Binomial distribution: notes and practice questions
- A discrete random variable's value depends on a random event, unknown until carried out.
- A discrete probability distribution lists all possible discrete outcomes and their probabilities.
- The Binomial Distribution counts "successes" in a fixed number of independent trials.
- Notation: .
- Valid values for are integers: .
- Conditions for a Binomial Model:
- A defined "Trial" (repeated action).
- A defined "Success" (desired outcome).
- Fixed number of trials ().
- Fixed probability of success () for each trial.
- Independent trials.
- Binomial Probability Formula (exact successes):
for
- Combinations formula:
- Expected Value (Mean):
- Variance:
- To set up a binomial model: Identify , , define in context, state .
- Use GDC's Binomial Cumulative Distribution (Cdf) for probabilities.
- GDC input for Cdf: Lower value, Upper value, , .
- For : GDC Lower=, Upper=.
- For : GDC Lower=, Upper=.
- For : Use GDC's Binomial Probability Distribution (Pdf) or Cdf with Lower=, Upper=.
- Adjust strict inequalities for GDC: becomes .
- State contextual assumptions for the binomial model (e.g., independence, constant ).
How it is examined
Almost always a GDC question: identify and , decide between "exactly", "at most" and "at least", then read off the value. The "at least" case needs and the off-by-one is the classic error. Justifying that a binomial model is appropriate (fixed trials, two outcomes, constant probability, independence) is a separate mark that students skip.
The binomial distribution with its mean and variance .
- The binomial distribution.
- The mean and variance of the binomial distribution.
Not required: formal proof of mean and variance.
Linking questions
- Aim 8: Pascal's triangle, and attributing the origin of a mathematical discovery to the wrong mathematician.
- International-mindedness: the so-called "Pascal's triangle" was known to the Chinese mathematician Yang Hui much earlier than Pascal.
- TOK: what criteria can we use to decide between different models?
- Enrichment only, so not examinable at SL: hypothesis testing using the binomial distribution. It becomes examinable at HL under AHL 4.18.
Practice questions
37 questions · 28 medium · 9 hardQuestion 1
MediumPaper 1 · calculator8 marksA renowned artisanal bakery claims that only 5% of its specialty sourdough loaves have minor cosmetic imperfections (e.g., slight cracks, uneven browning). A local restaurant owner, who regularly purchases these loaves, decides to test this claim. For their latest delivery, the owner inspects a batch of 150 loaves and finds 12 loaves with cosmetic imperfections.
(a) Identify the type of sampling used by the restaurant owner.
(b) State the null and alternative hypotheses for this test.
(c) Calculate the p-value for this hypothesis test, assuming cosmetic imperfections occur independently.
(d) The restaurant owner performs the test at the 5% significance level. State the conclusion of the test, giving a reason.
Consider how the sample of loaves was chosen for inspection.
The null hypothesis represents the bakery's claim, while the alternative hypothesis reflects the restaurant owner's suspicion (that the proportion might be higher).
This is a binomial distribution problem. You need to calculate the probability of observing 12 or more imperfect loaves out of 150, given the null hypothesis.
Compare your calculated p-value with the given significance level to determine whether to reject or fail to reject the null hypothesis.
Question 2
HardPaper 1 · calculator6 marks(a) A popular cafe uses a specialty espresso machine that has an 85% probability of functioning correctly on any given morning. The probability of the machine functioning correctly on a given morning is independent of it functioning on any other morning.
Find the probability that the espresso machine functions correctly on exactly four mornings in a particular 5-day workweek.
(b) If the specialty espresso machine does not function correctly, the cafe owner must decide whether to use a backup machine or close early for the day. The cafe owner closes early if the specialty machine does not function correctly AND the backup machine is not available. There is a 70% probability that the backup machine is available on any given morning, independent of the specialty machine's status or any other morning.
Find the probability that the cafe owner does not have to close early in a particular 5-day workweek.
Recall the formula for binomial probability: . Identify the number of trials (n), the probability of success (p), and the number of successes (k) for this scenario.
First, calculate the probability of closing early on a single day. This happens if the specialty machine fails AND the backup machine is unavailable. Then, find the probability of not closing early on a single day. Finally, use the binomial distribution for the 5-day workweek.
