 65 = 180) and solve it ('Solve m + 65 = 180 for the other angle. Write m = …'), for supplementary and complementary pairs. Not practised: no diagram accompanies any unknown-angle problem, no item splits a shown angle into parts that add to the whole, and none has a real-world setting. The app's drawn angles (the angle figure and the protractor items) sit at geometry L5 and below and carry no unknown-angle problem. Blocking clause: find unknown angles on a diagram |
| 8.G.4 | PARTIAL two-auditor sweep 03 Oct 2026 |
geometrySaplingQ L9 form 1 Vertical angles equal (supplement as foil); supplementary pairs (complement as foil). ▶ Play a live item from this skill |
26 Sep 2026 (round 2): the child types and solves m + a = 90 or 180 for supplementary and complementary angles; still PARTIAL because vertical and adjacent angles never appear in an equation, and there is no multi-step problem and no figure. Blocking clause: in a multi-step problem to write and use them to solve simple equations for an unknown angle in a figure |
| 8.G.5 | PARTIAL two-auditor sweep 03 Oct 2026 |
geometryG8Q kinds 1-2 Triangle angle sum (180 − a − b); corresponding and alternate-interior angles equal under parallel lines, supplement foil. ▶ Play a live item from this skill |
Geometry L10 computes with two of the facts: a triangle's third angle from the angle sum ('Two angles of a triangle measure 45° and 35°. What is the third?') and the exterior angle ('Two interior angles of a triangle measure 55° and 40°. What is the EXTERIOR angle at the third vertex?'). The transversal item ('A transversal crosses two PARALLEL lines. One angle measures 75°. What is its CORRESPONDING angle?') only ever asks for an angle equal to the one given, so its answer is always the number in the question. No item asks for an informal argument — the three-copies-form-a-line arrangement, or transversal reasoning as proof — and the angle-angle criterion for similar triangles does not appear in the Grade 6-8 content. Blocking clause: Use informal arguments to establish facts |
| 8.G.6 | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | No item has the child draw a triangle or any other shape from given side lengths or angle measures, and no item asks whether three given measures make one triangle, more than one, or none. The app takes drawings elsewhere (a freehand shape-drawing tool in the Pre-K-2 Shapes skill, the protractor at geometry levels 3-5, points plotted on a grid), but none is used for this at the Grade 6-8 levels; the nearest Grade 6-8 item has the child plot the four given vertices of a rectangle at coordinate-grid level 5. |
| 8.G.7 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
The theorem is used in geometryG8Q at geometry L10 (missing hypotenuse and missing leg), and g8GeomKnowledgeQ at the same level adds a proof item and a converse item. Proof: "In the squares proof of the Pythagorean Theorem, which comparison finishes the argument?" (the area of the square on the hypotenuse equals the sum of the areas on the legs - against the hypotenuse's length equalling the two legs' lengths added, and the hypotenuse square equalling only the longer leg's square). Converse: "A triangle has sides 8, 15, and 17. Use the CONVERSE of the Pythagorean Theorem: is it a right triangle?" - half the triangles are true triples and half have one side off by one, and the options are yes (the squares match), no (they do not) and "it cannot be decided from the sides alone". English and Spanish. Limits: 'explain a proof' is served by choosing one of three statements - the child never explains or orders the steps of a proof, and there is one fixed proof question with no figure; the converse item draws on seven base triples; and each of the two forms is about one geometry L10 question in thirty. Blocking clause: Explain a proof of the Pythagorean Theorem and its converse |
| 8.G.8 | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
geometryG8Q kind 0 Find the hypotenuse from two legs, or a leg from hypotenuse + leg, over a 7-triple bank; legs-sum foil ('the walk around the corner') and near-miss foil; hypotenuse-is-longest check in the why. ▶ Play a live item from this skill |
Geometry L10 asks for an unknown side of a right triangle in both directions (the hypotenuse from two legs; a leg from the hypotenuse and the other leg). The legs' sum a + b is a wrong option in every find-a-leg item and in about two of three find-the-hypotenuse items. Limits: the sides are always one of seven integer Pythagorean triples, so no answer is irrational; apart from a coordinate-distance item that uses the first four of those triples, every find-a-side problem is a bare triangle described in words, and none has a real-world setting; and no item applies the theorem in three dimensions — a slant height appears only as a given number in the pyramid surface-area item ('A square pyramid has base side 4 and slant height 7…'), never as a length to find. Blocking clause: in real-world ... problems in two and three dimensions |
| 8.G.9 | FULL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Geometry L10 (g8GeomKnowledgeQ) asks for the distance between two points: "What is the distance between the points (1, 2) and (4, 6)?" (5), explained as "the changes are 3 in x and 4 in y - legs of a right triangle whose hypotenuse joins the points: √(3² + 4²) = √25 = 5", with the L-shaped walk 3 + 4 = 7 offered as a wrong answer. English and Spanish. As built: the two points always differ by one of four Pythagorean triples (3-4-5, 6-8-10, 5-12-13, 9-12-15), so every distance is a whole number; all coordinates are positive and the second point is always up and to the right of the first; no grid is drawn; the answer is chosen from three options; and the form is about one geometry L10 question in fifteen. (transformationsQ at coordinate-grid L6 moves coordinate points and does not ask a distance.) |
| 8.G.10 | PARTIAL two-auditor sweep 03 Oct 2026 |
arcsQ L11; volumeG8Q L6 Arc length (θ/360 × 2πr) and sector area (θ/360 × πr²), π-exact over a 9-pair bank; cone volume (third of the cylinder) and sphere volume (4/3 πr³). ▶ Play a live item from this skill |
Arc length and sector area are computed at geometry L11 (arcsQ), and cone and sphere volume at volume L6. Pyramids appear: volume L6 asks the volume of a rectangular-base pyramid (one third of its prism) and the cross-sections of a square pyramid, and geometry L10 asks the surface area of a square pyramid from its base side and slant height. Not served: the surface area of a cone or a sphere is not asked anywhere in the app (in the volume L6 sphere item, sphere surface area appears only in the explanation and, at radius 6, as a wrong option), and the pyramid surface-area item is a single square-base form with the formula SA = B + ½Pl printed in the prompt. Blocking clause: surface areas and volumes of pyramids, cones, and spheres |
| 8.G.11 | PARTIAL two-auditor sweep 03 Oct 2026 |
volumeG8Q L6 Cone volume via the one-third-of-the-enclosing-cylinder relationship; sphere volume. ▶ Play a live item from this skill |
Stem code; graded on its sub-codes. The cone-cylinder relationship is at least computed with (the distractor set includes the full cylinder); the investigate layer is why-carried. 11a PARTIAL, 11b PARTIAL. Blocking clause: Investigate the relationship between the formulas of three dimensional geometric shapes |
| 8.G.11a | PARTIAL two-auditor sweep 03 Oct 2026 |
volumeG8Q kind 1 Cone volume with the enclosing-cylinder volume and 2/3-of-it as foils; why tells the three-cones-of-water experiment. ▶ Play a live item from this skill |
The 1/3 relationship is exercised for cones — choosing cyl/3 over the full cylinder is the item — and, in a separate item of the same generator, for rectangular-base pyramids, where the child chooses prism/3 over the full prism and half of it. Each item computes one solid. The generalization (the same V = 1/3 Bh across both families) is carried by the why — the pyramid rationale calls it 'the same relationship the cone has with its cylinder, now with corners' — not asked. Blocking clause: Generalize the volume formula for pyramids and cones (V = 1/3 Bh) |
| 8.G.11b | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Geometry L10 (g8GeomKnowledgeQ) serves pyramid surface area with a slant height: "A square pyramid has base side 8 and slant height 11. What is its surface area? (SA = B + ½Pl)" (240). The explanation works the base, the perimeter and the lateral part (four triangles of ½ × side × slant height) and separates slant height from vertical height. English and Spanish. Limits: square pyramids only (base side 4, 6 or 8) - no surface-area item for a cone exists in the Grade 6-8 content, and no other pyramid base; the formula is printed in the prompt; no figure or net is drawn; the answer is chosen from three options; and the form is about one geometry L10 question in fifteen (one of five forms in a family that is about a third of that level). Blocking clause: Generalize the formula; cones |
| 8.G.12 | PARTIAL two-auditor sweep 03 Oct 2026 |
arcsQ L11; volumeG8Q L6 Arc-length, sector-area, cone-volume, and sphere-volume problems, all π-exact. ▶ Play a live item from this skill |
Arc length and sector area are computed at geometry L11 (π-exact), the volumes of a cone, a sphere and a rectangular-base pyramid at volume L6, and the surface area of a square pyramid at geometry L10 (from base side and slant height, with SA = B + ½Pl printed in the prompt). Not practised: the surface area of a cone or a sphere is never asked, and every one of these items is a bare shape with no real-world setting. Blocking clause: surface area of three-dimensional objects including pyramids, cones and spheres |
| 8.SP.1 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Statistics L8 serves scatter plots on the tap grid in English and Spanish. To construct, the child plots five listed (x, y) pairs, scored on the exact set of points, then reads their own plot's association; this is about two L8 items in five. To interpret, shown plots ask for positive, negative or no association. Shown plots with a straight model line ask what its slope says, or whether a candidate line fits the trend. The row is held PARTIAL. The point clouds are only linear rising, linear falling or patternless, so clustering, outliers and nonlinear association never appear. The patterns are picked from three descriptions and never described by the child. Blocking clause: Describe patterns such as clustering, outliers, positive or negative association, linear association, and nonlinear association |
| 8.SP.2 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Statistics L8 (scatterQ) shows a scatter plot - eight points on a grid with axes from 0 to 9 - and one of its forms draws a candidate line over it: "Scatter plot with a candidate line: practice minutes (x) vs. free throws made (y)." followed by "Does the orange line fit the cloud's trend?" (three options, the same every time: yes - the line runs the same way the points do; no - the line runs against the way the points go; no - a good line has to pass through every point). A second form draws a model line and asks what its slope says in context, and a third has the child plot five pairs to construct the scatter plot. English and Spanish. Limits: the child never fits a line - the drawn line is always one of the grid's two diagonals; fit is judged only by direction (the line runs with the trend or against it), never by how close the points lie to the line, which is the criterion the standard names; and the line forms appear only for the contexts with a positive or negative trend. (twoWayTableQ at statistics L7 reads a two-way table and involves no scatter plot or line.) Blocking clause: Informally fit a straight line, and judge fit by the closeness of the points to the line |
| 8.SP.3 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
functionsQ L4 y = rate·x + start in a rental context; the intercept interpreted as the starting cost, rate as the per-hour cost foil. ▶ Play a live item from this skill |
At functions L4 the child picks what either number of a given linear rule means in context (the cost of each hour or the starting cost in a kayak-rental equation) and the point where a savings equation reaches $0. These rules are deterministic, never a fitted model of measured bivariate data. A fitted-model setting exists only at statistics L8 (about one item in eight), where a model line is drawn over a scatter plot and the child picks what its slope says in context ('each extra 1 of hours of sleep brings ABOUT 1 more energy rating'). That line has no equation, its slope is always 1 or −1, and its intercept is never interpreted. In neither place is the model used to solve a prediction problem about data. Blocking clause: in the context of bivariate measurement data |
| Code | Verdict | Evidence (generator · forms) | Notes |
|---|---|---|---|
| RL.6.1 | PARTIAL two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-secondlap, p6-static) Evidence-bound inference items: why Jonah smiles, what Priya's passing 'without looking over' suggests, what the brother's giving up his turn reveals. ▶ Play a live item from this skill |
Inferences drawn from the text, and what it says outright, are genuinely assessed in the stories (why Jonah smiles, what Priya's passing without looking over suggests, what the brother's giving up his turn reveals), and each answer is anchored to a quoted line. In the stories, though, the quoting happens in the explanation after the child answers. Choosing the evidence that supports a claim is asked only on informational passages ('Which piece of evidence SUPPORTS the claim?'). With Writing Coach on (a paid plan or a live school sponsorship), the child writes a paragraph that names a story's theme and defends it with two specific moments. That story is one the child picks, not one read in the app, and a clause met only through Writing Coach is not marked FULL. Blocking clause: Cite textual evidence to support analysis |
| RL.6.2 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-secondlap) Which sentence best states the passage's theme? — theme-as-idea-under-the-plot, with topic-word and false-moral distractors. ▶ Play a live item from this skill |
