Measures of Spread (AI HL)

A measure of spread shows how widely data values are scattered, and the main ones are the interquartile range Q3−Q1Q_3-Q_1, the standard deviation and the variance. A value is an outlier if it lies more than 1.5×IQR1.5\times\text{IQR} beyond the nearest quartile. Adding a constant to every value leaves the standard deviation unchanged, while multiplying every value by k>0k>0 multiplies it by kk. It's part of the broader Statistics & Sampling topic.

12 questions on this sub-topic.

Syllabus

CodeSyllabus content
SL4.3Measures of central tendency (mean, median, mode), including estimating the mean of grouped data using mid-interval values and identifying the modal class. Measures of dispersion (interquartile range, standard deviation, variance) and quartiles of discrete data, calculated using technology, with awareness that different methods may give slightly different quartile values. The effect of constant changes (adding a constant, or scaling) on the mean and standard deviation of a data set.
SL4.1Concepts of population, sample, random sample, and discrete and continuous data. Reliability of data sources and bias in sampling. Interpretation of outliers - a data item more than 1.5×1.5\times IQR from the nearest quartile - and awareness that some outliers are valid while others may be recording errors. Sampling techniques (simple random, convenience, systematic, quota, stratified) and their effectiveness.

Official IB source: Mathematics: applications and interpretation subject brief (PDF, ibo.org).

Skills practised on this page:

IQR and outlier boundaries

Interquartile range

IQR=Q3−Q1\text{IQR} = Q_3 - Q_1

Q1Q_1 and Q3Q_3 come straight from your GDC's one-variable statistics screen - you're not expected to find them by hand once the data set is large.

Outlier boundaries

Q1−1.5×IQR  and  Q3+1.5×IQRQ_1 - 1.5\times\text{IQR}\ \ \text{and}\ \ Q_3 + 1.5\times\text{IQR}

A value is an outlier if it falls outside this range. This is standard technique that IB defines explicitly in the syllabus, rather than a formula-booklet entry.

Need the full syllabus wording and formula-booklet reference table? See Statistics & Sampling.

Key terms

What does a larger standard deviation tell you?

A larger standard deviation means the values are more spread out from the mean, and a smaller one means they are clustered closely around it. It is measured in the same units as the data.

e.g. Two classes both have a mean mark of 6060, with standard deviations 55 and 1515; the marks in the second class are more spread out.

How are variance and standard deviation related?

The variance is the square of the standard deviation, so the standard deviation is the square root of the variance. Both are found on the GDC.

e.g. A standard deviation of 44 gives a variance of 1616, and a variance of 2525 gives a standard deviation of 55.

What are the outlier boundaries?

The lower boundary is Q1−1.5×IQRQ_1-1.5\times\text{IQR} and the upper boundary is Q3+1.5×IQRQ_3+1.5\times\text{IQR}. A value beyond either boundary is an outlier.

e.g. If Q1=20Q_1=20 and Q3=30Q_3=30, then IQR=10\text{IQR}=10, so the boundaries are 20−15=520-15=5 and 30+15=4530+15=45.

How do constant changes affect the mean and standard deviation?

Adding a constant to every value adds the same constant to the mean but leaves the standard deviation unchanged. Multiplying every value by a positive constant kk multiplies both the mean and the standard deviation by kk.

e.g. A data set has mean 1010 and standard deviation 22. Adding 55 to every value gives mean 1515 and standard deviation 22; multiplying every value by 33 gives mean 3030 and standard deviation 66.

Worked examples

1
Easy
No calc
[4 marks]

A shop's monthly sales figures ($1000s) have Q1=70Q_1=70 and Q3=100.Q_3=100. One particular month had sales of $20,000. Using the 1.5×\timesIQR rule, determine whether this month should be treated as an outlier.

Worked solution

IQR=100−70=30.\text{IQR}=100-70=30. M1
Lower boundary =Q1−1.5×IQR=70−1.5(30)=70−45.=Q_1-1.5\times\text{IQR}=70-1.5(30)=70-45. A1
20<25,20<25, so the value lies below the lower boundary. R1
The month's sales of $20,000 are an outlier. A1

M1 IQR A1 Lower boundary = 25 R1 Compare $20,000 to the boundary of $25,000 A1 Correct conclusion
2
Hard
GDC
[5 marks]

The scores, out of 50, of 12 students on a test are:

5, 22, 25, 28, 30, 31, 33, 35, 36, 38, 40, 49

(a) Use your GDC to find the interquartile range.

(b) Find the lower and upper outlier boundaries.

(c) State which score(s), if any, are outliers.

Worked solution

(a) Q1=26.5, Q3=37.Q_1=26.5,\ Q_3=37. M1
IQR=37−26.5=10.5.\text{IQR}=37-26.5=10.5. A1

(b) Lower =26.5−1.5(10.5)=26.5−15.75,=26.5-1.5(10.5)=26.5-15.75, Upper =37+1.5(10.5)=37+15.75.=37+1.5(10.5)=37+15.75. M1
=10.75=10.75 and 52.75.52.75. A1

(c) Only 55 lies below 10.75;10.75; no score exceeds 52.75.52.75. So 55 is the only outlier. A1

One-variable statistics on the GDC - TI-84 STAT▸CALC▸1-Var Stats · Casio 1VAR · Nspire One-Variable Statistics.

