S1 data covers types of data, measures of location (mean, median, mode, quartiles and percentiles, with interpolation for grouped data), measures of spread (range, interquartile range, variance and standard deviation), coding, and outliers. The variance is "the mean of the squares minus the square of the mean".
Qualitative data are descriptions (colours, brands); quantitative data are numbers. Discrete data take separate values (number of siblings, shoe size); continuous data can take any value in a range (height, time).
Mean, median and quartiles
Mean \(\bar x = \frac{\sum x}{n}\), or \(\frac{\sum fx}{\sum f}\) from a frequency table. For grouped data, estimate the median and quartiles by interpolation: \(Q_2 = L + \frac{\frac n2 - F}{f} \times w\), where \(L\) is the lower boundary of the median class, \(F\) the cumulative frequency before it, \(f\) its frequency and \(w\) its width.
Variance and standard deviation
\(\sigma^2 = \frac{\sum x^2}{n} - \bar x^2\) and \(\sigma = \sqrt{\sigma^2}\). For \(2, 4, 4, 4, 5, 5, 7, 9\): \(\bar x = 5\), \(\sigma^2 = 4\) and \(\sigma = 2\). The interquartile range \(Q_3 - Q_1\) measures the spread of the middle half and is not affected by extreme values.
Coding
If \(y = \frac{x - a}{b}\), then \(\bar x = a + b\bar y\) and \(\sigma_x = b\sigma_y\). Adding a constant moves the mean but not the spread; multiplying scales both. With \(y = \frac{x - 100}{5}\), \(\bar y = 2.4\) and \(\sigma_y = 1.2\) give \(\bar x = 112\) and \(\sigma_x = 6\).
Outliers and skewness
Use the rule the question gives, often a value more than \(1.5 \times\) IQR below \(Q_1\) or above \(Q_3\). With \(Q_1 = 20\) and \(Q_3 = 30\), the limits are \(5\) and \(45\). Positive skew has a longer tail to the right, so the mean is usually above the median.
In the formula booklet = printed in Pearson's IAL formulae booklet, so know where to find it. Learn it = not printed there, so learn it by heart. All S1 formulas on one page (free PDF) →
S1 Data and Summary Statistics worked examples
Each mark scheme is checked line by line before it is published here. M marks are for method, A marks for accurate answers that follow from them, B marks for independent results.
Worked example 1 · 9 marks · hard
Two different groups of students took a memory test.
Group A consists of \(20\) students. Their mean score is \(15\) and the standard deviation is \(3\).
Group B consists of \(30\) students. Their mean score is \(20\) and the standard deviation is \(4\).
The two groups are combined into a single group of \(50\) students.
Find the combined mean score.
By calculating the total sum of squares (\(\sum x^2\)) for each group separately, find the overall standard deviation of the combined group.
Mark scheme
Sum A \(= 20 \times 15 = 300\). Sum B \(= 30 \times 20 = 600\). M1
Total Sum \(= 300 + 600 = 900\). Total students \(= 50\).
Combined mean \(= \frac{900}{50} = 18\). A1
The mean of three numbers, \(a\), \(b\), and \(c\), is exactly \(8\). The variance of these three numbers is \(10\).
Using the formula for variance, prove that \(a^2 + b^2 + c^2 = 222\).
Given that \(c = 10\), find the mean and the variance of the remaining two numbers, \(a\) and \(b\).
Mark scheme
Let the sum of squares be \(\sum x^2 = a^2 + b^2 + c^2\).
Using the variance formula: \(\text{Var}(X) = \frac{\sum x^2}{n} - \mu^2\). M1 \(10 = \frac{a^2 + b^2 + c^2}{3} - 8^2\). M1 \(10 = \frac{a^2 + b^2 + c^2}{3} - 64 \implies 74 = \frac{a^2 + b^2 + c^2}{3}\). A1 \(a^2 + b^2 + c^2 = 3 \times 74 = 222\). A1 *
If \(c = 10\), then \(a + b + 10 = 3 \times 8 = 24 \implies a + b = 14\). M1
New mean for just \(a\) and \(b\): \(\frac{14}{2} = 7\). A1
Substitute \(c=10\) into sum of squares: \(a^2 + b^2 + 100 = 222 \implies a^2 + b^2 = 122\). M1
New Variance for \(a\) and \(b\)\(= \frac{a^2 + b^2}{2} - (\text{New Mean})^2\). M1 \(= \frac{122}{2} - 7^2 = 61 - 49 = 12\). A1
2 steps
Worked example 3 · 5 marks · medium
A dataset of exam scores has a mean of \(42\) and a standard deviation of \(5\).
The teacher decides to adjust the grades by multiplying every score by \(2\), and then adding \(10\) to each score.
Find the exact value of:
the new mean.
the new standard deviation.
the new variance.
Mark scheme
When data is multiplied by \(a\) and \(b\) is added (\(y = ax + b\)), the new mean is \(a\mu + b\). M1
New mean \(= 2(42) + 10 = 84 + 10 = 94\). A1
Standard deviation is affected by multiplication but not by addition.
New SD \(= 2 \times 5 = 10\). A1
Variance is the square of the new standard deviation. M1
New Variance \(= 10^2 = 100\). A1
3 steps
Free account: +5 more practice questions on data and summary statistics with instant feedback, and full worked answers on the free questions. Free, no card. Join free →
Common S1 data and summary statistics mistakes
Forgetting to subtract the mean squared. \(\sigma^2 = \frac{\sum x^2}{n} - \bar x^2\): leaving out \(\bar x^2\) gives a value far too big.
Coding the spread with a shift. Adding or subtracting \(a\) does not change \(\sigma\); only the scale factor \(b\) does.
Class boundaries. Interpolate with the true class boundaries (for example 9.5 to 19.5 for 10–19 in whole numbers), not the stated limits.
Mixing up variance and standard deviation. Variance is in squared units; take its square root for the standard deviation.