A researcher investigates whether a new fertilizer increases the mean height of a species of plant. Under the null hypothesis, simulated random samples are generated. Of these, produce an increase at least as large as the increase observed in the researcher's sample.
Estimate the -value from the simulation.
State the decision at the significance level and interpret it in context.
Explain what the simulations represent.
0
A company compares the population mean battery lives of two brands. A two-tailed test gives a -value of .
State the hypotheses, using and for the two population means.
State the decision at the significance level.
State the decision at the significance level.
Explain why the -value is not the probability that the null hypothesis is true.
0
A random sample of commuters is classified by their usual method of travel and whether they live in an urban or rural area. The results are shown in the table. A test is used to determine whether usual method of travel is independent of area of residence.
Area of residence | Car | Public transport | Bicycle | Total |
|---|---|---|---|---|
Urban | 20 | 25 | 25 | 70 |
Rural | 30 | 15 | 15 | 60 |
Total | 50 | 40 | 40 | 130 |
State the null and alternative hypotheses.
Calculate the expected frequency of urban commuters who travel by car.
The test gives and . Determine whether usual method of travel is independent of area of residence at the significance level. Justify your answer.
0
A manufacturer claims that packets of fruit sweets contain red, orange, yellow and green sweets in the proportions , , and , respectively. A random sample of sweets is classified by colour. The observed frequencies are shown in the table.
Colour | Observed frequency |
|---|---|
Red | 30 |
Orange | 32 |
Yellow | 34 |
Green | 24 |
Total | 120 |
Calculate the expected frequency of green sweets under the manufacturer's claim.
Calculate the value of the test statistic.
Find the -value and state whether the data are consistent with the manufacturer's claim at the significance level.
0
Visitors to a nature reserve are classified by age group and by the activity they choose. The observed frequencies are shown in a contingency table. A test for independence gives . At the significance level, the relevant critical value is .
Age group | Hiking | Cycling | Wildlife watching | Total |
|---|---|---|---|---|
18-30 | 32 | 20 | 20 | 72 |
31-50 | 17 | 15 | 16 | 48 |
51+ | 11 | 25 | 24 | 60 |
Total | 60 | 60 | 60 | 180 |
State the alternative hypothesis.
Calculate the expected frequency for the specified cell with row total , column total and grand total .
Write down the number of degrees of freedom and determine the conclusion of the test.
0
A coach compares the recovery times of athletes using treatment A with those using treatment B. The coach claims that the population mean recovery time for treatment A is greater than that for treatment B. A pooled two-sample -test, calculated as , gives with degrees of freedom. The calculator displays a two-tailed -value of .
State the null and alternative hypotheses.
Find the -value appropriate for the coach's claim.
Determine the conclusion at the significance level.
0
A streaming service predicts that viewers choose four film genres in the proportions , , and . In a random sample of viewers, the observed frequencies are , , and , respectively. A goodness-of-fit test is performed at the significance level. The critical value is .
Calculate the expected frequency for the first genre.
Calculate the contribution of the first genre to the statistic.
The complete test statistic is . Determine whether the predicted genre distribution is a satisfactory fit.
0
The mass of cereal in a box is normally distributed with known population standard deviation . The labelled population mean is . A random sample of boxes has mean mass . A consumer organization tests whether the population mean mass is greater than the labelled value.
State the hypotheses.
Calculate the value of the test statistic.
Find the -value and state the conclusion at the significance level.
0
A manufacturer claims that the population mean operating time of a sensor is hours. The population standard deviation is unknown. A random sample of sensors has mean operating time hours and sample standard deviation hours. Operating times are assumed to be normally distributed. An engineer tests whether the population mean exceeds hours.
Explain why a -test, rather than a -test, is appropriate.
Calculate the test statistic.
Find the -value and state the conclusion at the significance level.
0
A basketball player claims that the probability of scoring a free throw is greater than . During independent free throws, the player scores times. The probability of scoring is assumed to remain constant.
State the hypotheses and the distribution used under the null hypothesis.
Calculate the -value.
State the conclusion at the significance level.
0
Two independent groups of students use different revision methods. Their scores on the same test are summarized below. Scores in both populations are assumed to be normally distributed with equal unknown variances.
Method A: , ,
Method B: , ,
A teacher wishes to test whether the population mean scores for the two methods are different.
State the hypotheses for the test.
Carry out an appropriate test and find the -value.
State the conclusion of the test at the significance level.
0
Eight participants complete a task before and after a training session. For each participant, the difference is defined as
The observed differences are . The differences are assumed to come from a normal population.
State suitable hypotheses to test whether the training increases the population mean score.
Explain why the data should be analysed using a one-sample test rather than an independent two-sample test.
Carry out an appropriate test and state the conclusion at the significance level.
0
The number of faults recorded by a monitoring system in one hour is modelled by a Poisson distribution. The usual population mean is faults per hour. During eight independent hours after a software update, a total of faults is recorded. An engineer tests whether the update has increased the mean fault rate.
State the hypotheses and the distribution of the total number of faults under the null hypothesis.
Calculate the -value.
State the conclusion at the significance level.
Describe a Type I error in this context.
0