Determining a theme is directly assessed, on 'The Second Lap' ('Which sentence best states the passage's theme?') and, added on 31 Oct 2026 and not yet audited as evidence, on 'Enough of Each'. The explanation then ties the theme to particular details. A summary is asked by selection: 'Which is the BEST summary of the passage?' on three stories ('Enough of Each', added on 31 Oct 2026 and not yet audited as evidence, is the third), with a theme statement as a wrong option on 'The Second Lap'. Held PARTIAL: 'provide a summary' is a production verb met only by choosing one, and no writing prompt asks for a summary. How the theme is conveyed through details is explained to the child after answering. With Writing Coach on, the child does write how two moments of a chosen story show its theme, but a clause met only through Writing Coach is not marked FULL. Blocking clause: provide a summary of the text distinct from personal opinions or judgments |
| RL.6.3 | PARTIAL two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-secondlap, p6-static) Why does Jonah smile at the end; what the brother's action reveals; why the ending silence is 'exactly enough' (reversal from the opening). ▶ Play a live item from this skill |
How characters respond and change as the plot moves toward resolution is genuinely probed by selection: the narrator's annoyance turning into wonder in 'Static', Jonah beating his urge to quit, and the brother giving up his turn. The 'Which is the BEST summary' items on three stories ('The Second Lap', 'Static' and, added on 31 Oct 2026 and not yet audited as evidence, 'Enough of Each') ask the child to pick the retelling that follows the whole arc (start, turn and ending). Still, no item analyzes how the plot unfolds episode by episode, 'describe' is a production verb that every item meets only by choosing an answer, and the Grade-6 set has no drama. Blocking clause: how a particular story's or drama's plot unfolds in a series of episodes |
| RL.6.4 | PARTIAL adversarial: DOWNGRADE · re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-secondlap item 2, p6-static item 2) 'Rehearsed' most nearly means (context-fixed meaning); the 'like a held breath' simile — what it makes Saturn FEEL like (impact of a word choice on tone). ▶ Play a live item from this skill |
Word-in-context meaning ('rehearsed' most nearly means…) and the effect of a specific figurative choice ('like a held breath' makes Saturn feel…; money 'was shirts') are each directly the question inside literary passages. Connotation is assessed at vocabulary L6 ('Which has the more NEGATIVE connotation…'), but as standalone word pairs with no text attached, while this standard asks for meanings 'as they are used in a text'. Held PARTIAL: the in-text evidence is a few fixed items, and no item isolates the effect of word choice on tone across options. |
| RL.6.5 | FULL two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-secondlap item 5, p6-static item 4) How does the question 'the race was already decided, wasn't it?' work in the story (doubt staged so its defeat means something); why the closing silence is 'exactly enough' (reversal of the noisy opening). ▶ Play a live item from this skill |
Three of the questions ask exactly how a particular sentence fits the whole. In 'The Second Lap', the doubting question 'the race was already decided, wasn't it?' lets the reader hear the voice Jonah must defeat. In 'Static', the closing silence reverses the noisy opening. In 'The Lantern Schedule', the closing line about electric bulbs 'in every room' builds the 1911 setting by contrast. Together they make up about one in seventeen Grade-6 reading questions, in English and Spanish. Weakness: chapter, scene and stanza scales never appear, because the Grade-6 passage set has no drama and no poem. |
| RL.6.6 | PARTIAL two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-static item 1) From whose point of view is the story told, and why does that matter? — first-person skeptical sibling, so the reader feels annoyance turn to wonder. ▶ Play a live item from this skill |
Identifying the narrator's POV and its effect is assessed. HOW the author develops it (first-person interiority, ironic detail selection) lives in the why; explain is a production verb and the item is selection of a why-it-matters, not of a how-it-is-built. One item. Blocking clause: Explain how an author develops the point of view |
| RL.6.7 | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Comparing reading a text with listening to or watching it needs the child to meet both versions and contrast them. Grade-6 passages do carry a 'Read to me' button that reads the passage aloud while each word highlights, so a child can hear the same text. But no item asks the child to compare reading it with hearing it, and there is no video or live performance. The app's listen-first stories sit only in the early comprehension levels, and even there nothing compares reading with listening. This is a multimedia wall, and the verdict is not stretched to cover it. |
| RL.6.8 | NOT COVERED re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— | KS's own statement is 'Not relevant to literature' — a deliberate placeholder (the argument-evaluation slot has no literature analogue). Nothing to cover and nothing claimed; excluded from any coverage percentage in marketing copy. |
| RL.6.9 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Two texts do now sit side by side: 'Two Views of the Storm' presents Passage A (a journal) beside Passage B (a news brief) with cross-text items on tone, purpose, what appears in both, and what only one tells. Held PARTIAL because the pair is a personal journal against an informational report (so is the second pair, a travel journal beside a science-magazine article in 'Two Ways to Hear a Whisper', added on 31 Oct 2026 and not yet audited as evidence) rather than two literary forms — no poem or drama exists in the Grade-6 set — and approaches to a similar THEME are not what is compared. |
| RL.6.10 | PARTIAL two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Grade-6 grammar asks the child to read a sentence through its conventions: which part is extra, nonessential information, judged by the paired commas, dashes or parentheses that fence it off; and why 'When Leo met Sam, he was late' is unclear (a vague pronoun). Vocabulary also asks which transition fits the logic between two sentences. Held PARTIAL: these are standalone sentences shared with the writing-conventions rows, and no item applies conventions while reading a passage, which is what 'when reading to improve comprehension' asks. (The informational twin RI.6.10 should move the same way for consistency.) Blocking clause: When reading to improve comprehension |
| RL.6.11 | PARTIAL two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (word-in-context); vocabularySaplingQ (SAPLING_ROOTS) 'Rehearsed' in context; root/affix meaning and word-decode items. ▶ Play a live item from this skill |
The stem is graded on its sub-codes. (a) context use and (b) word roots are both PARTIAL: context use in stories rests on few fixed items, and the roots bank is shared. (c) and (d) are not covered. Multiple-meaning words appear only in the early vocabulary levels, never in Grade-6 content. Blocking clause: unknown and multiple-meaning words |
| RL.6.11.a | PARTIAL adversarial: DOWNGRADE · re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-secondlap item 2) In this passage, 'rehearsed' most nearly means… — context ('in his head all week', the 'what he had not rehearsed' contrast) fixes the meaning. ▶ Play a live item from this skill |
Using context to find a meaning inside a story is asked twice on audited items. 'Rehearsed' most nearly means… (context: 'in his head all week'), though a sixth grader may already know that word, so context use is not forced. Also, what money 'was shirts' reveals in 'The Lantern Schedule', where only the surrounding detail of collars sewn at the kitchen table gives the meaning. Held PARTIAL: two fixed items, one passable by prior word knowledge, and the other is the same item credited to figurative language. A third, added on 31 Oct 2026 and not yet audited as evidence, asks what Grandpa means when he says the dough 'answers' in 'Enough of Each', where the thumb test in the dough gives the meaning. The row stays at its audited verdict until a refute-first two-auditor re-check counts the new items. |
| RL.6.11.b | PARTIAL adversarial: NOTE-ONLY · two-auditor sweep 03 Oct 2026 |
vocabularySaplingQ (SAPLING_ROOTS, 15 entries × 2 forms) What does the root/prefix/suffix X mean (distractors are other parts' real meanings); using its word parts, 'telescope/portable/predict…' means… ▶ Play a live item from this skill |
The clause is genuinely exercised in isolation — tele/port/dict/aud/bio/spect/chron plus affixes, decode form included, etymologies fact-checked. Capped PARTIAL by house rule: the identical SAPLING_ROOTS bank is the sole credit for BOTH RL.6.11.b and RI.6.11.b (and feeds SL.6.8), and the words are bank-standalone — never a word met in a literary text being read. Blocking clause: Use common, grade-appropriate Greek or Latin affixes and roots to define the meaning of a word |
| RL.6.11.c | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | No Grade-6 item uses reference materials: no dictionary lookup, no pronunciation key, no part-of-speech determination. The app's dictionary questions sit only in early levels: guide words, and reading a glossary entry for its meaning and part of speech, in the early comprehension levels; and a picture-dictionary page, where the child uses the page to find a word, in early Sight Words. None asks for a pronunciation. Interaction-mode wall. |
| RL.6.11.d | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Verifying a preliminary meaning against context or a dictionary is a multi-step checking behavior no item hosts; the closest content (context confirmation of 'rehearsed') performs the check FOR the child in the why. |
| RL.6.12 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-static item 3) The 'held breath' simile item. ▶ Play a live item from this skill |
Stem graded on its sub-codes. (a) figures of speech are interpreted inside passages; (b) word relationships and (c) connotation are now real questions at vocabulary L6 ('What relationship joins "cello" and "instrument"?' with cause/effect, part/whole and item/category options; 'Which has the more NEGATIVE connotation…'). Held PARTIAL because (b) and (c) are standalone vocabulary items in a bank shared across the literature and informational strands, never anchored in a literary text. Blocking clause: word relationships and nuances in word meanings |
| RL.6.12.a | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-static item 3) The comparison 'like a held breath' makes Saturn feel… — still/fragile vs loud/fast and imaginary foils; why names what a simile transfers. ▶ Play a live item from this skill |
Interpreting a figure of speech in its context is squarely the question, with honest misreadings as the wrong options: the simile 'like a held breath' in 'Static' and money that 'was shirts' in the 1911 'Lantern Schedule' story. Held PARTIAL: two audited fixed items, both also credited to RL.6.4. A third, added on 31 Oct 2026 and not yet audited as evidence, asks what Grandpa means when he says the dough 'answers' in 'Enough of Each'. Personification appears in the Grade-6 passages (Saturn 'cooperating' in 'Static', thunder that 'played percussion' in Leo's journal), but no audited question asks the child to interpret it, and no question asks about an idiom. The row stays at its audited verdict until a refute-first two-auditor re-check counts the new items. |
| RL.6.12.b | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED-AFTER-FIX) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Word relationships are now the question at vocabulary L6: 'What relationship joins "cello" and "instrument"?', choosing among cause and effect, part and whole, and item and category. Held PARTIAL because the items are standalone (not anchored in a literary text), sit in the vocabulary bank shared across strands, and the 'to better understand each word' clause is why-carried. |
| RL.6.12.c | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Connotations of words with similar denotations are directly distinguished at vocabulary L6, in English and Spanish ('Both words can describe the same behavior. Which has the more NEGATIVE connotation…'). Held PARTIAL because the same bank is credited to the informational twin and the items are standalone vocabulary, never anchored in a literary text. |
| RL.6.13 | PARTIAL two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-secondlap, p6-static) Two literary passages read by the child (passageSelfRead: true — reading it yourself IS the task), each with 4-5 analysis items. ▶ Play a live item from this skill |
Independent reading of well-made literary prose genuinely happens: four short stories ('The Second Lap', 'Static', 'The Lantern Schedule' and, added on 31 Oct 2026, 'Enough of Each') and two journals, each paired with an informational text, all read by the child with three to six analysis questions. Blocked on complexity and range: the Grade-6 passages run about 100 to 160 words, far below Grade-6 text length, and 'dramas, prose and poetry' is met on one leg only. The Grade-6 set has no drama and no poetry; poems appear only in the K-5 reading levels. Blocking clause: of appropriate quantitative and qualitative complexity for sixth grade |
| RI.6.1 | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-octopus items 2, 5) Which detail supports the claim that the disguise happens fast? (true-but-irrelevant details as foils); what can you infer about why two hearts serve the gills (smallest-step inference). ▶ Play a live item from this skill |
The child picks the evidence for a stated point on the informational passages: the detail that shows the octopus's disguise is fast, plus, on 'The Case for Quiet Hallways', the evidence that supports the author's claim and the claim that has no evidence behind it. Inferences are asked too (why two hearts serve the gills, why the Pony Express existed at all), but the evidence for an inference is quoted in the explanation after the child answers. No item asks the child to pick or give the evidence that supports an inference, so the row is held PARTIAL on that leg. |
