In 0607, the GDC does the arithmetic of statistics: averages from a list, an estimated mean from grouped data and, for Extended students, the equation of the regression line. The marks that remain are for entering data correctly, reading the right values off the screen and saying what the numbers mean. This guide covers both halves with four worked examples.
The syllabus references are C10.5 and E10.5 (Averages on a calculator) for both tiers, and E10.7.4 (use a GDC to find and use the equation of linear regression) for Extended only. If your child keeps losing these "easy" marks, our 0607 specialist tutors can work through them on the student's own calculator.
What is on the list, by tier
| Skill | Core | Extended | Reference |
|---|---|---|---|
| Mean, median and quartiles for discrete data on the GDC | Yes | Yes | C10.5.1 / E10.5.1 |
| Mean for grouped data on the GDC | Yes | Yes | C10.5.2 / E10.5.2 |
| Scatter diagram and a straight trend line drawn by eye | Yes | Yes | C10.7 / E10.7 |
| Equation of linear regression on the GDC | No | Yes | E10.7.4 |
The Core syllabus notes that the coefficient of correlation is not required. Both tiers describe correlation in words: positive, negative or zero.
E10.5 and E10.7.4 are on the list of outcomes that appear on calculator papers only. You will meet them on Paper 3 or 4, and they can feed into an investigation or modelling task on Paper 5 or 6. The wider statistics topic, including charts and cumulative frequency, is in our 0607 statistics guide. For where this sits in the course, see the full 0607 syllabus overview.
Entering data without errors
- Clear old lists first. Leftover values from last week are the most common reason a mean looks wrong.
- Check the count. After entering, the GDC shows n. If the question has 11 values and the screen says 10, one is missing.
- For a frequency table, use two lists. Put values (or midpoints) in one list and frequencies in the other, then tell the one-variable statistics tool which list holds the frequencies.
- For regression, keep x and y in the right lists. The explanatory variable goes in x. Swapping them gives a different line.
On TI-84 models, lists are under STAT, EDIT and the calculations under STAT, CALC. On Casio fx-CG models, use the Statistics menu and then CALC. On TI-Nspire models, use a Lists and Spreadsheet page and the Statistics menu.
Worked example 1: averages from a list
The number of minutes 11 students spent on homework was: 12, 15, 9, 22, 18, 14, 11, 25, 16, 13, 20. Find the mean, median, quartiles and interquartile range.
Enter the 11 values into one list and run one-variable statistics. Check that n = 11.
- Mean = 15.909…, so 15.9 minutes (3 s.f.)
- Median = 15
- Lower quartile Q1 = 12 and upper quartile Q3 = 20
- IQR = 20 − 12 = 8 minutes
Check: in order the data are 9, 11, 12, 13, 14, 15, 16, 18, 20, 22, 25. The 6th value is 15, the 3rd is 12 and the 9th is 20. With some data sets, a calculator's quartile rule and a textbook's can give slightly different answers. If yours disagree, ask your teacher which convention your school's mark schemes follow.
Worked example 2: an estimated mean for grouped data
Time spent waiting at a clinic (minutes):
| Time t | 0 ≤ t < 10 | 10 ≤ t < 20 | 20 ≤ t < 30 | 30 ≤ t < 40 | 40 ≤ t < 50 |
|---|---|---|---|---|---|
| Frequency | 4 | 9 | 13 | 8 | 6 |
Estimate the mean.
Enter the midpoints 5, 15, 25, 35, 45 in list 1 and the frequencies 4, 9, 13, 8, 6 in list 2. Set the frequency to list 2 and run one-variable statistics. n should be 40.
Estimated mean = 25.75, so 25.8 minutes (3 s.f.).
Say "estimate". Using midpoints assumes each value sits in the middle of its class. If you type the class boundaries instead of midpoints, you get a wrong answer, and this is the error we see most.
Worked example 3: the regression line and a prediction (Extended)
Eight students recorded hours of revision x and test score y (out of 100):
| x | 2 | 3 | 5 | 6 | 8 | 9 | 11 | 12 |
|---|---|---|---|---|---|---|---|---|
| y | 41 | 45 | 52 | 58 | 63 | 66 | 75 | 78 |
(a) Find the equation of the regression line. (b) Estimate the score for 10 hours of revision. (c) Comment on using the line for 25 hours.
(a) Enter x in list 1 and y in list 2. Run linear regression in the form y = ax + b. The GDC gives a = 3.6739… and b = 34.032…, so:
y = 3.67x + 34.0 (3 s.f.)
(b) Use the unrounded values: y = 3.6739… × 10 + 34.032… = 70.77…, so the estimate is 70.8.