For a random sample of cities, a researcher records annual public transport expenditure per resident and the percentage of journeys made by public transport. The sample product moment correlation coefficient is . The paired variables are assumed to follow a bivariate normal distribution.
State hypotheses to test whether there is a non-zero linear correlation in the population.
Use an appropriate test to find the -value.
State the conclusion at the significance level.
Explain why this conclusion does not establish that greater expenditure causes a higher percentage of public transport journeys.
0
A company tests
using a random sample of independent items. Let be the number of defective items. The significance level is .
Determine the critical region, choosing the region that maximizes the probability of a Type I error while keeping it below .
Write down the actual probability of a Type I error.
Given that the true proportion of defective items is , calculate the probability of a Type II error.
Find the power of the test when the true proportion is .
0
The lifetime, in hours, of a component is normally distributed with known population standard deviation hours. A test of
uses the mean lifetime of a random sample of components. The significance level is .
Calculate the critical value of and state the critical region.
When the true population mean is hours, calculate the probability of a Type II error.
Calculate the power of the test when the true population mean is hours.
State the effect on the probability of a Type II error if the significance level is reduced while the sample size remains unchanged.
0
A museum records the payment method and age group of a random sample of 200 visitors. The observed frequencies are shown in the table. A test for independence is performed at the significance level.
Payment method | Under 25 | 25–49 | 50 or over | Row total |
|---|---|---|---|---|
Cash | 18 | 24 | 30 | 72 |
Card | 32 | 28 | 20 | 80 |
App | 25 | 15 | 8 | 48 |
Column total | 75 | 67 | 58 | 200 |
State the null and alternative hypotheses.
Calculate the expected frequency of visitors aged under 25 who paid with cash.
The observed rows are , and . Carry out the test and state the conclusion in context.
0
A seed supplier claims that packets contain seeds from five varieties in the proportions , , , and . A random sample of 200 seeds gives observed frequencies , , , and .
Seed variety | Claimed proportion | Observed frequency |
|---|---|---|
Variety 1 | 0.10 | 14 |
Variety 2 | 0.20 | 48 |
Variety 3 | 0.40 | 72 |
Variety 4 | 0.20 | 43 |
Variety 5 | 0.10 | 23 |
State suitable hypotheses for a goodness-of-fit test.
Find the expected frequencies.
Carry out the test at the significance level and interpret the result.
0
A delivery company claims that unsuccessful deliveries occur equally often on each of the five working days. During 200 unsuccessful deliveries, the frequencies from Monday to Friday are , , , and .
Weekday | Unsuccessful deliveries |
|---|---|
Monday | 32 |
Tuesday | 41 |
Wednesday | 29 |
Thursday | 50 |
Friday | 48 |
Find the expected frequency for each day under the company's claim.
Calculate the statistic and its degrees of freedom.
Find the -value and explain why conclusions at the and significance levels are different.
0
A conservationist compares the mean time taken to identify a bird species using two field guides. Independent random samples give the following summaries.
Guide A: , seconds, seconds.
Guide B: , seconds, seconds.
Times in both populations are assumed normal with equal unknown variances.
State hypotheses to test whether the population mean time using guide A is greater than that using guide B.
Explain why a pooled two-sample -test is appropriate.
Carry out the test at the significance level and interpret the result.
0
Two independent laboratories measure the concentration of a mineral in samples from two water sources. The measurements are assumed to come from normal populations with equal unknown variances.
Source P: , , .
Source Q: , , .
State hypotheses to test whether the two population means are different.
Find the test statistic and the -value.
State and compare the conclusions at the and significance levels.
0
The mass , in grams, of a product is claimed to follow . A sample of 300 products is grouped into the intervals , , and . The observed frequencies are , , and .
Mass interval [g] | Observed frequency |
|---|---|
M < 42 | 37 |
42 ≤ M < 50 | 112 |
50 ≤ M < 58 | 111 |
M ≥ 58 | 40 |
Find and .
Hence find the four expected frequencies, correct to one decimal place.
State hypotheses for a goodness-of-fit test.
Carry out the test at the significance level and interpret the result. If you did not obtain the expected frequencies in part (a), use .
0
A school compares the mean waiting times at two canteens. The observed difference in sample means is minutes. Under the null hypothesis of equal population means, 5000 random reallocations of the recorded waiting times are performed. In 118 reallocations, the difference is at least minutes in the specified direction.
Simulation outcome | Count (out of 5000) |
|---|---|
Difference < 2.8 min | 4882 |
Difference ≥ 2.8 min | 118 |
Total | 5000 |
Estimate the one-tailed -value.
State the decisions at the and significance levels.
Explain what the simulation represents and why the estimated -value is not the probability that the null hypothesis is true.
0
An online course provider records the device used and whether each of 190 randomly selected learners completed a course. The observed frequencies are shown in the table.
Device | Completed | Did not complete | Total |
|---|---|---|---|
Laptop | 50 | 20 | 70 |
Tablet | 35 | 25 | 60 |
Phone | 25 | 35 | 60 |
Total | 110 | 80 | 190 |
Given that the laptop row is , calculate the expected frequencies for this row.
Write down the degrees of freedom and state the alternative hypothesis.
The tablet row is and the phone row is . Carry out a test at the significance level.
0
The breaking force of a type of cable is assumed to be normally distributed. The population standard deviation is unknown. A sample of 16 cables has mean breaking force and sample standard deviation . A manufacturer claims that the population mean is .
State hypotheses to test the manufacturer's claim against a two-sided alternative.
Explain why a one-sample -test is appropriate.
Carry out the test at the significance level and interpret the result.
0
For a random sample of 12 farms, a researcher records annual rainfall and crop yield. The sample product moment correlation coefficient is . The researcher claims that greater rainfall is associated with greater crop yield.