| RI.6.2 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-octopus item 1) What is the central idea of the passage? — must cover camouflage + blue blood + three hearts + brief life, with single-detail and overreach foils. ▶ Play a live item from this skill |
Central-idea determination is directly assessed and the why shows how the details converge on it, and a summary is now asked on the informational octopus passage ('Which is the BEST summary of the passage?'). Held PARTIAL because the summary is selected rather than written, and no writing prompt asks for one. Blocking clause: provide a summary of the text distinct from personal opinions or judgments |
| RI.6.3 | PARTIAL two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-octopus item 4) Why does the author most likely mention that the blood is blue? — the next exhibit in the case about strangeness. ▶ Play a live item from this skill |
How an idea is elaborated is asked directly: why the author mentions the octopus's blue blood (to build the case that the animal is deeply unusual) and why the Quiet Hallways author mentions factory shifts (to show the bell is a leftover habit). How an idea is introduced is not asked by any audited item. The district example in the Quiet Hallways argument is picked as evidence, but no item analyzes it as an illustration. Held PARTIAL, and two fixed items are not analysis 'in detail'. Added on 31 Oct 2026 and not yet audited as evidence: 'Books on Horseback' asks why the author mentions the scrapbooks of recipes and quilt patterns, and 'The Color Underneath' asks why the author opens by saying the maples 'seem to repaint themselves'. The row stays at its audited verdict until a refute-first two-auditor re-check counts the new items. Blocking clause: how a key individual, event or idea is introduced, illustrated and elaborated |
| RI.6.4 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-octopus item 3) Based on the passage, chromatophores are… — the definition sits beside the word; the why teaches where technical texts pay for their hard words. ▶ Play a live item from this skill |
Technical meaning in context is assessed ('chromatophores are…'). Figurative meaning in an informational text is assessed too: on the Pony Express passage, the child interprets 'beaten not by faster horses, but by a wire' in context. Held PARTIAL because connotative meaning is asked only through standalone word pairs at vocabulary L6, never about a word as it is used in an informational passage. Blocking clause: including figurative, connotative and technical meanings |
| RI.6.5 | PARTIAL two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
The child is asked what a particular sentence contributes to the development of the ideas: 'Why does the author mention factory shifts?' (to show the bell is a leftover habit, part of the case for softer sounds), and why the octopus passage mentions blue blood (to build the case that the animal is deeply unusual). Held PARTIAL: no audited item asks how a sentence, paragraph or section fits the overall structure of the text. That reading appears only in the explanations, and both items are also credited to other standards. A new item on 'The Color Underneath', added on 31 Oct 2026 and not yet audited as evidence, asks why the author opens by saying the maples 'seem to repaint themselves' (to name a first impression that the facts then correct). The row stays at its audited verdict until a refute-first two-auditor re-check counts the new items. Blocking clause: Fits into the overall structure of a text |
| RI.6.6 | PARTIAL two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-octopus item 4) Author's most likely purpose for mentioning the blue blood — building the case for strangeness, vs warning and mechanism foils. ▶ Play a live item from this skill |
An author's purpose is determined both for one detail (why the octopus passage mentions blue blood) and for whole texts ('How do the two passages differ in PURPOSE?'). The author's position is identified on the argument passage ('What is the author's CLAIM?'), and the two writers' attitudes are compared on the storm pair ('How do the TONES of the two passages compare?'). Held PARTIAL because 'explain how they are conveyed' asks the child to produce an explanation. Here the child picks answers, and the explanation of how purpose and point of view come through is given to them afterward. Blocking clause: Determine an author's point of view |
| RI.6.7 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
A quantitative visual is integrated with prose at Grade 6: the Pony Express passage carries a rendered bar chart, and 'Using the CHART together with the text, about how many days did the Pony Express save…' cannot be answered from the text alone. Held PARTIAL: it is one chart on one passage, and in Spanish the chart's bar labels ('Stagecoach', 'Telegraph') are still in English while the passage and question are in Spanish. |
| RI.6.8 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED-RECORD-CORRECTED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
An argument is now traced and evaluated on 'The Case for Quiet Hallways': 'What is the author's CLAIM?', 'Which piece of evidence SUPPORTS the claim?', 'Which claim is NOT supported with evidence?' and 'Why does the author mention factory shifts?'. Held PARTIAL at audit: a single argument passage, and the same bank is also credited to the Missouri twin. A second argument, 'Every School Needs a Garden', was added on 31 Oct 2026 with four more items, including one that asks how the author answers people who say gardens cost too much; it has not been audited as evidence. The row stays at its audited verdict until a refute-first two-auditor re-check counts the new items. Blocking clause: a second argument passage — every audited item sits on 'The Case for Quiet Hallways' |
| RI.6.9 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
A genuine cross-text comparison now serves: 'Two Views of the Storm' sets a journal beside a neutral news brief of the same outage, with items on tone, purpose, what appears in both, and what only one tells. Held PARTIAL at audit: one passage pair. A second pair, 'Two Ways to Hear a Whisper' (a science-magazine article beside a travel journal), was added on 31 Oct 2026 and has not been audited as evidence. The row stays at its audited verdict until a refute-first two-auditor re-check counts the new items. Blocking clause: a second pair of texts — every audited item sits on 'Two Views of the Storm' |
| RI.6.10 | PARTIAL two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Grammar L6 now asks the child to use conventions to understand a sentence. The child reads a sentence and uses its paired commas, dashes or parentheses to pick out the extra (nonessential) information, and picks the reason 'When Leo met Sam, he was late' is unclear: the pronoun could point to either boy. Held PARTIAL: these are standalone sentences in the grammar bank, which is also credited to the Writing conventions standards, and no item has the child apply conventions while reading an informational passage. Blocking clause: When reading to improve comprehension (applied while reading an informational text) |
| RI.6.11 | PARTIAL two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (chromatophores item); vocabularySaplingQ (SAPLING_ROOTS) Technical word defined from context; root/affix bank. ▶ Play a live item from this skill |
Stem graded on sub-codes: a covered, b capped PARTIAL (shared bank), c and d not covered. Multiple-meaning words never appear at Grade 6. The app's multiple-meaning items (bark, bat, trunk…) are early-grade words, dealt only at the vocabulary levels below Grade 6. Blocking clause: unknown and multiple-meaning words |
| RI.6.11.a | FULL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-octopus item 3) Chromatophores defined from the appositive right beside the word — 'tiny sacs of pigment called chromatophores.' ▶ Play a live item from this skill |
The chromatophores item is exactly using context (an in-text definition) to determine a word's meaning, in English and Spanish, and the why teaches the move the child can reuse. The Pony Express passage adds a phrase whose meaning the child reads from context ('beaten not by faster horses, but by a wire'). Thin FULL: one word and one phrase, on two informational passages, at audit. Two more words met in informational passages, 'carotenoids' in 'The Color Underneath' and 'donated' in 'Books on Horseback', were added on 31 Oct 2026 and have not been audited as evidence. |
| RI.6.11.b | PARTIAL two-auditor sweep 03 Oct 2026 |
vocabularySaplingQ (SAPLING_ROOTS) Same 15-entry roots/affixes bank as RL.6.11.b, both forms. ▶ Play a live item from this skill |
Identical bank, identical cap: one shared item bank is the sole credit for both strands' .11.b. The child picks a word's meaning from its parts rather than defining it, and no Grade-6 item asks the child to use a root or affix on a word met inside an informational passage. The Pony Express passage's 'telegraph' is built from two bank parts, tele- and graph, but no question touches it. One of the 14 parts, -less, is an English suffix rather than a Greek or Latin one, and its Spanish word ('sin esfuerzo') has no part to decode. Capped PARTIAL per the shared-bank rule. Forms: Same 14-entry roots/affixes bank as RL.6.11.b, both forms. Blocking clause: Use common, grade-appropriate Greek or Latin affixes and roots to define the meaning of a word |
| RI.6.11.c | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | No Grade-6 item uses reference materials. The app's dictionary and glossary items (finding a word between guide words, reading a glossary entry for a word's meaning and part of speech, a picture dictionary) sit only in the earlier grade levels, and the app has no live dictionary lookup. |
| RI.6.11.d | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | The verify-your-inference checking behavior is never elicited; the whys perform it for the child. |
| RI.6.12 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Two of the stem's three legs are now real questions at vocabulary L6 (connotation pairs; word relationships), and a figure of speech is interpreted inside the informational Pony Express passage. Held PARTIAL because the vocabulary items are standalone and shared across the literature/informational twins. |
| RI.6.12.a | PARTIAL two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
A figure of speech is interpreted in an informational text. On the Pony Express passage, the child chooses what 'beaten not by faster horses, but by a wire' means (a new technology, not a rival service, ended the Pony Express), in English and Spanish. Held PARTIAL: at audit it was the only figure interpreted in a Grade-6 informational text, one fixed item, while the standard asks about figures of speech in general. A new item on 'The Color Underneath', added on 31 Oct 2026 and not yet audited as evidence, asks why the author opens by saying the maples 'seem to repaint themselves' (to name a first impression that the facts then correct). The row stays at its audited verdict until a refute-first two-auditor re-check counts the new items. Blocking clause: Interpret figures of speech (plural): only one figure is interpreted in an audited Grade-6 informational item |
| RI.6.12.b | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED-AFTER-FIX) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Word relationships are now the question at vocabulary L6 ('What relationship joins "cello" and "instrument"?' — cause and effect, part and whole, item and category). Held PARTIAL: a thin family, the bank is shared with the literature twin, and the items never sit in an informational text. |
| RI.6.12.c | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Connotation among similar denotations is directly assessed at vocabulary L6, in English and Spanish. Same cap as the literature twin for the same evidence: a shared bank, and the items never sit in an informational text. |
| RI.6.13 | PARTIAL two-auditor sweep 03 Oct 2026 |
comprehensionSaplingQ (p6-octopus) One informational passage read by the child (passageSelfRead), five items across central idea, evidence, technical vocabulary, purpose, inference. ▶ Play a live item from this skill |
The child reads eight short Grade-6 informational texts and answers questions on each: a science passage (the octopus), a history passage with a bar chart (the Pony Express), an argument (quiet school hallways) and a journal paired with a news brief (the storm), plus four added on 31 Oct 2026 and not yet audited as evidence: a science passage (fall leaf color), an argument (school gardens), a science-magazine article paired with a travel journal (a whispering gallery) and a history passage (Kentucky's pack-horse librarians). A 'Read to me' button is offered on every passage. Held PARTIAL on complexity: each text runs about 100-160 words, far below the length and density of a grade-6 informational text. Blocking clause: of appropriate quantitative and qualitative complexity for sixth grade |
| W.6.1 | PARTIAL adversarial: DOWNGRADE · re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-argument (writing-coach L4) Should your school start an hour later? — real written production, min 250 chars, LLM-judged against rubric: arguable claim, two distinct pieces of evidence/reasoning, answers an opposing view, organized paragraph. |
The Writing Coach has the child write a real argument paragraph on one of two fixed topics (a later school start; phones in class), judged against a rubric, in English and Spanish. Both rubrics require a clear, arguable claim and two distinct pieces of evidence or reasoning; the first adds an answer to an opposing view, a formal style and a concluding statement, the second adds connecting words that link claim to reasons. Held PARTIAL: this writing happens only inside Writing Coach — a paid feature that runs only when the grading service is reachable and that a school can switch off — and no row met only through Writing Coach is marked FULL. The evidence is the child's own reasoning; no source is supplied. |
| W.6.1.a | PARTIAL adversarial: DOWNGRADE · re-audited ×1 · re-audited ×2 (CONFIRMED-EVIDENCE-REANCHORED) · two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-argument rubric lines 1 and 4 'A clear, arguable claim' and 'Complete sentences, organized as one paragraph' are judged on the child's own writing. |