Question 3
MediumPaper 1 · calculator7 marksA manufacturing plant produces a specific electronic component. Due to the complexity of the manufacturing process, there is a 15% chance that any given component will be defective. A batch of 12 components is randomly selected for quality control inspection. Each component's defect status is independent of others.
(a) Calculate the expected number of components in the batch that are non-defective.
(b) Calculate the probability that exactly 3 components in the batch are defective.
(c) Determine the probability that more than 9 components in the batch are non-defective.
Recall the formula for the expected value of a binomial distribution. First, identify the probability of a component being non-defective.
This is a binomial probability problem. Identify the number of trials (n), the number of successes (k), and the probability of success (p) for a defective component.
Consider the number of non-defective components. 'More than 9' means 10, 11, or 12 non-defective components. You can use the binomial cumulative distribution function (CDF) or calculate individual probabilities and sum them.
Question 4
HardPaper 1 · calculator12 marksThe lifespan of a certain brand of smartphone battery is normally distributed with a mean of days and a standard deviation of days.
(a) The probability that a randomly selected battery lasts less than days is . Find the value of .
(b) A battery is randomly selected. It is known that the battery lasts longer than days. What is the probability that it lasts longer than days?
(c) A retailer orders batteries. What is the probability that at most of them have a lifespan less than days?
For a normally distributed variable with mean and standard deviation , if you are given and need to find , you should use the inverse normal function on your GDC. Remember to input the area, mean, and standard deviation correctly.
This is a conditional probability problem. Recall the formula . In this case, is the event that the battery lasts longer than days, and is the event that it lasts longer than days. Since lasting longer than days implies lasting longer than days, simplifies to .
This problem involves a large number of trials, suggesting a binomial distribution that can be approximated by a normal distribution. First, calculate the probability of a single battery having a lifespan less than days using the normal distribution. Then, use this probability to define the parameters (mean and variance) of the normal approximation to the binomial distribution. Remember to apply a continuity correction when approximating a discrete distribution with a continuous one.
Question 5
MediumPaper 1 · calculator6 marksLeo is a student who takes a short mathematics quiz every day for five days each week. He has a 70% probability of passing any given quiz. The outcome of one quiz is independent of any other quiz.
(a) Find the probability that Leo passes exactly three quizzes in a particular five-day week.
Leo studies for his quiz with a 60% probability on any given morning, independent of whether he passes the quiz or not. Leo is considered 'well-prepared' for a quiz if he passes the quiz AND he studied for it.
(b) Find the probability that Leo is not well-prepared on any day in a particular five-day week.
This problem involves a fixed number of trials (quizzes), each with two possible outcomes (pass or fail), and a constant probability of success. Consider which probability distribution is appropriate for this scenario.
First, calculate the probability that Leo is 'well-prepared' on a single day. Then, find the probability that he is 'not well-prepared'. Finally, use this probability to determine the chance of this event happening for all five days.
Question 6
HardPaper 2 · calculator16 marksA factory produces specialized electronic components. The total "quality score" of a component, , is a combination of scores from three independent inspection stages:
- Stage 1: Automated visual inspection. The score from this stage has an expectation of and a standard deviation of .
- Stage 2: Manual functional test. A batch of critical functions are tested, and the score is the number of functions that pass. Each function has a probability of passing, independently.
- Stage 3: Environmental stress test. The score from this stage, representing the number of successful stress cycles, follows a Poisson distribution with a mean of .
The overall quality score for a component is given by .
Calculate the expected value and variance of the total quality score .
Given that the distribution of can be approximated by a Normal distribution, find the probability that a randomly selected component has a total quality score between and (inclusive of , exclusive of ).
The factory manager wants to ensure that the mean total quality score of a sample of components is within units of the true mean, with a probability of at least . Find the minimum sample size required.
Recall the formulas for the expectation and variance of Binomial and Poisson distributions. For independent random variables and constants , and . Remember that .
When approximating a discrete distribution with a continuous Normal distribution, remember to apply a continuity correction. For , the continuous approximation would be .