State hypotheses for an appropriate test of the researcher's claim.
State the assumption about the joint population distribution that is required, and define the parameter being tested.
Carry out the test at the significance level and explain one limitation of the conclusion.
0
A bottling company fills containers with juice. Fill volume is normally distributed with population mean and known population standard deviation . A random sample of containers is used to test
The significance level is .

State what a Type I error would mean in this context.
Given that the sample mean is , calculate the test statistic and state the conclusion.
Determine the critical value of and state the critical region.
When the true population mean is , calculate the probability of a Type II error and hence the power of the test.
0
Ten musicians test a new breathing exercise. For each musician, the increase in the time for which a note can be sustained is calculated as
The sample of differences has mean and sample standard deviation . The differences are assumed to come from a normal population.
Musician | Before / s | After / s | / s |
|---|---|---|---|
1 | 18.0 | 21.5 | 3.5 |
2 | 19.4 | 22.7 | 3.3 |
3 | 20.1 | 23.2 | 3.1 |
4 | 17.8 | 20.6 | 2.8 |
5 | 21.0 | 23.8 | 2.8 |
6 | 24.5 | 24.6 | 0.1 |
7 | 23.1 | 23.4 | 0.3 |
8 | 22.2 | 22.8 | 0.6 |
9 | 18.7 | 19.4 | 0.7 |
10 | 26.0 | 26.8 | 0.8 |
Explain why a one-sample -test of the differences is appropriate.
Calculate the test statistic for testing whether the exercise increases the population mean sustaining time.
The one-tailed -value is . State the conclusion at the significance level.
The conductor considers an increase practically useful only if the population mean increase exceeds . Test this claim at the significance level and comment on the difference between statistical and practical significance.
0
Two independent cooling systems are tested using separate electronic devices. Cooling times are assumed to be normally distributed with equal but unknown population variances. Summary statistics are shown for the two systems.
System | Sample size n | Mean cooling time [min] | Sample standard deviation [min] |
|---|---|---|---|
A | 12 | 52.4 | 4.1 |
B | 10 | 48.6 | 3.7 |
State hypotheses to test whether system A has a greater population mean cooling time than system B.
Calculate the pooled estimate of the common population standard deviation.
Carry out the test at the significance level.
difference is considered operationally important only if minutes. Test this claim and interpret the result.
0
A marine scientist investigates the relationship between water temperature and the growth rate of an algae species at locations. The paired variables are assumed to follow a bivariate normal distribution. For all locations, the sample product moment correlation coefficient is .