Introducing a claim is graded on real writing: both Grade-6 argument prompts' rubrics begin with 'A clear, arguable claim', in English and Spanish. Organizing the reasons and evidence is not graded — neither argument rubric has an organization line, and the prompts ask only for 'a paragraph'. Held PARTIAL; this writing also happens only inside Writing Coach, so it could not be FULL. |
| W.6.1.b | PARTIAL two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-argument rubric line 2 'Two distinct pieces of evidence or reasoning' judged on the child's writing. |
Supporting a claim with reasons and evidence is genuinely produced and graded. No sources exist in the flow — the child argues from experience, so credible-source use and demonstrating understanding of a TEXT are never met. Blocking clause: using credible sources |
| W.6.1.c | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
One Grade-6 argument prompt (phones in class) carries the graded rubric line 'Cohesive words (because, therefore, however) link claim to reasons', in English and Spanish, although the prompt itself does not ask for them. Outside Writing Coach, Grade-6 Vocabulary asks the child to pick the transition that best connects two sentences — recognition, not use. Held PARTIAL: the child uses these words in their own writing only inside Writing Coach, and no row met only through Writing Coach is marked FULL. Blocking clause: Use words, phrases and clauses to clarify the relationships among claim(s) and reasons — met only through Writing Coach (published rule: no FULL for Writing Coach-only rows) |
| W.6.1.d | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
One Grade-6 argument prompt's rubric grades 'Formal style — no slang, no chatty asides' across the child's paragraph, and one explanatory prompt's rubric grades topic words used 'in a formal style'. Held PARTIAL: formal style is checked only inside Writing Coach, and no row met only through Writing Coach is marked FULL. Neither prompt tells the child to write formally (the requirement is in the rubric only), and 'maintain' is tested over a single short paragraph. |
| W.6.1.e | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
One Grade-6 argument prompt (a later school start) carries the graded rubric line 'A concluding statement that follows from the argument', in English and Spanish. Held PARTIAL: it is checked only inside Writing Coach, and no row met only through Writing Coach is marked FULL. The prompt the child reads does not ask for a conclusion — the requirement is in the rubric only — and the second argument prompt has no conclusion line. Blocking clause: Provide a concluding statement — met only through Writing Coach (published rule) |
| W.6.2 | PARTIAL adversarial: NOTE-ONLY · re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-inform (writing-coach L4) Explain something you know how to do better than most people — framed opening, three+ precise steps/facts in logical order, domain-specific words, closing that says why it matters; min 250 chars, LLM-judged. |
OFFICIAL DEFECT, stated per instructions: KS's printed W.6.2 statement is a copy-paste of RL.6.2 (theme/summary — a READING standard); this verdict judges W.6.2 by its sub-codes, which carry the real informative/explanatory writing standard. Real informative writing is rubric-graded on two prompts: a how-to/explain prompt whose rubric names domain words in a formal style, and a compare-and-contrast prompt whose rubric grades two similarities, two differences and 'Transition words connect the points'. Held PARTIAL: this writing happens only inside Writing Coach (no row met only through Writing Coach is marked FULL), headings, charts and multimedia are impossible in the plain text box, and only two organizational shapes are ever asked for. Blocking clause: Use appropriate transitions to clarify the relationships among ideas and concepts |
| W.6.2.a | PARTIAL two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-inform rubric lines 1-2 'An opening that frames the topic' and 'in a logical order' judged on production. |
Introducing a topic and organizing ideas are graded on real writing in two prompts: an explain prompt ('An opening that frames the topic', steps 'in a logical order') and a compare-and-contrast prompt whose rubric grades two similarities AND two differences. Held PARTIAL: definition, classification and cause/effect are never asked for; headings, charts and multimedia are impossible in the plain text box; and this writing happens only inside Writing Coach. Blocking clause: using strategies such as definition, classification, comparison/contrast and cause/effect |
| W.6.2.b | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-inform rubric line 2 'Three or more precise steps or facts, in a logical order' judged on the child's own explanatory writing. |
Developing a topic with relevant facts and concrete details is graded on the child's own writing: the explain prompt's rubric requires 'Three or more precise steps or facts, in a logical order', and the compare prompt requires two similarities and two differences. Quotations and definitions never come up (the standard's list is illustrative). Held PARTIAL: this writing happens only inside Writing Coach, and no row met only through Writing Coach is marked FULL. Blocking clause: Develop the topic with relevant facts ... — met only through Writing Coach (published rule) |
| W.6.2.c | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
The compare-and-contrast prompt asks the child to 'Use transition words to steer the reader between points' and its rubric grades 'Transition words connect the points', in English and Spanish. Outside Writing Coach, Grade-6 Vocabulary has the child pick the transition that fits two sentences — recognition, not use. Held PARTIAL: transitions are used in the child's own writing only inside Writing Coach, and no row met only through Writing Coach is marked FULL. Blocking clause: Use appropriate transitions — met only through Writing Coach (published rule) |
| W.6.2.d | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-inform rubric line 3 'Domain-specific words used correctly' judged on the child's chosen expertise topic. |
The explain prompt asks for 'precise steps or facts', and its rubric also grades 'Domain-specific words used correctly, in a formal style' on the child's chosen topic, in English and Spanish. Held PARTIAL: this is checked only inside Writing Coach, and no row met only through Writing Coach is marked FULL; the domain-word requirement is in the rubric, not the prompt the child reads, and it appears on one prompt only. Blocking clause: Use precise language and domain-specific vocabulary — met only through Writing Coach (published rule) |
| W.6.2.e | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Formal style appears in one explanatory prompt's rubric, folded into a single line — 'Domain-specific words used correctly, in a formal style' — in English and Spanish. Held PARTIAL: the prompt the child reads never asks for a formal style, the compare-and-contrast prompt does not check it, and it is graded only inside Writing Coach. |
| W.6.2.f | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-inform rubric line 4 'A closing that says why it matters' judged on production. |
Both Grade-6 explanatory prompts ask for and grade a closing — 'A closing that says why it matters' and, on the compare prompt, 'A closing that matches player to game' — in English and Spanish. Held PARTIAL: closings are written and graded only inside Writing Coach, and no row met only through Writing Coach is marked FULL. Blocking clause: Provide a concluding statement — met only through Writing Coach (published rule) |
| W.6.3 | PARTIAL adversarial: DOWNGRADE · re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-narrative (writing-coach L4) Write a scene, not a summary: two characters, one moment of tension, two+ lines of correctly punctuated dialogue, pacing slows at the key moment, sensory/concrete detail; min 280 chars, LLM-judged. |
Real narrative writing with technique (dialogue, pacing, description) is rubric-graded on two scene prompts; the second adds transition words that move time forward and a change of mind the scene must earn. Held PARTIAL: both prompts confine the child to a single scene, so event sequences stay short; pacing is graded on one prompt only; and narrative writing happens only inside Writing Coach, where no row is marked FULL. |
| W.6.3.a. | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-narrative rubric line 1 'A single scene with two characters and real tension' judged on production. |
Introducing characters and establishing a situation are graded via the scene requirement, and the second narrative prompt's 'At least two transition words moving time forward' gives a sliver of event-sequence organization. Held PARTIAL because no rubric line on either prompt grades reader orientation or establishing a context. Blocking clause: organize an event sequence that unfolds naturally and logically |
| W.6.3.b | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-narrative rubric lines 2-4 Dialogue (two+ correctly punctuated lines), pacing (the key moment gets the most words), description (sensory/concrete detail, not labels) — each a graded rubric line. |
The standard's three named techniques are each a graded rubric line on real writing: dialogue (two correctly punctuated lines), pacing (the key moment gets the most words) and description (sensory or concrete detail, not labels). Dialogue and description are graded on both narrative prompts; pacing on one. Held PARTIAL: narrative writing happens only inside Writing Coach, and no row met only through Writing Coach is marked FULL. Blocking clause: Use narrative techniques — met only through Writing Coach (published rule) |
| W.6.3.c | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
The second narrative prompt grades 'At least two transition words moving time forward' on real production, so time-frame transitions are exercised. Held PARTIAL because shifts of setting and a VARIETY of transitional phrases and clauses are not addressed. |
| W.6.3.d | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-narrative rubric line 4 'Sensory or concrete detail, not labels' judged on production. |
Descriptive detail is graded on the child's own narrative writing on both scene prompts: 'Sensory or concrete detail, not labels' and 'Concrete detail, not labels'. Held PARTIAL: the rubric accepts concrete OR sensory detail, so sensory language itself is never required; precise word choice is checked only as 'not labels'; and this writing happens only inside Writing Coach. |
| W.6.3.e | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
The first narrative prompt's rubric grades 'An ending that follows from the scene's events' on real writing, in English and Spanish (the second narrative prompt has no ending line). Held PARTIAL: it is checked only inside Writing Coach, and no row met only through Writing Coach is marked FULL; the prompt the child reads does not ask for an ending. Blocking clause: Provide a conclusion — met only through Writing Coach (published rule) |
| W.6.4 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-argument/inform/narrative/analyze Four distinct task types (argue, inform, narrate, analyze), each rubric-judged for development and organization on real production. |
Clear, coherent writing is produced and graded across four purposes — argue, inform, narrate, analyze — with two prompts each. Held PARTIAL: no prompt names or varies an AUDIENCE. Organization appears in the rubric only on the explain prompt ('in a logical order'), the compare prompt (its two-similarities, two-differences shape joined by transitions) and the two analysis prompts ('One organized paragraph'); neither argument prompt grades organization. Style is graded only where a rubric line names it, and this writing happens only inside Writing Coach. Blocking clause: appropriate to task, purpose and audience |
| W.6.5 | PARTIAL two-auditor sweep 03 Oct 2026 |
WriteBox try-again flow (one revision invited after coaching feedback) On a below-rubric submission the coach shows targeted feedback (grow items) and invites exactly one revision before resolution. |
Revision with feedback happens: when a first draft misses the rubric, the coach shows up to three short, pre-written tips and invites one revision, which can be a full rewrite. Held PARTIAL: only one revision, offered only after a failing draft; planning is never asked for; the tips are fixed lines about sentence mechanics, detail, staying on topic, the order of steps or events, point of view and giving reasons — none speaks to a claim, an opposing view or a conclusion; there is no adult or peer step; and this happens only inside Writing Coach. Blocking clause: by planning, revising, editing, rewriting or trying a new approach |
| W.6.6 | PARTIAL adversarial: DOWNGRADE · two-auditor sweep 03 Oct 2026 |
WriteBox (typed production in-app) All Grade-6 writing is composed by keyboard/touchscreen in the app. |
The child produces writing with technology: every Grade-6 Writing Coach answer is composed on a keyboard and graded. Held PARTIAL: writing is never published, shared or used to collaborate (it goes to the grader, and only to a private parent notice if a safety concern is flagged); the answer box holds at most 700 characters, so the three-pages-in-a-sitting keyboarding benchmark cannot be reached; and all of this happens only inside Writing Coach. Blocking clause: To produce and publish writing as well as to interact and collaborate with others; three pages in a single sitting |
| W.6.7 | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Short research projects drawing on several sources — research wall. No Grade-6 item poses a research question, has the child gather from several sources, or refocus an inquiry; the in-app passages and chart are read only to answer single comprehension questions. |
| W.6.8 | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Gathering information from print/digital sources, assessing credibility, quoting/paraphrasing with attribution — research wall, nothing behind any verb. |
| W.6.9 | PARTIAL adversarial: RECLASSIFY · two-auditor sweep 03 Oct 2026 |
WRITE_PROMPTS wp-six-analyze Name the theme of a book you finished this year and defend the reading with two specific moments — rubric: theme as idea not topic word, two specific moments as evidence, explains HOW each moment shows the theme. |
