The normal distribution: mean, standard deviation and standardisation
The normal distribution is the most important continuous model in the IB exam: it describes heights, weights, times, measurement errors and marks. It is identified with just two numbers, the mean and the standard deviation , and it has the shape of a symmetrical bell. To work with it, it helps to know how to read the bell (where the centre is and how wide it spreads), to know the 68-95-99.7 rule for estimating probabilities at a glance, and to know how to standardise, that is, to convert a value into how many standard deviations it lies from the mean.
What a normal distribution is
A continuous random variable follows a normal distribution when its density function has the shape of a bell. It is written
where is the mean and is the standard deviation. Here is the standard deviation, not the variance: in the standard deviation is 4 and the variance is 16.
For example, the time a person takes to walk to school can be modelled by , in minutes: the most common value is about 15 minutes, and it is very rare to take less than 5 or more than 25.
The two parameters move the bell in different ways:
- Changing shifts the bell to the right or to the left, without changing its shape.
- Changing changes the width: a small gives a tall, narrow bell (the data cluster near the mean); a large gives a low, wide bell (the data are more spread out). The area is still 1.
The 68-95-99.7 rule
There are three probabilities that recur in every normal distribution, whatever and are. They tell you what proportion of the data lies within 1, 2 and 3 standard deviations of the mean.
Symmetry lets you find other probabilities without a calculator. For example, the central 68% leaves out 32%, shared equally between the two tails: .
Standardisation: the variable
To compare values from different normal distributions, or to use a single table of probabilities, each value is measured with a common rule: how many standard deviations it lies from the mean.
For example, if , the value has : it is 2 standard deviations above the mean. And the value has .
Standardising is useful for two things in the exam. The first is to compare results that come from different distributions: the value with the larger is the one that is relatively higher. The second is to read the 68-95-99.7 rule: , and are exactly the positions , and .
Worked examples
Plants in a greenhouse
The height, in centimetres, of tomato plants in a greenhouse follows a normal distribution .
a) What is the value of ? Justify it without a calculator.
b) Find with the GDC, to three significant figures.
c) Find with the GDC, to three significant figures.
Solution
a) The bell is symmetrical about the mean , so half of the area lies to the right:
b) The interval is , because and . By the 68-95-99.7 rule the probability should be around 0.95. The GDC gives:
c) By symmetry, the probability left outside is shared equally between the two tails: . The GDC gives a more precise value:
Standardising the marks of a test
The marks of a level test, out of 100 points, follow a normal distribution .
a) Calculate the standardised value of a mark of 88 points.
b) Calculate the standardised value of a mark of 60 points.
c) Find the mark that corresponds to .
d) Find with the GDC, to three significant figures, and interpret it.
e) Find with the GDC, to three significant figures.
Solution
a) . A mark of 88 is 2 standard deviations above the mean.
b) . A mark of 60 is one and a half standard deviations below the mean.
c) Solving for : .
d) As , this probability coincides with in the standard normal. The GDC gives: Interpretation: approximately 97.7% of the people who take the test score less than 88 points. This is consistent with the 68-95-99.7 rule: below lies the central 95% plus half of the remaining 5%, that is, roughly 97.5%.
e) Enter the limits 60 and 80, the mean 72 and the standard deviation 8 (not 64). The GDC gives:
Which subject does she stand out in more?
In an assessment, the Mathematics marks follow a normal distribution and the Physics marks a normal distribution . A student has scored 68 in Mathematics and 75 in Physics.
a) Calculate the standardised value of each of her two marks.
b) Find the probability that a student chosen at random scores less than her in Mathematics, and the probability that a student scores less than her in Physics. Give both to three significant figures.
c) In which subject has she stood out more relative to her group? Answer with the word "Matematicas" or "Fisica" (as written).
d) Find the probability that a student scores between 48 and 72 in Mathematics, to three significant figures, and explain why it coincides with a value from the 68-95-99.7 rule.
Solution
a) Each mark is standardised with its own mean and standard deviation:
b) With the GDC, entering the mean and standard deviation of each subject: That is, she outperforms 90.9% of the group in Mathematics and 84.1% in Physics.
c) In Physics she is 10 points above the mean and in Mathematics only 8, but the Physics marks are more spread out ( compared with ). If you compare in standard deviations, : her Mathematics mark is relatively higher up. She has stood out more in Mathematics, and the percentage of students she beats (90.9% compared with 84.1%) confirms it.
d) The limits are and , that is, . The GDC gives: which is the precise value of the 95% rule.
In the exam: when you are asked to compare two values from different distributions, comparing their differences from the mean is not enough: you must calculate the standardised values (or the probabilities), compare those numbers and conclude in the context of the subjects. Always write down which mean and standard deviation you used in each calculation.
Common mistakes
Summary
- : a bell curve symmetrical about , with the standard deviation (not the variance); the total area under the curve is 1.
- Mean, median and mode coincide at ; ; because it is continuous.
- Changing shifts the bell; a larger makes it wider and lower.
- 68-95-99.7 rule: within , and of the mean lie approximately 68%, 95% and 99.7% of the values (more precisely: , , ).
- Standardising: indicates how many standard deviations lies from the mean; ; and the probabilities are preserved.
- To compare values from different distributions, compare their -values.
Want to work on it together?
Free first consultation: we look at your level and your exam, and build a plan. Álvaro replies on WhatsApp.
Message us on WhatsApp