State hypotheses for testing whether there is a positive linear correlation in the population.
Calculate the test statistic.
The one-tailed -value is . State the conclusion at the significance level.
After the potentially influential location is removed, the remaining data give and a two-tailed -value of . Discuss what this suggests about the original claim.
0
A national park classifies volunteers by age group and preferred conservation activity. The observed frequencies are shown in the table. A chi-squared test is used to investigate whether age group and preferred activity are independent.
Age group | Trail maintenance | Habitat restoration | Wildlife monitoring | Total |
|---|---|---|---|---|
Junior | 24 | 10 | 26 | 60 |
Senior | 19 | 25 | 26 | 70 |
Adult | 27 | 35 | 8 | 70 |
Total | 70 | 70 | 60 | 200 |
State the null and alternative hypotheses.
Calculate the expected frequency for junior volunteers who prefer trail maintenance.
GDC gives . Find the degrees of freedom and state the conclusion at the significance level.
The largest contribution to is , from adult volunteers choosing wildlife monitoring. Interpret this contribution and calculate Cramér's measure .
0
A renewable-energy company predicts that maintenance requests occur in five categories with probabilities , , , and . A random sample of requests gives observed frequencies , , , and , respectively.
Category | Predicted probability | Observed frequency |
|---|---|---|
1 | 0.10 | 12 |
2 | 0.20 | 48 |
3 | 0.30 | 71 |
4 | 0.25 | 43 |
5 | 0.15 | 26 |
Calculate the expected frequencies under the company's model.
State the number of degrees of freedom.
Calculate the chi-squared statistic and its -value.
Compare the conclusions at the and significance levels and explain why the significance level must be selected before examining the data.
0
A language-learning application claims that a learner answers each vocabulary question correctly with probability . A researcher tests whether an adaptive feature increases this probability. Each learner answers independent questions. One learner answers correctly. A simulation under the null hypothesis produces samples, of which contain at least correct answers.

Estimate the -value using the simulation.
Calculate the exact binomial -value.
Determine the critical region for a significance level and state the conclusion for the learner.
When the true success probability is , calculate the probability of a Type II error and hence calculate the power of the test, and explain what the simulation represents.
0
A digital randomizer is intended to produce each of six symbols with equal probability. In trials, the observed frequencies are , , , , and .
Symbol | Observed frequency |
|---|---|
Symbol 1 | 11 |
Symbol 2 | 17 |
Symbol 3 | 22 |
Symbol 4 | 19 |
Symbol 5 | 21 |
Symbol 6 | 30 |
Total | 120 |
State suitable hypotheses for a chi-squared goodness-of-fit test.
Calculate the expected frequency for each symbol and state the number of degrees of freedom.
Carry out the goodness-of-fit test at the significance level.
After inspecting the data, an analyst proposes a separate upper-tailed binomial test only for the sixth symbol. Explain why using this post hoc test without adjustment may be misleading.
0
A call centre models the number of complaints received in a -minute interval by a Poisson distribution with mean . During intervals, the observed numbers of complaints are grouped as , , , , and at least , with observed frequencies , , , and .
Complaints in a 30-minute interval | Observed frequency |
|---|---|
0 | 10 |
1 | 25 |
2 | 31 |
3 | 17 |
At least 4 | 17 |
Total | 100 |
Calculate the expected frequency for intervals with exactly two complaints.
Calculate the expected frequency for intervals with at least four complaints.
The complete expected frequencies are , , , and . Carry out a chi-squared goodness-of-fit test at the significance level.
Explain why the conclusion does not prove that complaints follow a Poisson distribution, and state one contextual assumption needed for the model.
0
The thickness of a glass panel is normally distributed with known population standard deviation . A sample of 25 panels is used to test
at the significance level.

