| Component | Max | 4 | 5 | 6 | 7 |
|---|---|---|---|---|---|
| Overall | 100 | 45 | 58 | 70 | 83 |
SL Paper 1
Routine statistics and substitution went well, but distributions, rational function ranges, tangents, part-cylinder measurement and an unscaffolded optimisation proved hard, and late questions were often blank.
Found difficult
- binomial distribution
- normal distribution
- tree diagrams
- domain and range
- tangents and normals
- surface area and volume of solids
- optimisation
Well prepared
- descriptive statistics
- evaluating functions
- chi-squared test
- voronoi diagrams
- midpoint
- basic probability
| Q | Topic | How it went | Where marks were lost |
|---|---|---|---|
| 1 | mode, median, standard deviation and range of a list | Mixed | gave the mean when the median was asked; could not find the standard deviation; gave the range as an inequality or as the IQRTip: The range is a single number: largest value minus smallest value. |
| 2 | logarithmic vocabulary model | Mixed | evaluated log of 10 to the power 10 instead of log 10; ignored the four significant figure instruction; comments made no use of the equation; did not reject a negative word count as unrealisticTip: Base comments on what the model predicts and question results that make no sense, such as negative counts. |
| 3 | chi-squared test for independence | Mixed | mixed up chi-squared and t-test hypothesis wording; gave the p-value to one significant figure without the full value; no explicit comparison with the significance level; conclusion not in the required contextTip: Always write the full p-value, compare it with the significance level, then conclude in context. |
| 4 | normal distribution with symmetry, rounded heights and inverse normal | Most struggled | missed that half the distribution lies below the mean; used calculator notation as the only working; used 24 or 26 instead of 25 as the boundary; used the wrong tail for the inverse normalTip: Draw a labelled normal curve and shade the region before using normalcdf or invnorm. |
| 5 | Voronoi diagram: nearest site and bisectors | Mixed | ignored one of the sites when comparing distances; gave y = 10 or a point instead of the line x = 7; could not find the perpendicular gradientTip: Two sites with the same y-coordinate have a vertical bisector of the form x = constant. |
| 6 | rational function: value, range and inverse value | Most struggled | gave the range as a closed interval between two values; computed f(-2) instead of the inverse at -2; read the inverse notation as a reciprocal or derivativeTip: f inverse of a value means the x giving that output; do not confuse it with 1/f or f'. |
| 7 | tree diagram with unknown branch probability | Mixed | added instead of multiplying along branches; did not identify all branches leading to the event; assumed or subtracted percentages to get bTip: Mark every path leading to the event, add their products, and set the total equal to the given probability. |
| 8 | quadratic model with symmetric vertex | Most struggled | did not set y = 0 at the x-intercept; forgot to use the value of c; could not form a second equationTip: Use the vertex, the intercepts and symmetry to write enough equations for every parameter. |
| 9 | equation of a tangent line | Mixed | found f(1) and f'(1) but no line equation; arithmetic slips evaluating f(1); differentiated only one term correctly; did not substitute x = 1 into the derivativeTip: Finish tangent questions by writing y - y1 = m(x - x1) using f(1) and f'(1). |
| 10 | binomial expected value and cumulative probability | Most struggled | gave n instead of np for the expected value; calculated probabilities by hand without recognising a binomial; mishandled P(X >= 7); truncated to two significant figuresTip: E(X) = np is in the formula booklet; use the binomial cdf for at least probabilities via 1 - P(X <= 6). |
| 11 | volume and surface area of a slice of a cylinder | Most struggled | missing or wrong units; used a pyramid volume formula; forgot the arc length for the curved surfaceTip: For part of a cylinder, the curved face is a rectangle with arc length times height. |
| 12 | exponential model with initial value | Mixed | unclear that b to the power 0 equals 1; rounded b to 1, oversimplifying later workTip: Keep growth factors unrounded; rounding b to 1 destroys the model. |
| 13 | minimising surface area of a box for a fixed volume | Most struggled | question often left blank; did not express area in terms of width only; did not set the derivative to zero or graph the functionTip: Use the volume to eliminate a variable, then minimise the area graphically or with the derivative. |
What to practise
- Follow accuracy instructions such as four significant figures.
- Show working; correct answers with wrong methods are not credited.
- Cross out rejected solutions, since only the first is marked.
- Write the unrounded answer before rounding.