Drawing evidence from a text to support analysis is written and graded on two prompts: a story's theme defended with two specific moments, and an informational text's main idea backed by two specific details, each explained. Grade-6 Reading also has the child pick the evidence that supports a claim in an in-app passage (selection). Held PARTIAL: in the writing prompts the text is one the child chose outside the app, so the evidence cannot be checked against it; reflection and research are never asked for; and the writing happens only inside Writing Coach. Blocking clause: analysis, reflection and research |
| W.6.10 | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED-EVIDENCE-REANCHORED) · two-auditor sweep 03 Oct 2026 |
writing-coach L4 rubrics ('Complete sentences') + WriteBox coaching Every wp-six rubric demands complete sentences on production; coaching flags convention breaks in the child's own draft. |
Stem graded on its sub-codes. Grammar and usage are practised at grammar L6 by recognition: subject and object pronoun case, an intensive pronoun and a vague pronoun. In the child's own writing they are practised through Writing Coach. One Grade-6 argument prompt grades 'Consistent point of view — pronouns do not drift', another grades 'Formal style — no slang, no chatty asides', and the coach can flag pronoun case and pronoun shifts. No Grade-6 rubric line asks for complete sentences. Sentence variety is never asked for, possessive case was not assessed at audit, and the writing half runs only through Writing Coach, so no sub-code is fully met. A possessive-pronoun item ('The blue backpack is hers, and the red one is mine.'), added on 31 Oct 2026, has not been audited as evidence. Blocking clause: Ensure that pronouns are in the proper case (subjective, objective, possessive) |
| W.6.10.a | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Sentence-pattern variety is never demanded by a prompt or graded by a rubric. |
| W.6.10.b | PARTIAL re-audited ×1 · re-audited ×2 (UPGRADED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Subject and object pronoun case are assessed at grammar L6 with 'Which sentence uses the pronoun correctly?' ('My brother and I built the fort'; 'The coach chose Maya and me for the relay'), and the drop-the-other-name test is taught in the explanation, in English and Spanish. Held PARTIAL: possessive case was not assessed at audit, and in the child's own writing pronoun case is only a coaching note, not a graded line. A possessive-pronoun item ('The blue backpack is hers, and the red one is mine.'), added on 31 Oct 2026, has not been audited as evidence. The row stays at its audited verdict until a refute-first two-auditor re-check counts the new items. Blocking clause: Possessive |
| W.6.10.c | PARTIAL re-audited ×1 · re-audited ×2 (UPGRADED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Intensive pronouns are recognized at grammar L6: 'Which sentence uses an INTENSIVE pronoun to add emphasis?' ('The mayor herself answered our letter'). Held PARTIAL: the child picks the sentence that has one but never uses an intensive pronoun in writing, and at audit it was a single fixed item. Two more, added on 31 Oct 2026 and not yet audited as evidence, ask the same question with the keys 'The cook herself tasted every dish.' and 'We painted the class banner ourselves.' The row stays at its audited verdict until a refute-first two-auditor re-check counts the new items. Blocking clause: Use intensive pronouns |
| W.6.10.d | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
One Grade-6 argument prompt grades 'Consistent point of view — pronouns do not drift' on the child's own writing, and the coach can flag a drift from 'I' to 'you'. Held PARTIAL: this runs only through Writing Coach. No Grade-6 grammar item presents a pronoun shift; the one consistency item ('Pick the version where the pronouns stay CONSISTENT') sits at the grade-5 grammar level. A shift in pronoun number is never presented. |
| W.6.10.e | PARTIAL re-audited ×1 · re-audited ×2 (UPGRADED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
A vague-pronoun item serves at grammar L6: "Why is this sentence UNCLEAR? 'When Leo met Sam, he was late.'" The child picks the diagnosis (the pronoun could mean either boy), in English and Spanish. Held PARTIAL: in the audited item the corrected sentence is shown in the explanation but never produced or chosen by the child, and at audit it was one fixed item; three more vague-pronoun items were added on 31 Oct 2026 and have not been audited as evidence. One of them asks the child to choose the corrected sentence ('Which revision fixes the UNCLEAR pronoun?'). The row stays at its audited verdict until a refute-first two-auditor re-check counts the new items. Blocking clause: And correct |
| W.6.10.f | PARTIAL two-auditor sweep 03 Oct 2026 |
WriteBox coaching feedback + one-revision flow The coach identifies variations from standard English in the child's OWN draft (grow-flagged feedback) and the child may revise once. |
The child's own writing gets coaching on convention slips and one revision when a draft falls short. But the coach does the spotting: it names the kind of slip and the rule to apply (for example 'Your spelling is close — read it out loud and fix the word that looks off'), and the child only makes the change. This runs only through Writing Coach. Others' writing appears as single test sentences at grammar L6, where the child picks the sentence that uses the pronoun or the punctuation correctly, or says why a pronoun is unclear. The strategies (drop the other name; delete the extra part) are taught in the explanations, never identified by the child. Blocking clause: in their own and others' writing |
| W.6.10.g | PARTIAL two-auditor sweep 03 Oct 2026 |
WriteBox coaching feedback + one-revision flow Continuation fragment of W.6.10.f ('…conventional language.') — the extraction preserves KS's own line break as a separate code. |
Not a standalone standard: this code is the tail of W.6.10.f's sentence in the KS layout. Graded identically to W.6.10.f and stated as a fragment so no one counts it as independent coverage. Blocking clause: in their own and others' writing |
| W.6.10.h | PARTIAL two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Two Grade-6 Writing Coach prompts grade a steady formal style in the child's own writing ('Formal style — no slang, no chatty asides'; 'Domain-specific words used correctly, in a formal style'). Held PARTIAL: this runs only through Writing Coach, tone is never graded on its own, and no coaching note targets style or tone. Blocking clause: Maintain consistency in ... tone (style graded only through Writing Coach) |
| W.6.11 | PARTIAL two-auditor sweep 03 Oct 2026 |
wp-six-narrative rubric ('correctly punctuated dialogue'); all wp-six rubrics ('Complete sentences') Dialogue punctuation graded on production; sentence-boundary capitalization/punctuation enforced by every rubric. |
Stem graded on sub-codes. Punctuation is practised in real writing, because correctly punctuated dialogue is a graded line on both Grade-6 narrative prompts. Commas, parentheses and dashes around nonessential information are assessed at grammar L6 by recognition. No Grade-6 rubric line grades capital letters, end marks or spelling; the coach can only flag them. Held PARTIAL. Blocking clause: Use punctuation (commas, parentheses, dashes) to set off nonrestrictive/parenthetical elements |
| W.6.11.a | PARTIAL re-audited ×1 · re-audited ×2 (CONFIRMED) · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Grammar L6 assesses this standard with all three mark types: identify the nonessential part ('My uncle, who lives in Denver, collects maps.'; 'The recipe — an old family secret — takes three days.'; 'Saturn (the sixth planet) has visible rings.') and 'Which sentence punctuates the EXTRA information correctly?', in English and Spanish. Held PARTIAL because 'use' is a production verb met here only by selection, over twelve fixed sentences, eight of them added on 31 Oct 2026. |
| W.6.11.b | PARTIAL adversarial: UPGRADE · two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Spelling is never a graded line at Grade 6: no Grade-6 rubric names it, and the writing judge is told not to fail a strong answer for minor spelling or punctuation slips. The coach can still flag a misspelling in the child's own draft ('Your spelling is close — read it out loud and fix the word that looks off'), and if the draft falls short the child gets one revision. The Spelling skill belongs to the grades 3–5 band, has no Grade-6 level, and its words are English only. |
| W.6.12 | PARTIAL two-auditor sweep 03 Oct 2026 |
writing-coach L4 (four prompts, repeatable single-sitting sessions) Short-time-frame written production across four discipline-relevant purposes, routinely available in the play loop. |
Single-sitting writing for a range of tasks and purposes genuinely happens, across eight Grade-6 Writing Coach prompts in argument, explanation, narrative and analysis. Extended-time-frame writing (multi-day pieces with research and sustained revision) is structurally impossible in the app. A draft lives only on the screen it was written on, is capped at 700 characters, and gets at most one revision, offered only when it falls short. Blocking clause: extended time frames (time for research, reflection and revision) |
| SL.6.1 | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Collaborative discussion with partners — interaction-mode wall (no peer or adult interlocutor exists). Not stretched; applies to the whole SL.6.1 family. |
| SL.6.1a | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Coming to discussions prepared presupposes a discussion; none exists. |
| SL.6.1b | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Rules, goals, and roles for collegial discussion — interaction-mode wall. |
| SL.6.1c | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Posing and responding to questions in live discussion — interaction-mode wall. |
| SL.6.1d | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Reviewing others' ideas and paraphrasing multiple perspectives — interaction-mode wall. |
| SL.6.2 | PARTIAL two-auditor sweep 03 Oct 2026 |
— No generator is named on this row yet; the note says what is practised. |
Interpreting information in different formats is practised at Grade 6 on one passage. 'Ten Days to California' (the Pony Express) comes with a rendered bar chart, and two of its questions need it: reading the days saved from the bars together with the text, and saying what the telegraph bar shows that the text puts in words. That covers the visual and quantitative part of the standard. The questions are in English and Spanish, but the chart's bar labels (Stagecoach, Pony Express, Telegraph) stay in English in Spanish mode. The 'explain how it contributes' part is met only by choosing an answer. No Grade-6 item presents an oral or video source (the Read-to-me button only voices the printed passage), and the K-5 listening branch sits below the band. Blocking clause: Explain how it contributes to a topic, text or issue under study |
| SL.6.3 | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Delineating a SPEAKER'S argument needs an argument the child listens to. The Grade-6 argument passage ('The Case for Quiet Hallways') does ask which claim lacks evidence, but it is a text to read. The optional Read-to-me button only voices the printed page, and no item asks the child to listen. That reading work is not stretched to this listening standard. |
| SL.6.4 | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Oral presentation of claims and findings — speaking wall. The written argument prompt is credited to W.6.1, not stretched here. |
| SL.6.5 | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Multimedia components in presentations — multimedia + speaking wall. |
| SL.6.6 | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Adapting SPEECH to contexts and tasks — speaking wall. |
| SL.6.7 | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Grammar conventions WHEN SPEAKING — speaking wall; applies to the whole SL.6.7 family regardless of the sub-skills' overlap with W.6.10. |
| SL.6.7.a | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Sentence variety in speech — speaking wall. |
| SL.6.7.b | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Intensive pronouns in SPEECH — speaking wall. (In writing, an intensive-pronoun item does serve at grammar L6 and is credited to W.6.10.c.) |
| SL.6.7.c | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Pronoun-shift correction in speech — speaking wall. |
| SL.6.7.d | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Vague-pronoun correction in speech — speaking wall. |
| SL.6.7.e | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Recognizing variations from standard English in speaking — speaking wall. |
| SL.6.7.f | NOT COVERED two-auditor sweep 03 Oct 2026 |
— | Style and tone consistency in speech — speaking wall. |
| SL.6.8 | PARTIAL two-auditor sweep 03 Oct 2026 |
vocabularySaplingQ (SAPLING_ROOTS); wp-six-inform rubric line 3 Academic word-part acquisition across 15 roots/affixes; 'Domain-specific words used correctly' graded on the child's own informative writing. ▶ Play a live item from this skill |
KS houses the acquire-and-use-vocabulary standard in SL. Acquisition is practised at vocabulary L6 in English and Spanish: 14 roots and affixes, transition words and phrases (however, as a result, for example, in addition, otherwise), connotation pairs and word relationships. Gathering word knowledge is practised too: reading items ask what 'rehearsed' and 'chromatophores' mean from the passage, and decode items build a word's meaning from its parts. Accurate USE in the child's own words is graded only by the metered writing grader: the Writing Coach inform prompt's domain-vocabulary line, and the short written math explanations that ask for terms such as 'per', 'factor' and 'term'. By the published rule that cannot carry a FULL, so the row stays partial. Blocking clause: gather vocabulary knowledge when considering a word or phrase important to comprehension or expression |
These are the rulings the adversarial passes changed, kept in the open: a report that never shows its corrections is asking to be trusted on faith.