The Central Limit Theorem states that for a sufficiently large sample size , the sample mean is approximately normally distributed with mean and variance . You will need to use the inverse normal function to find the critical Z-value for the given probability.
Question 7
MediumPaper 1 · calculator8 marksA manufacturing company produces electronic components. Historically, the defect rate for a specific component has been 15%. A new production line is implemented, and the quality control manager wants to test if the defect rate has decreased. She assumes that the defect status of each component is independent of others.
(a) Write down suitable hypotheses for this test.
(b) The quality control manager decides to take a random sample of 120 components. She will reject the null hypothesis if fewer than 12 components are found to be defective.
Find the probability that she makes a Type I error.
(c) In fact, the new production line successfully reduced the defect rate to 10%.
Find the probability that she makes a Type II error.
Remember to define your parameter (e.g., p for probability) and state both the null and alternative hypotheses using appropriate notation. Consider what 'decreased' implies for the alternative hypothesis.
A Type I error occurs when you reject the null hypothesis () when it is actually true. Use the binomial distribution with the parameters defined by and the rejection region given.
A Type II error occurs when you accept the null hypothesis () when it is false. This means using the true defect rate (10%) and the acceptance region (the complement of the rejection region from part (b) ).
Question 8
HardPaper 2 · calculator16 marksA factory produces electronic components. A quality control inspector randomly selects a batch of components and checks for defects. This process is repeated for batches. The number of defective components in each batch is recorded as shown in the table below.
| Number of defects () | ||||
|---|---|---|---|---|
| Frequency |
The manager claims that the number of defective components in a batch of follows a binomial distribution .
Show that the expected frequency for a batch having defective components is .
Find the table of expected frequencies for the number of defective components in batches, assuming the binomial distribution .
State the null and alternative hypotheses for a goodness-of-fit test. Justify any necessary adjustments to the categories for the test and state the degrees of freedom.
Calculate the Chi-squared test statistic for this data, using the adjusted categories and assuming a significance level. The critical value for this test is .
State the conclusion for the test, justifying your answer.
To find the expected frequency, first calculate the probability of getting 0 defective components using the binomial probability formula . Then multiply this probability by the total number of batches.
Calculate the binomial probability for and defects, then multiply each probability by the total number of batches () to get the expected frequencies.
Remember the conditions for a Chi-squared goodness-of-fit test, particularly regarding expected frequencies. The degrees of freedom for a goodness-of-fit test are , where is the number of categories and is the number of parameters estimated from the data.
Use the formula with the combined observed and expected frequencies. Remember to use the combined categories: .
Compare your calculated Chi-squared test statistic with the given critical value. If the test statistic is less than the critical value, you do not reject the null hypothesis.
Question 9
MediumPaper 1 · calculator6 marksA company manufactures microchips. On average, 1 in 20 microchips fails the final quality control test. The failure of any microchip is independent of others.
In a randomly selected batch of 30 microchips, calculate the probability that exactly three microchips fail quality control.
In a randomly selected batch of 30 microchips, calculate the probability that more than two microchips fail quality control.
The company sells functional microchips for 20 each.
Calculate the expected revenue, in dollars, from selling a randomly selected batch of 30 microchips.
This scenario involves a fixed number of trials (microchips), two possible outcomes (fail/pass), a constant probability of failure, and independent trials. Which probability distribution models this situation?
To find the probability of 'more than two', consider the complement event. How can you express 'more than two' in terms of cumulative probabilities?
First, calculate the expected number of functional microchips and the expected number of failed microchips in the batch. Then, multiply these expected numbers by their respective selling prices and sum them up.
Question 10
HardPaper 1 · calculator9 marksA call centre receives calls at an average rate of calls per minute during a specific period of the day.
The number of calls received in a given minute can be modelled by a Poisson distribution.
(a) Find the probability that the call centre receives exactly one call in a randomly selected minute.
(b) Find the probability that the call centre receives at least one call in a randomly selected minute.
The call centre supervisor observes the calls received over a period of independent minutes.
(c) Find the probability that a total of six calls are received during these minutes.
(d) Find the probability that exactly one call is received in each of these minutes.
(e) Find the probability that at least one call is received in exactly of the minutes.