M1 Attempt one-variable statistics on the GDC A1 Correct IQR of 10.5 M1 Apply Q1-1.5IQR and Q3+1.5IQR A1 Correct boundaries of 10.75 and 52.75 A1 Correctly identify 5 as the sole outlier
3
Medium
No calc
[4 marks]

A set of temperatures recorded in ∘^\circC has a mean of 18∘18^\circC and a standard deviation of 3∘3^\circC. Each temperature is converted to ∘^\circF using F=1.8C+32.F=1.8C+32.

(a)  Find the mean of the temperatures in ∘^\circF.

(b)  Find the standard deviation of the temperatures in ∘^\circF.

Worked solution

(a) F¯=1.8(18)+32.\bar{F}=1.8(18)+32. M1
F¯=64.4∘\bar{F} = 64.4^\circF. A1

(b) Adding a constant (32) does not change the spread; only the multiplier (1.8) scales the standard deviation. M1
New standard deviation =1.8(3)=5.4∘=1.8(3)=5.4^\circF. A1

M1 Apply F=1.8C+32 to the mean A1 Correct mean of 64.4 M1 Recognise only the multiplier scales the standard deviation A1 Correct standard deviation of 5.4

Common mistakes

  • Comparing the raw value to Q1Q_1 or Q3Q_3 instead of the boundary. A value below Q1Q_1 is not automatically an outlier - it has to fall below Q1−1.5×IQRQ_1 - 1.5\times\text{IQR}, which is often a good deal lower again.
  • Using the wrong quartile for the wrong boundary. The lower boundary subtracts 1.5×IQR1.5\times\text{IQR} from Q1Q_1; the upper boundary adds it to Q3Q_3. Mixing these up gives boundaries the wrong way round.
  • Reading Q1Q_1 and Q3Q_3 off the GDC incorrectly. Different calculator models label these slightly differently in the one-variable statistics output - always check you're reading Q1Q_1 and Q3Q_3, not the min/max or the median, before subtracting.

Using your GDC

Every step below is a real button sequence, not a vague "use your calculator" hint - covering the TI-84 Plus, TI-Nspire, and Casio fx-9860/fx-CG50. Pick your model to filter down to just the steps that apply to you.

Show steps for:
One-variable statistics (mean, median, standard deviation)

Instant summary statistics from a list - no formulas to compute by hand.

  1. Enter the data into a list.
  2. STAT → Edit → type values into L1. Then STAT → CALC → 1:1-Var Stats, choose L1, Calculate.TI-84
  3. Add a Lists & Spreadsheet page, name a column and enter data; then a Calculator page → menu → Statistics → Stat Calculations → One-Variable Statistics.Nspire
  4. Statistics menu → enter data in List 1 → CALC (F2) → 1-VAR.Casio
  5. Read x¯\bar x (mean), SxS_x (sample sd) or σx\sigma_x (population sd), and the five-number summary (min, Q1Q_1, median, Q3Q_3, max).
  6. For frequency data, put values in one list and frequencies in another and set the frequency list.

Tip: SxS_x vs σx\sigma_x: use σx\sigma_x (population) for a complete data set, SxS_x (sample) for a sample. IB usually wants σx\sigma_x.

Build and manipulate lists

Lists let you store data sets, generate sequences, and compute statistics without re-entering values - essential for grouped data and frequency tables.

  1. Enter data manually: go to the list editor and type values one by one.
  2. 2nd → STAT (LIST) → OPS → 5:seq(expression, variable, start, end) generates a list from a formula. For grouped data, put midpoints in L1 and frequencies in L2, then run 1-Var Stats L1, L2.TI-84
  3. In a Lists & Spreadsheet page, type a formula in the header row to fill the column automatically. Use two columns for midpoints and frequencies.Nspire
  4. In Statistics list editor, enter data directly; OPTN → LIST → Seq( builds a list from a formula. For frequency data, put midpoints in List 1 and frequencies in List 2.Casio
  5. For frequency data (grouped), enter midpoints in one list and frequencies in another, then reference both lists when running 1-Var Stats.

Tip: For cumulative frequency or working out Σfx\Sigma fx by hand - don't. Store midpoints in L1, frequencies in L2, and let 1-Var Stats do it all.

See the full GDC guide for more calculator models and topics.

Quick questions

How do you find the interquartile range?

IQR=Q3−Q1\text{IQR} = Q_3 - Q_1. On the IB course you find Q1Q_1 and Q3Q_3 using the one-variable statistics function on your GDC rather than by hand.

How do you test whether a value is an outlier?

A value is an outlier if it lies below Q1−1.5×IQRQ_1 - 1.5\times\text{IQR} or above Q3+1.5×IQRQ_3 + 1.5\times\text{IQR}.

Why is the interquartile range less affected by outliers than the standard deviation?

The interquartile range depends only on the quartiles, so a very large or very small value changes it little. The standard deviation uses every value, so an extreme value pulls it up.

Should an outlier always be removed from the data?

No. An outlier may be a valid value or a recording error, so it should only be removed if there is a reason to think it is a mistake. A student height of 17 m is clearly an error, but an unusually tall student is not.

Why do different calculators give slightly different quartiles?

There are several accepted methods for finding quartiles of discrete data, so results can differ slightly between technologies. Use the values from your own GDC, and do not mix methods within one question.

Ready to practise properly?

12 measures-of-spread questions, marked instantly like the real exam.