| Code | Change | Reason |
|---|---|---|
| 6.RP.2 | FULL → PARTIAL | The standard's lead verb phrase is 'Use unit rate language (“for each one”, “for every one” and “per”) and unit rate notation to demonstrate understanding'. The mappers' own house rule graded KS 6.RP.1 PARTIAL because 'Use ratio language to describe' is production — the identical verb construction here got FULL. ratiosSaplingQ L6 (content.ts ≈9472-9511): the child computes/compares unit rates and RECOGNIZES rate language in answer options; no item has the child use the language, and the a/b fraction-notation clause ('3/4 cup of flour for each cup of sugar') has zero content — every generated rate is a whole or half-dollar amount. The FULL note itself concedes both gaps. Cross-state: the same generator is PARTIAL under MO 6.RP.A.2's weaker 'describe the meaning' clause. blockingClause: 'Use unit rate language … and unit rate notation'. |
| 6.NS.6 | FULL → PARTIAL | Statement: 'Understand a rational number as a point on the number line and a coordinate pair as a location on a coordinate plane.' Verified in code: integersQ L1 n = 2 + rnd(18) (integers only), coordinateSaplingQ nz() = ±(1+rnd(8)) (integers only). No fraction or decimal is ever placed anywhere in either generator, and no number line or plane is ever rendered — the number-half of the standard is one verbal-description form (‘n units LEFT of zero’). The draft's own precedent grades umbrellas against their statement text (6.EE.2, 6.SP.5 both graded down); 'a rational number as a point' witnessed only through integer cases with no line is the same kind of gap the mapper's own note flags and defers to 6.NS.6c. The pair-as-location half is genuinely covered; the number-as-point half is not — that is the definition of PARTIAL. |
| 6.NS.7a | FULL → PARTIAL | Statement verb: 'INTERPRET statements of inequality as statements about the relative position of two numbers on a number line diagram.' The cited items run the opposite direction: integersQ L2 form 0 asks the child to PRODUCE the order ('Which is LESS: −a or −b?'), and L3 form 1 has the child endorse a true statement — that is 6.NS.7d's distinguish task (already claimed there as FULL). No item ever presents a written inequality (−3 > −7) and asks what it says about position; the positional reading lives entirely in the whys ('−lo sits farther left'), which the mapper's own note admits. Feedback text is not assessed content anywhere else in this crosswalk — 6.NS.7c was capped at PARTIAL for exactly the explanation-lives-in-the-feedback reason. blockingClause: 'Interpret statements of inequality as statements about the relative position of two numbers on a number line diagram'. |
| 6.EE.1 | PARTIAL → PARTIAL | Forms-description mismatch with code: the draft says exponent bases run 'b 2-6'; content.ts 10272 is `const b = 2 + rnd(4)` — bases 2-5 (the MO draft says 2-5, correctly). Exponent e is 2-4 for b≤3, 2-3 for b 4-5. Verdict (PARTIAL, blocked on 'Write') is unaffected; correct the forms string. |
| 6.EE.2c | FULL → PARTIAL | The standard contains a named operational clause: 'Perform arithmetic operations, INCLUDING THOSE INVOLVING WHOLE-NUMBER EXPONENTS, in the conventional order.' Verified in expressionsQ L2 (content.ts 10270-10291): the exponent branch is a bare single-operation b^e item; the evaluate branch is ax + b with no exponent. No generated expression anywhere mixes an exponent with another operation, so order-of-operations WITH exponents — the clause that makes this a Grade-6 standard rather than a Grade-5 one — is never exercised. The mapper's own note names this exact gap and graded FULL anyway. Evaluate-at-values and the formula clause (volumeSaplingQ V=lwh/V=Bh at x.5) are genuinely covered; the exponent-order clause is not. blockingClause: 'including those involving whole-number exponents, in the conventional order'. |
| 6.EE.8b | NOT COVERED → PARTIAL | Content the mappers missed: expressionsQ L3 branch 1 (content.ts 10304-10311) — 'Each ticket costs $per and you pay a flat $k for shipping. Which expression gives the cost of b tickets?' correct: per·b + k. This has the child select the expression for the DEPENDENT quantity (total cost) in terms of the INDEPENDENT variable (b, tickets) — the exact d = 65t-family relationship the standard names, minus the equation form ('c =') and minus 'write' (selection). Under this crosswalk's own conventions, selection covering part of a write clause yields PARTIAL, not NOT COVERED (cf. 6.EE.7, graded PARTIAL on write-as-selection of h > c). Blocked remainder: the equation form is never posed, nothing is written, and the list-and-graph-ordered-pairs example content is absent — so PARTIAL, no higher. |
| 6.G.1 | FULL → FULL | Verdict held under attack, but a hostile evaluator will press the closing clause 'apply these techniques in the context of solving REAL-WORLD and mathematical problems': verified in geometrySaplingQ L6-L7, every area item is a bare labeled shape (cm/m units, no situation — no garden, wall, or floor context anywhere). All named shape classes (triangle, parallelogram, trapezoid, kite, rectilinear L) and compose/decompose are genuinely present, which carries the standard's core; recommend one contexted area form to close the flank. Same exposure on 6.G.2 (bare prisms). |
| 6.SP.5c | FULL → PARTIAL | Internal-consistency overturn. The statement's second half is a production clause: 'as well as DESCRIBING any overall pattern and any striking deviations from the overall pattern with reference to the context.' The mapper's note concedes 'no item has the child characterize a pattern' — the identical concession that produced PARTIAL at 6.NS.5b ('explaining'), 6.NS.7c ('explain'), 6.SP.1 ('generate'), and 6.EE.2a ('write'). The outlier item (statisticsQ L4) identifies a striking deviation by recognition, which keeps this above the give-measures-only floor, but describing an overall pattern has zero content (cluster language appears only in a why). Giving mean/median/mode/range is fully covered; the describing clause is blocked by the house production wall. blockingClause: 'as well as describing any overall pattern … with reference to the context in which the data were gathered'. |
| 6.SP.4 | NOT COVERED → PARTIAL | statisticsDisplayQ L5 build register constructs a real dot plot from listed data (the create verb, on the shipped plot builder). BLOCKING: "histograms, stem-and-leaf plots, and box plots" remain undisplayable. SELF-ASSESSED after the adversarial pass — re-adversarial required at go-live. |
| 6.SP.5a | NOT COVERED → PARTIAL | L5 dot-plot/histogram reading items ask for the total number of observations from a rendered display. BLOCKING: "reporting" as free production. SELF-ASSESSED after the adversarial pass — re-adversarial required at go-live. |
| 6.EE.2a | PARTIAL → FULL | Two separate adversarial auditors run by this project (a verb-fidelity auditor and a refutation auditor, round 1, wf_001cb4de-a84) both marked this FULL after the write/type/draw builds: 26 Sep 2026 (round 1, two auditors agreed FULL): expressionsQ L3 types 2n − k and per·b + k from words; the polynomial judge refuses k − 2n and 2(n − k) and accepts commuted forms; ~half of live L3 deals, EN+ES; typedKeyTruth re-derives every key from its prompt. |
| 6.EE.7 | PARTIAL → FULL | Two separate adversarial auditors run by this project (a verb-fidelity auditor and a refutation auditor, round 1, wf_001cb4de-a84) both marked this FULL after the write/type/draw builds: 26 Sep 2026 (round 1, two auditors agreed FULL): Write typed at L1 (~half of deals), 'infinitely many' asked directly at L3, and RayPick has the child BUILD the boundary, open/closed circle and direction (~37% of L3); checkAlgebra refuses h ≥ 40 against h > 40 on direction. |
| 6.EE.3 | PARTIAL → FULL | Two separate adversarial auditors run by this project (a verb-fidelity auditor and a refutation auditor, round 2, wf_44c82963-87a) both marked this FULL after the write/type/draw builds: 26 Sep 2026 (round 2): the child types equivalent expressions — a(x + b) expanded with brackets refused, ax + bx combined into one term — with the given form, bracketed and uncombined answers refused, in about half of live Level 4 items in English and Spanish. |
| 6.EE.8b | PARTIAL → FULL | Two separate adversarial auditors run by this project (a verb-fidelity auditor and a refutation auditor, round 2, wf_44c82963-87a) both marked this FULL after the write/type/draw builds: 26 Sep 2026 (round 2): the child types d = kt from a table of distances and hours — the standard's own d = 65t pattern — with the dependent variable d required alone on one side; the ordered pairs are given, not listed or graphed by the child. |
| 6.NS.5a | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: Use positive and negative numbers to represent quantities in real-world contexts — the child only picks the signed number from three choices |
| 6.NS.7 | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: ordering and absolute value of rational numbers — at the Grade 6-8 levels only integers are ordered or taken absolute value of |
| 6.G.1 | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: apply these techniques in the context of solving real-world ... problems |
| 6.SP.5d | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: Relating the choice of measures of variability to the distribution of the data |
| 6.EE.4 | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: which values from a specified set, if any, make the equation or inequality true |
| 6.EE.8a | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: Identify the independent ... variable |
| 6.EE.8c | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: Analyze the relationship ... using graphs ... and relate these to the equation |
| 6.NS.4 | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: Use the distributive property to express a sum of two whole numbers 1–100 with a common factor as a multiple of a sum |
| 6.NS.6b | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: recognize that when two ordered pairs differ only by signs, the locations of the points are related by reflections across one or both axes (no item ever asks about the both-axes case; the L4 reflection item gives its answer away by form) |
| 6.G.2 | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: in the context of solving real-world and mathematical problems (no real-world item); V = Bh item answerable from its choices |
| 6.EE.8c | NOTE-ONLY | Note refresh, 10 Oct 2026 (after commit 9c765be): the shown-graph item's slope slip is dealt on either side of k instead of always (k + 2)t; the note names the shipped slips and keeps the remaining shares-form-and-number tell. Verdict unchanged. |
| 6.EE.8c | NOTE-ONLY | Note refresh, 10 Oct 2026 (r6/twovar): the table and shown-graph multiple choice now deal four equations (both forms × both numbers) around a proportional OR additive relationship, so the shares-form-and-number tell is gone — the pairing rule fell from 75% to 25% (= chance) over 600+ deals per language, locked at the 95% bound in saplingGrade6Gaps.test.ts; generator and forms filled in. The note now says what the item still does not do (graph ↔ table; values read off the graph). Verdict unchanged — held at PARTIAL for the second re-audit. |