For a Poisson distribution with mean , the probability of observing events is given by .
The probability of 'at least one' is minus the probability of 'zero'.
If the mean rate is per minute, the mean rate over minutes is . The number of calls over minutes still follows a Poisson distribution.
Since the minutes are independent, you can multiply the probabilities of the individual events.
This involves a binomial distribution. First, identify the probability of success (at least one call in a minute) and the number of trials.
Question 11
MediumPaper 1 · calculator7 marks9. [Maximum mark: 7]
A factory produces electronic components. Historical data shows that 15% of the components produced are defective.
A quality control inspector randomly selects a batch of 25 components for testing.
As part of its quality control, the factory uses the model , where is the number of defective components in a randomly selected batch.
(a) Calculate the
(i) mean of .
(ii) variance of .
(b) Find the probability that
(i) exactly 3 components are defective.
(ii) fewer than 5 components are defective.
(c) State one assumption that the factory makes in using this model.
For a binomial distribution , the mean is given by the formula .
For a binomial distribution , the variance is given by the formula .
Use the binomial probability mass function (PMF) or a GDC's binomial PD function.
This means , which is equivalent to . Use the binomial cumulative distribution function (CDF) or sum individual probabilities.
Consider the conditions required for a binomial distribution model to be appropriate.
Question 12
HardPaper 2 · calculator14 marksThe lifespan of a new model of LED light bulb, , is normally distributed with a mean of hours and a standard deviation of hours.
Sketch a diagram showing this information.
Find the proportion of these LED bulbs that have a lifespan between hours and hours.
A large batch of LED bulbs is purchased. Determine the expected number of bulbs in this batch that will have a lifespan of less than hours.
It is observed that of the LED bulbs last longer than hours. Estimate the value of .
Ten of these LED bulbs are chosen at random. Find the probability that exactly two of them last longer than hours.
Draw a bell-shaped curve. Label the mean at the center and indicate the standard deviation by marking points on the horizontal axis at one standard deviation away from the mean.
Use your GDC's normal CDF function. Remember to input the lower bound, upper bound, mean, and standard deviation.
First, calculate the probability that a single bulb lasts less than hours using the normal CDF. Then, multiply this probability by the total number of bulbs in the batch.
Since last longer than , this means last less than . Use the inverse normal function on your GDC with a cumulative probability of .
First, calculate the probability that a single bulb lasts longer than hours. Then, recognize that this is a binomial distribution problem with trials and the probability you just calculated as . Use the binomial probability formula or GDC function for exactly successes.
Question 13
MediumPaper 1 · calculator8 marksAlex is taking a multiple-choice quiz with questions. For each question, there are four possible answers, and Alex guesses randomly. The probability that Alex answers any single question correctly is .
(a) Find the probability that Alex answers at least one question correctly.
(b) Find the probability that Alex answers exactly one question correctly.
(c) Find the probability that Alex answers at least two questions correctly.
Consider the complementary event. What is the probability that Alex answers zero questions correctly? Use the binomial probability formula .
Use the binomial probability formula with .
You can calculate by summing , or by using the complementary event: .
Question 14
HardPaper 2 · calculator14 marksThe diameter of a certain type of industrial component is modelled by a normal distribution with a mean of mm and a standard deviation of mm.
Find the probability that a randomly selected component has a diameter less than mm.
Find the probability that a randomly selected component has a diameter greater than mm.
Assuming the diameters of components are independent, find the probability that two consecutive components both have a diameter greater than mm.
A component is classified as 'premium' if its diameter is between mm and mm. A batch of three components is considered 'high quality' if all three components in the batch are premium. Find the probability that a randomly selected batch is NOT high quality.
In a production run, batches of three components are produced. Find the probability that at least of these batches are high quality.
Find the probability that between and (exclusive of ) of these batches are high quality.
Given that at least batches are high quality, find the probability that less than batches are high quality.
Use the normal cumulative distribution function (CDF) on your GDC. Remember that is directly calculated by the CDF.
Use the normal cumulative distribution function (CDF) or the normal survival function (SF) on your GDC. Remember that .
For independent events and , the probability of both occurring is .
