Revision notes

Statistics: Practice

The key points, rules to remember, tables and diagrams from all 9 lessons, on one page for quick revision. To keep a copy, use your browser's Print and choose "Save as PDF".

Lesson 1

Averages at exam speed: corrections, combined means and missing frequencies

Total = n × mean. Combined mean = (n₁x̄₁ + n₂x̄₂)/(n₁ + n₂). Corrected mean = (old total − wrong value + right value)/n.

x510152025
f24p63
x123456
f357645

Key takeaways. Convert every mean into a total before doing anything else. Weight combined means by group size. For a raw-data median with an even count, use both middle positions. When solving for a missing frequency in the median formula, check afterwards that the median class you assumed is correct.

Read the full lesson

Lesson 2

Mean deviation: frequency tables and grouped data

x3691213152122
f34524543
cf37121418232730
Class0-1010-2020-3030-4040-5050-60
f68141642
x51525354555

Σ|x − a| is least when a is the median. So the mean deviation about the median is less than or equal to the mean deviation about any other point, including the mean.

Key takeaways. Use one fixed table layout every time. Find the centre the question asks for, not the one you prefer. Check signed deviations about the mean add to zero. Remember that the median gives the smallest mean deviation.

Read the full lesson

Lesson 3

Variance shortcuts and corrected values

Σx = n x̄ and Σx² = n(σ² + x̄²). Fix both totals, then x̄ = Σx/n and σ² = Σx²/n − x̄² again.

Class30-4040-5050-6060-7070-8080-9090-100
f371215832
u = (x − 65)/10−3−2−10123

Key takeaways. Move between (n, x̄, σ²) and (n, Σx, Σx²) freely. Correct both totals, then recompute with the new mean. Use step deviation with an assumed mean near the centre of large tables.

Read the full lesson

Lesson 4

Combined variance and finding unknown observations

Combined variance σ² = [n₁(σ₁² + d₁²) + n₂(σ₂² + d₂²)]/(n₁ + n₂), where d₁ = x̄₁ − x̄ and d₂ = x̄₂ − x̄. Equivalently, add the two Σx² totals and use Σx²/n − x̄².

Key takeaways. Combine groups through totals, or with the d² correction; never just average the variances. Two unknown observations come from a sum and a sum of squares, solved as a quadratic. Always check by recomputing.

Read the full lesson

Lesson 5

Coefficient of variation and changes of origin and scale

y = ax + b: ȳ = a x̄ + b, σ_y = |a|σ_x, σ_y² = a²σ_x², MD_y = |a|MD_x. CV_y = |a|σ_x/(a x̄ + b) × 100, so the CV survives a change of scale (a > 0) but not a change of origin.

Key takeaways. Under y = ax + b, averages follow ax + b and spreads scale by |a|. The CV is unit-free and survives multiplying by a positive constant, but it changes when a constant is added. Compare consistency by CV whenever the means differ.

Read the full lesson

Lesson 6

Averages in special settings: weighted means, data in AP and average speed

Weighted mean = Σwx/Σw. Equal distances: average speed = harmonic mean = 2uv/(u + v) for two legs. AP data: mean = median = (first + last)/2. Adding an observation: new total = old total + x; removing one: new total = old total − x.

Key takeaways: work with totals; weight the values when counts differ; use the harmonic mean for equal distances; use (first + last)/2 for data in AP.

Read the full lesson

Lesson 7

Relative measures of dispersion: coefficients of range, quartile deviation and mean deviation

Absolute measureCoefficient (relative measure)
Range = L − S(L − S)/(L + S), where L and S are the largest and smallest values
Quartile deviation = (Q₃ − Q₁)/2(Q₃ − Q₁)/(Q₃ + Q₁)
Mean deviation about amean deviation/a, where a is the mean or median used
Standard deviation σCV = (σ/mean) × 100%

Coefficient of range = (L − S)/(L + S). Coefficient of QD = (Q₃ − Q₁)/(Q₃ + Q₁). Coefficient of MD = MD/(the average used). Relative measures have no units; a change of unit (multiplying every value by k > 0) does not change them.

Key takeaways: dividing by a measure of centre removes the units. The coefficient of range is (L − S)/(L + S), the coefficient of quartile deviation is (Q₃ − Q₁)/(Q₃ + Q₁), and the coefficient of mean deviation is MD divided by the average used. A smaller coefficient means less relative spread.

Read the full lesson

Lesson 8

Deviation sums: using Σ(x − a) and Σ(x − a)² to find means and variances

x̄ = a + Σ(x − a)/n. σ² = Σ(x − a)²/n − (x̄ − a)². The deviations from the mean always add to zero. The sum of squares about the mean is the smallest sum of squares about any point.

Key takeaways: the mean is a + Σ(x − a)/n; the variance is Σ(x − a)²/n − (x̄ − a)². Deviations from the mean sum to zero, which lets you find a missing deviation.

Read the full lesson

Lesson 9

Mixed practice: statistics under time pressure

Question saysFirst step
wrong value correctedfix the total (and Σx² for the variance)
more consistentcompute both CVs
every value changed by y = ax + bmean: a·x̄ + b; σ: |a|σ; variance: a²σ²
mean deviation about the medianfind the median first, then the distances
groups combinedadd the totals, divide by the total number

Fix totals first. CV = σ/mean × 100. Under y = ax + b: mean a·x̄ + b, σ multiplied by |a|, variance by a². The combined mean is a weighted mean of the group means, weighted by group sizes.

Key takeaways: identify the tool from the wording, write down the totals or the formula, and check with one quick estimate, such as whether the answer lies between the group means or has the right size.

Read the full lesson

Privacy · Terms · Refunds · Delivery