- Practise from specimen and past papers.
SL Paper 2
A well-graded paper with entry points for all levels and good technology use; some questions were skipped and marks were lost to two significant figure answers.
Found difficult
- percentage error
- arithmetic and geometric sequences
- area of a parallelogram
- hypothesis testing
- reading contextual problems
- loans and amortization
- sinusoidal functions
- trapezoidal rule
- integration with boundary conditions
Well prepared
- efficient GDC use
- accuracy and rounding
- arithmetic and geometric sequences
- t-test
- sine and cosine rules
- area of a triangle
- right-angled trigonometry
- definite integrals and area
- optimisation
| Q | Topic | How it went | Where marks were lost |
|---|---|---|---|
| 1 | rounding, percentage error, arithmetic and geometric sequences of running distances | Mixed | divided by the approximate value in percentage error; gave a decimal instead of a percentage; substituted 42 for n instead of setting u_n = 42; wrong common ratio such as 0.2Tip: Distinguish n (position), u_n (term) and S_n (sum) before choosing a formula. |
| 2 | correlation coefficient, area and volume of a letter shape, t-test | Mixed | gave r squared instead of r, or truncated; struggled with the parallelogram area; recomputed the volume instead of multiplying area by depth; wrong hypotheses or unpooled settingTip: Reuse earlier results: a prism's volume is its cross-section area times its length. |
| 3 | triangle area, distances, cost and loan interest | Most struggled | did not give the answer in standard form; added distances instead of using geometry; forgot to double before multiplying; found an interest rate when interest paid was askedTip: Interest paid is an amount of money; the nominal annual rate is a percentage. Read which is wanted. |
| 4 | sinusoidal tide model and times a road is usable | Most struggled | question often omitted; swapped parameter values; gave b instead of the period; reversed the meaning of the condition or missed some timesTip: Period equals 360 divided by b; draw the graph with the threshold line to find every crossing time. |
| 5 | area under a curve, maximum, angle of elevation, trapezoidal rule and integration | Mixed | omitted dx; used the wrong trig ratio for the angle of elevation; used h = 30 or 15 instead of 20; forgot the constant of integrationTip: Always add and then determine the constant of integration from the given information. |
What to practise
- Round only the final answer.
- For multi-step problems, write in words what you need to find.
- Practise mixed questions combining several topics.
- Practise writing justified responses in words.
- Budget time so every question is attempted.
Internal assessment
What moderators raised about the exploration this session, criterion by criterion.
- A. title, research question and aim presented separately and inconsistently, plus long rationales
- A. abrupt transitions; results not interpreted straight after each process
- B. missing subscripts, inconsistent italics and case, unlabelled axes, poorly presented piecewise functions
- B. regression lines added to scatter plots by default
- C. high marks given for stated interest or effort alone
- D. conclusions based on assumptions; reflection left to a final generic list
- E. PMCC and regression done mechanically; outliers removed or ignored without justification
- E. models chosen only by R squared, with unjustified extrapolation and no domain restrictions
About this summary
After every session the IB sends schools a subject report: how candidates did on each paper and question, and what teachers should work on. This page is my reading of the Maths AI SL report for November 2025 TZ3, in my own words. No text from the report or from the exam questions is reproduced, and it is not endorsed by the International Baccalaureate.
Session codes: TZ is the time zone variant of the papers. Other variants of the same session:
Frequently asked questions
Which questions were hardest in Maths AI SL November 2025 TZ3?
+
The questions most candidates struggled with were on normal distribution with symmetry, rounded heights and inverse normal, rational function: value, range and inverse value, quadratic model with symmetric vertex and binomial expected value and cumulative probability. The tables above list every question discussed in the report, paper by paper.
What did candidates find difficult in November 2025 TZ3?
+
Across the papers the examiners listed binomial distribution, normal distribution, tree diagrams, domain and range, tangents and normals, surface area and volume of solids, optimisation and percentage error, among others.
What were the grade boundaries for Maths AI SL November 2025 TZ3?
+
Maths AI SL: a 7 from 83 out of 100 overall in November 2025 TZ3. The component boundaries are in the table on this page.
Is this the official IB subject report?
+
No. It is my summary of the report the IB issues to schools after each session, rewritten in my own words with a teacher's reading of it. No text from the report or from exam questions is reproduced. Your school's IB coordinator can access the original.