First, find the probability that a single component is 'premium'. Then, use this to find the probability that a batch of three is 'high quality'. Finally, calculate the complementary probability.
This scenario involves a fixed number of trials (batches), each with two possible outcomes (high quality or not), and the trials are independent. This suggests a binomial distribution. Remember .
This means finding , which is equivalent to .
This is a conditional probability problem: . Here, is 'less than batches are high quality' and is 'at least batches are high quality'. The intersection is 'between and (exclusive of ) batches are high quality'.
Question 15
MediumPaper 1 · calculator11 marksA factory produces electronic components. A quality control inspector randomly selects batches of components and records the number of defective items. Over a month, such batches were inspected. The observed frequencies are shown in the table below:
| Number of defective items | or more | ||
|---|---|---|---|
| Frequency |
It is assumed that the number of defective items in a batch follows a binomial distribution , where is the probability of a single component being defective.
(a) Calculate the expected frequencies for obtaining , , and or more defective items in a batch of , based on the binomial distribution and a total of batches.
(b) Perform a goodness of fit test at the significance level to determine if the observed data fits the assumed binomial distribution. You should state the null and alternative hypotheses, calculate the test statistic, and justify your conclusion. The critical value for this test is .
Recall the binomial probability formula . Calculate the probabilities for and . For '2 or more', calculate . Then, multiply these probabilities by the total number of batches () to find the expected frequencies.
Remember to state your null and alternative hypotheses clearly. Use the formula to calculate the test statistic. The degrees of freedom for this test are , where is the number of categories, since the parameter was given and not estimated from the data. Compare your calculated value with the critical value to draw a conclusion in context.
Question 16
HardPaper 2 · calculator14 marks(a) The number of customer service emails received by 'TechSolutions' per hour follows a Poisson distribution with a mean of . Using this model, find the probability that TechSolutions receives exactly emails in a given hour.
(b) Over a period of consecutive hours, find the probability that TechSolutions receives:
(i) exactly emails.
(ii) emails during the first and third hour only (i.e., at least one email in the first hour, no emails in the second hour, and at least one email in the third hour).
(c) Over a working day of hours, find the probability that there are exactly hours during which TechSolutions receives no emails.
(d) TechSolutions expands its operations and opens a new department, 'SupportPlus', which also receives emails. The number of emails received by each 'SupportPlus' agent per hour follows a Poisson distribution with a mean of . Assuming these are independent of the main TechSolutions department and each other, determine the least number of SupportPlus agents required so that the total probability of receiving at least emails across all departments (main TechSolutions and the SupportPlus agents) in an hour is greater than .
Recall the probability mass function for a Poisson distribution: . Identify the mean and the specific number of events from the question.
When combining independent Poisson processes, their means add up. Calculate the new mean for the -hour period and then apply the Poisson probability mass function.
This involves combining probabilities of independent events. Calculate and for a single hour, then multiply these probabilities for the specific sequence of events.
This scenario can be modeled by a binomial distribution. First, find the probability of 'no emails' in a single hour using the Poisson distribution. This will be the 'success' probability for your binomial model.
Let be the number of SupportPlus agents. The total mean number of emails per hour will be the sum of the mean from TechSolutions and times the mean from each SupportPlus agent. You need to find the smallest integer such that . You can do this by iterating through values of or by using a GDC's table/graphing function.
Question 17
MediumPaper 2 · calculator15 marksA factory produces light bulbs, and it is known that the probability of a light bulb being defective is . A quality control inspector takes a sample of light bulbs from a production line and records the number of defective bulbs. This process is repeated times, and the observed frequencies are shown in the table below.
| Number of defective bulbs | ||||
|---|---|---|---|---|
| Frequency |
Show that the expected frequency for getting defective bulbs in a sample of is .
Find the table of expected frequencies, pooling categories as necessary to ensure all expected frequencies are at least .
State the null and alternative hypotheses for a goodness of fit test.
Write down the number of degrees of freedom.
Calculate the test statistic.
The critical value for this test at the significance level is .
State the conclusion for the test and give a reason for your answer.
Recall the formula for binomial probability . The expected frequency is the total number of trials multiplied by the probability of the event.