| 6.EE.8b | NOTE-ONLY | Note refresh, 10 Oct 2026 (r6/twovar): the typed-equation table is proportional (d = kt) or additive (d = t + k) in equal shares since this round; note, generator and forms name both shapes. Verdict unchanged. |
| 6.EE.8c | NOTE-ONLY | Note refresh, 10 Oct 2026 (r6/twovar, after the separate review of round 6): the forms text listed one option set (d = 2t, d = 3t, d = t + 2, d = t + 3) while the item also dealt d = t, d = 4t and other equations that were never the key — d = t as a wrong in about a quarter of graph deals, so striking it was free ('strike any option never seen as the key' scored 40% on the graph, 29% on the table, chance 25%), and only d = 3t was plotted with three points. The visible numbers are now an adjacent pair drawn inside the key range ({1, 2} or {2, 3} on the graph, with d = t a real key; 2 to 8 on the table), every graph plots t = 1, 2, 3, and the never-key rule scores 25%; saplingGrade6Gaps.test.ts bounds every equation and number dealt as also the key. Forms and note describe HEAD. Verdict unchanged (PARTIAL). |
| 6.EE.7 | NOTE-ONLY | Note refresh, 11 Oct 2026 (after the separate review of widget/r7 at 949b7a3): the forms text described the solution-set picture item for x > c only, with the open-circle, shaded-right answer every time; since r7 it asks x > c, x < c, x ≥ c or x ≤ c, the key sits at each circle-and-direction corner in turn, and two of the other three pictures are dealt, each named. The 'how many solutions' options are quoted as shipped. Verdict unchanged (FULL). |
| 6.SP.3 | NOTE-ONLY | Note refresh, 11 Oct 2026 (after the separate review of widget/r7 at 949b7a3): the note described the L6 range item with the medians always equal; since r7 they are equal in half the deals and far apart in the other half, and the options cross a center claim with a spread claim, two of the three wrong combinations dealt and named. The two-group spread item's weakness is unchanged and now also names that its key is the shortest option in every deal (measured over 900 deals per language). Verdict unchanged (FULL). |
| 6.EE.7 | NOTE-ONLY | Note refresh, 11 Oct 2026 (widget/r7 after merging r7/batch1-content, r7/batch1-guard and main): the 'how many solutions does x > c have?' options were re-worded on r7/batch1-content so the three are one length and one word count (the key had sat at the middle length in every deal); the forms text quotes the live wording. Verdict unchanged (FULL). |
| 6.RP.2 | NOTE-ONLY | Note refresh, 24 Oct 2026 (keyfirst-a, measured over 450–600 real deals per language in English and Spanish; locked in app/src/lib/keyfirstGiveaway.test.ts): the better-buy options printed each package's true price per ounce ('A — $0.25 per oz'), so the key was the smaller printed price in every deal and 'they cost the same per ounce' was never the key. The options now name only the packages beside 'they cost the same per ounce', the per-ounce working is in the explanation, and the two packages cost the same per ounce in about a third of deals. Forms and note describe the live item. Verdict unchanged (PARTIAL, on the language clause). |
| 6.SP.2 | NOTE-ONLY | Note refresh, 24 Oct 2026 (keyfirst-a, measured over 450–600 real deals per language in English and Spanish; locked in app/src/lib/keyfirstGiveaway.test.ts): the note said the L5 histogram 'MOST' answer is always the middle interval and the L2 median item always prints the median in the middle; the tallest bar now stands over any of the three intervals (a third each) and the five median values are printed in a shuffled, never-sorted order (the median is the middle printed value in about 19% of deals), so both need the data read. Same-day repair: with only list neighbours offered, 'the choice nearest the mean' found the median in every choice deal (nearest the midpoint of the extremes, 86–89%); each list now has one far value and offers its mean (one of the five values) as a wrong choice. Round 3 (re-verification, same day): that list let 'one choice far from the other two → the opposite end, else the middle' win 60–66% of choice deals and a gap-pattern lookup 73–81%; the list's gaps and the wrong pair are now dealt together, the mean is offered in about five deals in eight, and no rule on the choices' order or spacing (alone or read from the far value's side) tops about 42%, exactly over every list it can deal and on 3,000 real deals per language and mode, nearest the mean or the midpoint of the extremes about 37%. Still open and now stated in the note: working out the mean exactly and taking the closest other choice finds the median every time. Verdict unchanged (FULL). |
| 6.G.2 | NOTE-ONLY | Note refresh, 24 Oct 2026 (keyfirst-a, measured over 450–600 real deals per language in English and Spanish; locked in app/src/lib/keyfirstGiveaway.test.ts): the V = Bh item's single-rule giveaways are closed, but it still gives its answer away to a determined guesser. Its wrong choices were always B × hWhole and B + h, both below the key, so 'a .5 choice → the larger whole number, else the middle' won every deal. The first repair's 'base counted twice' slip (2Bh) was always double the key ('the smaller of a doubled pair' won 67–70%) and 'the .5 one of the two closest choices' won 67–69%; the same-day repair deals a pair from seven named slips (one list per parity of B), never one choice a whole-number multiple of another. Over 1,500 deals per language and mode: each size rank about a third, the compound .5 rule 38–43%, the closest-two rules at most 44%, no doubled pair. Still open (re-verification, same day): how many choices end in .5, read together with which two sit closer or whether they are evenly spaced, finds the key in about 81% of deals (exact over every base area and height; chance 33%), so the blocking clause keeps the real-world clause and the V = Bh choices clause, re-worded to name that pattern. Verdict unchanged (PARTIAL). |
| 6.G.2 | NOTE-ONLY | Note refresh, 24 Oct 2026 (keyfirst-a, founder decision of 24 Oct 2026 recorded in GAPS.md 24OCT2026-b; measured over 3,000 real deals per language and mode in English and Spanish; locked in app/src/lib/keyfirstGiveaway.test.ts): V = Bh (volume L5) is now typed on the number pad in every interactive deal ('What is the volume, in cubic centimeters?'), with no choices on screen and a typed slip (the half dropped, B + h, or the height rounded up) answered with its own note. Printed sheets, the placement check and Duel cannot take typing and have no numeric write-in, so they keep the three choices, where the .5 count read with which two sit closer still finds the answer in about 81% of deals (82% measured; capped in the guard). The V = Bh part of the blocking clause is removed because practice now requires the formula; the paper residual is stated in the note; the real-world clause stays. Verdict unchanged (PARTIAL). |
| 6.SP.2 | NOTE-ONLY | Note refresh, 24 Oct 2026 (keyfirst-a, founder decision of 24 Oct 2026 recorded in GAPS.md 24OCT2026-b; measured over 3,000 real deals per language and mode in English and Spanish; locked in app/src/lib/keyfirstGiveaway.test.ts): the L2 median is typed on the number pad in every interactive deal (it was about half), with no choices on screen. Printed sheets, the placement check and Duel keep three list-value choices; the note now names everything open there, as the re-verification asked: the mean-neighbour route (every time), the no-arithmetic list-extreme rule (about 75%; about 80% with the list's range, up to about 89% with the choices' span), 'smallest choice 6 or less' (about 58%) and the second-closest choice to the midrange (about 63%). The earlier sentence 'No rule that reads only the choices ... more than about 42%' is narrowed to the choices' order and spacing, which is what was measured. Verdict unchanged (FULL). |
| 6.SP.5c | NOTE-ONLY | Note refresh, 24 Oct 2026 (keyfirst-a, founder decision of 24 Oct 2026 recorded in GAPS.md 24OCT2026-b; measured over 3,000 real deals per language and mode in English and Spanish; locked in app/src/lib/keyfirstGiveaway.test.ts): the median is typed in every interactive deal, the mean, mode and range still in about half; the note no longer says all four are typed in about half. Verdict unchanged (PARTIAL). |
| 7.RP.1 | PARTIAL → FULL | Two separate adversarial auditors run by this project (a verb-fidelity auditor and a refutation auditor, round 1, wf_001cb4de-a84) both marked this FULL after the write/type/draw builds: 26 Sep 2026 (round 1, two auditors agreed FULL): complexRateQ types the unit rate (numpad, unconditional in live rounds); key (a/b)/(c/d) re-derived from the prompt on 2,726 live items with 0 mismatches; like-unit, area and different-unit contexts all dealt; at least 2 per round. |
| 8.G.1b | PARTIAL → NOT COVERED | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached NOT COVERED. Blocking clause: An angle that turns through n one-degree angles is said to have an angle measure of n degrees |
| 7.SP.1b | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: justify your reasoning |
| 7.SP.4 | PARTIAL → FULL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached FULL. |
| 8.EE.7b | PARTIAL → FULL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached FULL. |
| 7.EE.1 | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: add, subtract, and factor linear expressions with rational coefficients — each item's options give the key away |
| 7.RP.2a | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: Determine whether two quantities are in a proportional relationship |
| 7.RP.2b | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: recognize that, in a proportional relationship, every pair of numbers has the same unit rate |
| 8.F.2 | FULL → PARTIAL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached PARTIAL. Blocking clause: Compare properties of two linear functions represented in a variety of ways (algebraically, graphically, numerically in tables, or by verbal descriptions) |
| 7.G.2 | PARTIAL → FULL | Two-auditor audit of every grades 6-8 math row, 03OCT2026: two separate auditors run by this project re-read the row against the live code and both reached FULL. |
| 7.EE.2 | NOTE-ONLY | Note refresh, 10 Oct 2026 (after commit ae4ee1a): the increase form's three choices now all open 'raising by N% …' and name the percent and the factor; the note no longer says the key alone repeats the prompt's percent. Verdict unchanged. |
| 7.EE.3 | NOTE-ONLY | Note refresh, 10 Oct 2026 (after commit ae4ee1a): the reasonableness item's off estimates are now products from the same range as the fair ones (1.7–12× off either way, every estimate a multiple of 10), so the note no longer says the estimate's size alone decides the item; the printed product is still named as the remaining limit. Verdict unchanged. |
| 7.RP.2d | NOTE-ONLY | Note refresh, 10 Oct 2026 (after commit 9c765be): the point-(1, r) item's second wrong option is a starting-cost reading rather than a second copy of the reversed rate, and the key carries no '— the unit rate' tag; the note now quotes the shipped options. Verdict unchanged. |
| 8.EE.5 | NOTE-ONLY | Note refresh, 10 Oct 2026 (after commit 9c765be): the slope-triangles item's wrong options now share wording with each other, and the ratios L8 equation item's slope slip sits on either side of k; the note quotes the shipped options. Verdict unchanged. |