Find the critical value of and state the critical region.
State the probability of a Type I error.
When the true population mean is , calculate the probability of a Type II error and the power of the test. If you did not obtain the critical value, use . Use the unrounded critical value in subsequent calculations; answers using are also accepted.
0
A germination test uses 30 independent seeds. A supplier tests
at the significance level. Let be the number of seeds that germinate.
State the distribution of under .
Determine the critical region that maximizes the probability of a Type I error while keeping it below ; state the actual probability.
When the true germination probability is , calculate the probability of a Type II error and the power of the test.
0
The number of emergency calls received each day is modelled by a Poisson distribution. The usual population mean is 4 calls per day. Assume that the numbers of calls on different days are independent. Calls over five days are used to test
at the significance level. Let be the total number of calls over five days.
State the distribution of under .
Determine the critical region and the actual probability of a Type I error.
Suppose the true population mean is 3 calls per day. Calculate the probability of a Type II error and the power of the test.
0
Ten patients have their systolic blood pressure measured before and after a relaxation programme. Define
The differences, in , are . The population of differences is assumed normal.
State hypotheses to test whether the programme reduces the population mean blood pressure.
Explain why the data should be analysed using a one-sample test rather than an independent two-sample test.
Carry out the test at the significance level and interpret the result.
0
A wildlife charity claims that more than of released turtles reach a protected feeding area. Of 40 independently tracked turtles, 30 reach the area. The probability of success is assumed constant.
State the hypotheses and the distribution used under the null hypothesis.
Explain why the test is upper-tailed.
Calculate the exact -value and state the conclusion at the significance level. Describe a Type I error in context.
0
A conservation programme normally finds that of tagged turtle nests survive a severe storm. After a new protective barrier is installed, a researcher tests
using independently selected nests. Let be the number of nests that survive. The significance level is .

State the distribution of under .
Determine the critical region.
Write down the actual probability of a Type I error and explain why it is not exactly .
When the true survival probability is , calculate the probability of a Type II error and hence the power of the test.
0
The number of emergency calls received by a rural clinic each night is modelled by a Poisson distribution with mean calls per night. A new routing system is intended to reduce the mean number of calls received by the clinic. Data are collected over eight nights. Let be the total number of calls received.
X | P(X≤x) |
|---|---|
5 | 0.0014 |
6 | 0.0040 |
7 | 0.0100 |
8 | 0.0220 |
9 | 0.0433 |
10 | 0.0774 |
11 | 0.1270 |
12 | 0.1931 |
State suitable hypotheses and the distribution of under the null hypothesis.
Determine the critical region for a significance level.
During the eight nights, calls are received. State the decision and interpret it in context.
When the true mean is calls per night, calculate the probability of a Type II error and hence the power of the test.
0
A laboratory claims that the population mean acidity index of a solution is . A random sample of batches has mean index and sample standard deviation . Acidity indices are assumed to be normally distributed.

Explain why a -test is appropriate for testing whether the population mean is less than .
Calculate the test statistic.
The one-tailed -value is . State the conclusion at the significance level.
For planning a future study, assume the population standard deviation is known to be . Determine the minimum sample size needed for a lower-tailed test to have at least power when the true mean is .
0
The breaking force of a safety clip is normally distributed with known population standard deviation . A manufacturer tests
using the mean of a random sample of clips. The significance level is .

Explain why the significance level is divided equally between two tails.
Determine the two critical values of .
State the critical region and the probability of a Type I error.
Calculate the probability of a Type II error and the power when the true population mean is .
0
A medical screening device normally gives a false alarm on of harmless samples. After recalibration, independent harmless samples are tested. Let be the number producing an alarm. The manufacturer tests
at the significance level.
Expression | Probability |
|---|---|
0.0867 | |
0.0321 | |
0.4159 |
Determine the critical region.
State the actual probability of a Type I error.
When the true alarm probability is , calculate the probability of a Type II error and the power.
The manufacturer states that a Type I error is more serious than a Type II error. Explain the practical consequence of each error in this context and suggest one change that gives greater protection against a Type I error.
0
A telescope normally records an average of transient signals per hour. After a detector upgrade, signals are counted over five hours. Let be the total number recorded. A one-tailed test is used to determine whether the upgrade has increased the mean signal rate.

State the hypotheses and the distribution of under .
Determine the critical region at the significance level.
total of signals is recorded. Calculate the -value and state the conclusion.
When the true mean rate is signals per hour, calculate the probability of a Type II error and hence the power.
0
The mass of medicine in a capsule is normally distributed with known population standard deviation . A random sample of 36 capsules is used to test
at the significance level.

Find the two critical values of and state the critical region.
Explain why the probability of a Type I error is even though the critical region has two parts.
Suppose the true population mean is . Calculate the probability of a Type II error and the power of the test. If you did not obtain the critical boundaries, use and .
0
A health researcher investigates three independent outcome variables after introducing a workplace programme. Under their respective null hypotheses, the tests produce -values , and . The researcher initially tests each outcome at the significance level.

State which outcomes are statistically significant at the level.
Assuming the three tests are independent and all three null hypotheses are true, calculate the probability of at least one Type I error.
Find the common significance level for each test such that the probability of at least one Type I error is , assuming independence.
Using the adjusted level from part (b), evaluate the researcher's evidence.
0