Calculate the probability for each number of defective bulbs (). Then multiply by the total number of samples to get expected frequencies. If any expected frequency is less than , combine it with an adjacent category.
The null hypothesis () assumes the data fits the specified distribution, while the alternative hypothesis () suggests it does not.
Degrees of freedom for a goodness of fit test are , where is the number of categories and is the number of parameters estimated from the data. In this case, the probability is known, so no parameters are estimated.
Use the formula , where are the observed frequencies and are the expected frequencies from part (b). Remember to use the pooled observed frequencies for defective bulbs.
Compare your calculated test statistic with the given critical value. If the test statistic is less than the critical value, you do not reject the null hypothesis.
Question 18
HardPaper 2 · calculator15 marksThe 'SoundWave' concert hall has a capacity of seats. Historical data shows that of ticket holders attend the concert. The management decides to sell tickets, hoping that no more than concertgoers will arrive.
The number of concertgoers who arrive is assumed to follow a binomial distribution with a probability of .
(a) Calculate the probability that more than concertgoers arrive for the performance.
(b) (i) Write down the expected number of concertgoers who will arrive if tickets are sold.
(ii) Find the maximum number of tickets that could be sold if the expected number of concertgoers who arrive must be less than or equal to .
Each ticket costs 200 in compensation to each concertgoer who cannot be seated.
(c) Find, to the nearest integer, the expected increase or decrease in the money made by the venue if they decide to sell tickets rather than .
For a binomial distribution , the probability of successes is given by . To find the probability that more than concertgoers arrive, consider the probabilities for concertgoers, or use the complement rule . You will need a GDC for this calculation.
The expected value of a binomial distribution is given by .
Let be the number of tickets sold. The expected number of attendees is . Set up an inequality with this expected value and the capacity, then solve for . Remember that the number of tickets must be an integer.
Calculate the total expected money made for each scenario (selling tickets and selling tickets). For the -ticket scenario, you need to consider the income from all tickets sold and subtract the expected compensation. The expected compensation is calculated by summing the products of the probability of a specific number of overbooked attendees and the corresponding compensation amount.
Question 19
MediumPaper 1 · calculator5 marksThe volume of liquid in bottles produced by a beverage company is normally distributed with a mean of mL and a standard deviation of mL. A bottle is considered 'underfilled' if its volume is less than mL.
(a) Find the percentage of bottles that are classified as underfilled.
Bottles are packed into crates, with each crate containing bottles.
(b) Find the probability that a random crate has at most two underfilled bottles.
For part (a), you need to calculate the Z-score for mL and then use the normal cumulative distribution function (CDF) to find the probability. Remember to convert the probability to a percentage.
For part (b), use the probability of an underfilled bottle found in part (a). This is a binomial distribution problem. Let be the number of underfilled bottles in a crate. You need to find .
Question 20
MediumPaper 2 · calculator12 marks(a) Show that there are factors of .
(b) A player spins a "Factor Frenzy" spinner times. The spinner has numbers from to . A player scores a "bonus point" if the number spun is a factor of .
(i) Find the probability that the player scores exactly three bonus points.
(ii) Find the probability that the player scores at least three bonus points.
(iii) Find the probability that the player scores at most three bonus points.
(iv) Find the probability that the player scores bonus points on the first three spins, and does not score a bonus point on the fourth spin.
To find the number of factors of an integer, first find its prime factorization. If a number is expressed as , the number of factors is .
This is a binomial probability problem. Identify the number of trials (), the probability of success (), and the number of successes (). The probability of success is the number of factors of 96 divided by 96.
For 'at least three', calculate . Use your GDC's binomial cumulative distribution function.
For 'at most three', calculate . Use your GDC's binomial cumulative distribution function.
Consider the probability of a specific sequence of independent events. A bonus point is a 'success' (probability ), and not scoring a bonus point is a 'failure' (probability ).
No question on this page matches those filters. Try another difficulty or paper.
17 more Binomial distribution questions in the app
Every answer is marked mark by mark, IB-style, and the AI tutor helps when you are stuck.
Where marks are lost
- Using your own wrong value after failing a "show that." All follow through is withdrawn for the rest of that question.