| 7.RP.2d | NOTE-ONLY | Note refresh, 10 Oct 2026 (after commit 282dd18, merged at fc8983c): proportionalGraphQ now plots (0, 0) with the other points and draws the line y = rx from the origin on every form that shows the grid; the note no longer says no line is drawn and (0, 0) is not plotted. Verdict unchanged. |
| 7.EE.1 | NOTE-ONLY | Note refresh, 11 Oct 2026 (after the separate review of widget/r7 at 949b7a3; measured on r7/batch1-guard with everyAskGuard's sign-aware word overlap and middle length rank): the note said the add item's key always shares its coefficient with one wrong answer and its constant with the other; since r7 the add item deals one of three pairs of named slips, and the option sharing the most with the other two is its key in 72–77% of deals (chance 33%). The combine and factor keys are that option in every deal, the combine key is also the middle-length option in every deal, and the expand −(x − b) key is the only option with a +. Note and blocking clause describe widget/r7. Verdict unchanged (PARTIAL). |
| 7.SP.4 | NOTE-ONLY | Note refresh, 11 Oct 2026 (after the separate review of widget/r7 at 949b7a3; measured on r7/batch1-guard with everyAskGuard's sign-aware word overlap and middle length rank): the note said one form holds the medians equal and compares ranges; since r7 that form prints medians that are equal in half the deals and far apart in the other half, and both it and the mode form cross two claims and deal two of the three wrong combinations, each named. Verdict unchanged (FULL). |
| 7.EE.1 | NOTE-ONLY | Note refresh, 11 Oct 2026 (widget/r7 after merging r7/batch1-content, r7/batch1-guard and main; measured with everyAskGuard's sign-aware word overlap and middle length rank, 900 deals per language): the add item is now a full 2×2 — the key and three named slips — and the option sharing the most with the others, the odd one out and the middle-length option are each its key in 25% of deals (chance 25%); the note and blocking clause no longer count the add item among the items whose options give the key away. Verdict unchanged (PARTIAL). |
| 7.EE.4b | NOTE-ONLY | Note refresh, 11 Oct 2026 (widget/r7 after merging r7/batch1-content, r7/batch1-guard and main; measured with everyAskGuard's sign-aware word overlap and middle length rank, 900 deals per language): the multiple-choice px + q > r and px < c solve items now deal four options (the direction right or turned wrongly × the boundary's sign right or lost), with a negative coefficient in three deals in five; the note no longer says their options give the key away. Verdict unchanged (PARTIAL). |
| 7.EE.4b | NOTE-ONLY | Note refresh, 11 Oct 2026 (widget/r7, ineqSolve2x2): the inequality multiple choice now deals a 2x2 — direction right or turned wrongly, crossed with the boundary's sign right or lost — with a negative coefficient in about 3 deals in 5; the why names each step and whether dividing flips the sign. The earlier 'tested by substitution in the why' wording no longer described the item. Verdict unchanged. |
| 7.EE.2 | NOTE-ONLY | Note refresh, 31 Oct 2026 (answer-giveaway round 8 review, revised in its third pass): the increase form's wrong readings changed. 'Nearly doubles the price: 1.05 ≈ 2' is gone; the two wrong choices are now drawn from multiplying by 0.05 or 1.50, adding 1.05 or 1.50 dollars, and dividing by 1.05, all the same length. The note names them and the surface leans left: picking the choice that names the printed 1.05 finds the right one in about half of English deals and a little over two in five Spanish ones, and knowing only that it is a multiplication in about 85%. Verdict unchanged. |
| Code | Attack | Why the claim held |
|---|---|---|
| 6.SP.4 | Rescue attempt (would flip NOT COVERED upward): K-5 dataQ has real display machinery — rendered pictographs/bar charts, a BUILD-the-line-plot interactive (consPlotQ, 3.MD.4 'make' clause), tally charts, and 4.MD.4 line-plot reading in fractions of a unit. A line plot is essentially a dot plot, so this looked like accidental 6.SP.4 coverage. | dataQ caps at level 3 (`Math.min(3, level)`, content.ts ≈2441) and has no sapling register in catalog.ts — not Sapling-reachable, so it cannot count for Grade-6 attribution under the house rule. Even ignoring reachability, KS 6.SP.4 names four display types (dot plots, histograms, stem-and-leaf, box plots) and the K-5 content touches at most the dot-plot cousin; histograms/stem-and-leaf/box plots appear nowhere in the repo. Same result for MO 6.DSP.B.4a/4b. NOT COVERED stands, all four codes. |
| 6.NS.2 | Fluency claim: every quotient is exact (nd = q0×dv by construction), dividends top out ~1218 (L6) / ~3465 (L7), MC with q±1 distractors verifies the answer, not an efficient algorithm; no remainders ever. | The standard's verb is 'divide multi-digit numbers' and multi-digit ÷ 2-digit is genuinely computed; the q±1 distractors cannot be beaten by estimation alone, forcing the actual division. The KS parenthetical defines fluently as efficiently/accurately/flexibly without mandating remainders or a shown algorithm, and no MC engine can observe algorithm choice — the house standard accepts computation-by-MC for fluency codes (same grade both states). The no-remainders ceiling is honestly recorded in the note. |
| 6.NS.4 | Bounds audit against the standard's explicit numeric limits, plus a production attack on 'express … as a multiple of a sum'. | Verified in code: GCF pool filter keeps both operands ≤100 (`g*m1 <= 100 && g*m2 <= 100`); every LCM pair ∈ {[4,6],[6,8],[4,10],[6,9],[8,12],[6,10],[8,10],[9,12]} has both operands ≤12; distributive branch guarantees coprime remainders (SAPLING_COPRIME) so g is genuinely the GCF and the rewrite matches the 18+48=6(3+8) exemplar, with the un-divided-addends distractor g(a+b) requiring the factoring to be performed. Express-by-selection is the house-accepted reading of 'use … to express' (graded identically at MO 6.NS.B.4b). Tie-check: a×b ≠ LCM for every pair (all share a factor >1), so the distractor never collides. |
| 6.NS.7 | Integers-only scope: the umbrella says 'ordering and absolute value of RATIONAL NUMBERS' and no fraction/decimal is ever ordered or absolute-valued (integersQ generates integers exclusively) — the same textual ground on which 6.NS.6 fell. | Deliberate asymmetry with 6.NS.6, on principle: 6.NS.6's statement makes a representational demand ('a point on the number line') that nothing renders, while 6.NS.7's statement names two concepts — ordering, absolute value — both directly assessed by generated items, and every worked example in its own children (−3 > −7, −3°C, |−30|) is an integer. An understand-standard whose own exemplars are integers survives integer-only instantiation; the rational-scope thinness is already carried by the PARTIALs at 7b/7c (and now 7a). Same reasoning preserves MO 6.NS.C.7 FULL (definition-of-absolute-value directly assessed, distance-from-zero why verbatim). |
| 6.NS.6b | The reflection item's source point is ALWAYS Quadrant I (x = 1+rnd(8), y = 1+rnd(8), both positive — verified) and the asked reflection is ALWAYS across the x-axis; the y-axis and both-axes relations are never the correct answer. | The standard's demand is to 'recognize that when two ordered pairs differ only by signs, the locations are related by reflections across one or both axes' — and the distractor set IS that recognition task: choosing (x,−y) over (−x,y) and (−x,−y) requires mapping each sign pattern to its reflection, and the why names all three. The quadrant branch independently covers the signs→quadrant clause across all four quadrants (nz() draws both signs). Same hold for MO 6.GM.A.3b. Recommend rotating the asked axis, but the FULL survives. |
| 6.EE.4 | Checked whether the 'specified set' framing is real or cosmetic, and whether the inequality half exists (the standard demands both equation AND inequality). | Both stems carry literal sets: equationsSaplingQ L6 '{x−1, x, x+2}' against ax + b = rhs, inequalitiesQ L2 '{c−a−1, c−a, yes}' against x + a > c, with the boundary value c−a excluded by strictness — the exact substitution logic of the standard, and the whys state 'testing each value from the set IS the method'. The cleanest FULL on the sheet; identical hold for MO 6.EEI.B.4. |
| 6.EE.8a | The generator only ever asks 'Which is the DEPENDENT variable?' — the standard says 'Identify the independent AND dependent variable', and the independent variable is never the asked-for answer. | The independent variable is always the discriminating distractor (plus a 'neither' option), so a child who cannot tell the two roles apart cannot answer; picking the dependent over the independent identifies both in one act, and the why teaches the changes-in-response test in both directions. Three rotating contexts. Thin, but the verb is met. |
| 6.G.2 | 'Fractional edge lengths' (plural) realized as the narrowest possible case: verified in volumeSaplingQ that the fractional edge is ALWAYS exactly +0.5 and ALWAYS the height (l, w, B whole); also the real-world-context clause — both items are bare prisms. | Both named formulas (V=lwh, V=Bh) are applied with a genuinely fractional edge, the half-layer meaning connects to the packing concept the KS parenthetical invokes, and the rounded-edge distractor (l×w×hWhole) is exactly the missing half-layer — the fractional edge is load-bearing, not decorative. The standard sets no breadth requirement on which or how many edges. Same hold for MO 6.GM.A.2b; the .5-only narrowness is recorded for the content team in the KS note already. |
| 6.RP.3c | Breadth: only two conversion pairs exist in the entire generator (inches→feet, hours→minutes) — no metric, no capacity, no weight, no between-system conversion, no compound units. | KS's text names no systems and no breadth ('Use ratio reasoning to convert measurement units; manipulate and transform units appropriately when multiplying or dividing quantities'): both transformation directions occur (÷12 small→large, ×60 large→small) with the .5h→50min ratio trap making the ratio-reasoning clause real. The contrast case proves the method: MO 6.RP.A.3d, whose text DOES demand 'between two systems', was correctly graded PARTIAL on the same generator. |
| 6.SP.2 | Shape clause without displays: no item addresses cluster, peak, gap, symmetry, or skew; data is never presented as a distribution to look at. | KS's shape list is verbatim '(cluster, peak, gap, symmetry, skew (data) AND/OR outlier)' — outlier alone satisfies the disjunction, and the L4 outlier items assess it directly, with cluster language in the why as a bonus. Center (mean/median/mode) and spread (range, 'and/or' again) are computed outright. The FULL rests on the and/or reading, which is the standard's own text. |
Some verbs in the standards cannot be honestly assessed by an app, and we do not pretend otherwise: conducting real surveys, experiments, or simulations (we assess the design-judgment half); freehand drawing and construction; and speaking & listening as spoken practice (taking part in a discussion, presenting, listening to a speaker). Codes built on those verbs appear above as NOT COVERED or with the limitation named in their notes.