Menu
Sign up Sign in

AI SL · 13.2

Probability Distributions

13.2.8

Normal probability calculations with the GDC

13.2.8

Normal probability calculations with the GDC

Target: I can find the probability that a normally distributed variable is above a value, below a value or between two values, using the calculator. (+4 more)

By the end of this section

Key points

Finding a normal probability

  • Write down the distribution first: \(X \sim N(\mu, \sigma^2)\), with what X measures.
  • The calculator's normal cdf needs four numbers: lower limit, upper limit, \(\mu\) and \(\sigma\).
  • For a tail with no limit, use a very large number such as 10^99 (for a lower tail, −10^99).
  • The 68–95–99.7 rule is only a check. Use the calculator for exact probabilities.

Watch out

  • In \(N(\mu, \sigma^2)\) the second number is the variance. \(N(120, 36)\) has \(\sigma = 6\), not 36.
  • For a continuous variable \(P(X < k) = P(X \leq k)\). The probability of one exact value is 0.
  • Lower and upper are in order: the smaller value goes in lower.
  • Sketch the curve and shade the region. The area you shade should match the size of your answer.

Is a normal model reasonable?

  • A normal distribution is symmetric. A box plot with a long upper whisker, or a median close to Q1, is skewed.
  • If the data is skewed, say so and say a normal model is poor.

On the calculator

Normal cdf

  • Choose the normal cdf from the distributions list.
  • Enter the lower limit, the upper limit, the mean \(\mu\) and the standard deviation \(\sigma\).
  • Read the probability and give 3 significant figures.

Screenshot and video to come

A tail with no limit

  • For \(P(X > k)\), enter k as the lower limit and a very large number, such as 10^99, as the upper limit.
  • For \(P(X < k)\), enter −10^99 as the lower limit and k as the upper limit.

Screenshot and video to come

Variance to standard deviation

  • If the question gives the variance, take its square root first.
  • Enter \(\sigma\), never the variance, in the calculator.

Screenshot and video to come

Explore: the normal distribution

A probability is an area under the curve. Drag the top of the curve to move the mean, the teal point on its side to change σ, and the orange points under the axis to choose the area. Or type exact values. Say how big the area is, then press Show answer.

Drag with the mouse or a finger, or select a point (Tab) and use the arrow keys. The axis fits itself to the curve when you let go.

Drag a point to change the shaded area.

  1. Drag the top of the curve, then the teal point. Which one moves the curve, and which one changes its shape? Why does a wider curve have to be lower?
  2. Press "Mark μ ± σ, 2σ, 3σ". Drag the orange points to μ − σ and μ + σ, then 2σ and 3σ. Say each area before you show it. Does it change when you change μ or σ?
  3. X ~ N(250, 16). What do you type for σ? Find P(X < 246).
  4. In "Find a value", choose Above and an area of 0.05. Is the value above or below the mean? Why does the calculator need the area 0.95?

Worked examples

Problems like the ones in class. Pick one, try it, then open the working.
Upper tail

The time a parcel takes to be delivered is normally distributed with mean 48 hours and standard deviation 9 hours. Find the probability that a parcel takes more than 60 hours.

Working
  1. Define X

    Let \(X\) be the delivery time in hours. \(X \sim N(48, 9^2)\)

    Write the distribution before you use the calculator.

  2. Upper tail

    From the calculator: normal cdf with lower = 60, upper = 10^99, μ = 48, σ = 9

    More than 60 has no upper limit, so use a very large number such as 10^99.

  3. Answer

    \(P(X > 60) = 0.0912\)

    About 9.1% of parcels take more than 60 hours.

Video walkthrough coming
Lower tail

The mass of flour in a bag is normally distributed with mean 1 000 g and standard deviation 12 g. Find the probability that a bag contains less than 985 g.

Working
  1. Define X

    \(X \sim N(1 000, 12^2)\), \(X\) in grams

    State the mean and the standard deviation.

  2. Lower tail

    From the calculator: normal cdf with lower = −10^99, upper = 985, μ = 1 000, σ = 12

    Less than 985 has no lower limit, so use a very large negative number such as −10^99.

  3. Answer

    \(P(X < 985) = 0.106\)

    For a continuous variable, P(X < 985) and P(X ≤ 985) are the same.

Video walkthrough coming
Between two values

The length of a fish of one species is normally distributed with mean 32 cm and standard deviation 4.5 cm. Find the probability that a fish is between 28 cm and 36 cm long.

Working
  1. Define X

    \(X \sim N(32, 4.5^2)\)

    Mean 32 and standard deviation 4.5.

  2. Between

    From the calculator: normal cdf with lower = 28, upper = 36, μ = 32, σ = 4.5

    Put the smaller value in lower and the larger in upper.

  3. Answer

    \(P(28 < X < 36) = 0.626\)

    The limits are symmetric about the mean, so this is also twice the area from 32 to 36.

Video walkthrough coming
Variance given

The volume of juice in a carton, \(X\) ml, is modelled by \(X \sim N(250, 16)\). Find the probability that a carton contains less than 246 ml.

Working
  1. Read the variance

    \(\sigma^2 = 16\), so \(\sigma = \sqrt{16} = 4\)

    The second number in N(250, 16) is the variance. The calculator needs the standard deviation.

  2. Lower tail

    From the calculator: normal cdf with lower = −10^99, upper = 246, μ = 250, σ = 4

    Enter σ = 4.

  3. Answer

    \(P(X < 246) = 0.159\)

    The standard deviation is 4, not 16.

Video walkthrough coming
Sketch and shade

The marks in a test are normally distributed with mean 62 and standard deviation 8. Sketch the normal curve, shade the region for P(X > 70) and find the probability.

24324048566472808896Mark
Working
  1. Sketch

    Bell curve centred at \(62\); mark \(62\) on the axis and \(70\) to the right.

    Shade to the right of 70. The shaded area is the probability.

  2. Calculator

    From the calculator: normal cdf with lower = 70, upper = 10^99, μ = 62, σ = 8

    The shaded region has no upper limit.

  3. Answer

    \(P(X > 70) = 0.159\)

    The shaded region is a small part of the curve, so a probability below 0.5 makes sense.

Video walkthrough coming
Judging normality

The box plot shows the number of minutes that 40 customers waited to be served. Is it reasonable to model the waiting time with a normal distribution?

Wait051015202530Waiting time (minutes)
Working
  1. Compare the halves

    Median − Q1 = 2 and Q3 − median = 6

    A symmetric distribution has the two halves of the box about equal.

  2. Compare the whiskers

    Lower whisker 2, upper whisker 18

    The upper tail is much longer: skewed to the right.

  3. Conclude

    Not reasonable.

    A normal curve is symmetric. A skewed box plot means a normal model is poor.

Video walkthrough coming