(c) At x = 25 the line gives 126, which is impossible for a test out of 100. Twenty-five hours is far outside the data range of 2 to 12 hours, so this is extrapolation. The prediction is not reliable.
A mark is often available for that last sentence, and it needs a reason, not just "unreliable". The line passes through the mean point (7, 59.75), which is a useful check if you also draw it on a scatter diagram.
Worked example 4: negative correlation and context (Extended)
A café records the midday temperature x (°C) and the number of hot drinks sold y:
| x | 8 | 11 | 14 | 17 | 20 | 23 | 26 | 29 |
|---|---|---|---|---|---|---|---|---|
| y | 96 | 90 | 81 | 74 | 66 | 60 | 52 | 45 |
Find the regression line, interpret the gradient and estimate sales at 18 °C.
The GDC gives a = −2.452… and b = 115.86…, so y = −2.45x + 116 (3 s.f.).
Interpretation: for each 1 °C rise in temperature, the café sells about 2.45 fewer hot drinks. Give the gradient's meaning in the units of the question. "The line goes down" is not enough.
At 18 °C: y = −2.452… × 18 + 115.86… = 71.7…, so about 72 drinks. Rounding to a whole number is sensible here because you cannot sell part of a drink. State that you rounded for context.
Mistakes our tutors see
- Typing class boundaries instead of midpoints for grouped data.
- Not linking the frequency list, so the GDC finds the mean of the frequencies.
- Using rounded a and b in a prediction, which can change the final digit.
- Writing the regression line with a rounded intercept, such as y = 3.7x + 34, when 3 s.f. was expected.
- Reading σx (standard deviation, not in the syllabus) instead of x̄.
- No comment on reliability when predicting outside the data range.
When to get help
- Answers are right on one question and wrong on the next, and the student cannot say why. This usually means list or frequency set-up habits.
- The student can find a regression line but cannot write a sentence on what the gradient means.
- Interpretation marks on Paper 4 or Paper 6 are lost consistently.
- The student has changed calculator model and is slower than before.
Self-check
- Find the mean, median and quartiles of 4, 7, 7, 8, 10, 11, 13.
- Midpoints 5, 15, 25, 35, 45 have frequencies 4, 9, 13, 8, 6. What value of n should your GDC show?
- Find the regression line for x = 1, 2, 3, 4, 5 and y = 3, 5, 8, 9, 12.
- A regression line is y = −2.45x + 116. What does −2.45 mean if x is temperature and y is drinks sold?
- Why is a prediction at x = 40 unreliable if the data run from x = 8 to x = 29?
Answers. 1: mean 8.57, median 8, Q1 7, Q3 11. 2: 40. 3: y = 2.2x + 0.8. 4: about 2.45 fewer drinks for each 1 °C rise. 5: it is extrapolation beyond the data, and the trend may not continue.
Where it leads
Regression grows in IB Mathematics: Applications and Interpretation, which adds correlation coefficients and other models. Our guide to IB AI regression and Spearman's rank shows the next step. Within 0607, regression often appears in Paper 6 modelling. Our guide to investigation and modelling on Paper 6 shows how, and the GDC graph tools are in zeros and intersections on the GDC.
How Maths Bodhi helps
The free demo includes one data question done on the student's own GDC, watched step by step. Most list errors show up in the first two minutes. After that, our tutors set short weekly sets that mix calculator statistics with interpretation sentences, and track which marks are still being lost. For extra help with the data topics, see our one-to-one statistics tutoring.
Home tuition is available in Gurgaon (Gurugram). Students elsewhere in India and abroad learn with us online. You can book a free demo class.
Written and checked by the Maths Bodhi team. Facts checked against official Cambridge International and Pearson sources, November 2026.
Frequently asked questions
- Is linear regression in IGCSE 0607 Core?
- No. Finding the equation of linear regression on a GDC is Extended only (E10.7.4). Core students draw a straight trend line by eye.
- How do I find the regression line on my GDC?
- Enter x values in one list and y values in another, then choose linear regression in the form y = ax + b from the statistics calculation menu and write the equation to 3 significant figures.
- Do I need the correlation coefficient for 0607?
- The Core syllabus says the coefficient of correlation is not required. Describe correlation as positive, negative or zero and interpret it in context.
- How do I find the mean of grouped data on a calculator?
- Put the class midpoints in one list and the frequencies in another, link the frequency list in one-variable statistics, and state that the answer is an estimate.
- Why is my GDC quartile different from my textbook answer?
- Calculators and textbooks sometimes use slightly different quartile rules for certain data sets. Check which convention your teacher and mark schemes use.
- Can GDC statistics come up on 0607 Paper 2?
- No. Averages on a calculator (E10.5) and the regression equation (E10.7.4) are tested on calculator papers only.
