P-value interpretation worksheet with answer key

8 problems on p-value interpretation, ready to print for class or homework. The answer key prints on its own page, so hand out everything before it.

Problem 1

A community solar program states that 45% of its subscribers check the monthly savings dashboard. A staff member takes a random sample of subscribers and runs a one-sample z-test of H0:p=0.45H_0: p = 0.45 against Ha:p>0.45H_a: p > 0.45, where pp is the true proportion of all subscribers who check the dashboard. The p-value is 0.02710.0271. Write a complete interpretation of this p-value in context.

Problem 2

An urban planner tests whether more than 30% of a city's commuters bike to work at least once a week. She runs a one-sample z-test of H0:p=0.30H_0: p = 0.30 against Ha:p>0.30H_a: p > 0.30 and gets a p-value of 0.0280.028. Three students describe it. (a) 'There is a 2.8% chance the null hypothesis is true.' (b) 'There is a 2.8% chance the result happened by chance.' (c) 'If we reject H0H_0, there is a 2.8% chance we are wrong.' Explain why each is wrong and give the correct statement.

Problem 3

A food co-op tests whether the proportion of members who use the bulk bins differs from 0.250.25. The test gives a p-value of 0.0430.043. (a) State the decision at α=0.05\alpha = 0.05. (b) State the decision at α=0.01\alpha = 0.01. (c) A second co-op runs its own test, sets α=0.02\alpha = 0.02 in advance, and gets a p-value of exactly 0.020.02. What is its decision? (d) Explain what changed between (a) and (b).

Problem 4

A ferry operator wants evidence that more than 12% of weekday passengers bring a bicycle aboard. From a random sample of weekday passengers, a one-sample z-test gives a test statistic of z=2.14z = 2.14. (a) Find the p-value. (b) Interpret it in context. (c) What would the p-value be if the alternative had been two-sided?

Problem 5

(a) A two-sided test of H0:p=0.60H_0: p = 0.60 gives a p-value of 0.0860.086, and the sample proportion came out above 0.600.60. What would the p-value have been for the one-sided alternative Ha:p>0.60H_a: p > 0.60? (b) A right-tailed test gives a p-value of 0.0170.017. What is the two-sided p-value for the same data? (c) A researcher tests Ha:p>0.40H_a: p > 0.40 and gets z=−0.90z = -0.90. Find the p-value and explain what it says.

Problem 6

A museum director claims that 35% of visitors come from outside the county. A random sample of 400 visitors contains 132 from outside the county. Test H0:p=0.35H_0: p = 0.35 against Ha:p≠0.35H_a: p \ne 0.35. (a) Find the test statistic and p-value. (b) State the decision at α=0.05\alpha = 0.05. (c) A trustee writes, 'the test proves that exactly 35% of visitors come from outside the county.' Correct that sentence.

Problem 7

A podcast platform moves the credits to the end of each episode, then checks whether the proportion of plays that reach the end differs from 0.5000.500. In a random sample of 40,000 plays, 20,280 reach the end. (a) Find the test statistic and the p-value. (b) State the decision at α=0.05\alpha = 0.05. (c) A 95% confidence interval for the true completion rate is (0.5021,0.5119)(0.5021, 0.5119). Use it to explain why a very small p-value does not mean the change matters.

Problem 8

A regional park installs new trailhead signs asking hikers to pack out their trash. Rangers then take a random sample of 250 trail users and find 212 who carry out all of their trash. Before the signs, the park's rate was 0.800.80. (a) State the hypotheses and check the conditions. (b) Find the test statistic and the p-value. (c) Interpret the p-value in context. (d) State a decision at α=0.05\alpha = 0.05, then correct this student sentence: 'The p-value of 0.02870.0287 is the probability that the signs had no effect.' (e) Give the decision at α=0.01\alpha = 0.01 and say what that does and does not change.

P-value interpretation: answer key

  1. Assuming the true proportion of subscribers who check the dashboard is 0.450.45, there is a 0.02710.0271 probability that random sampling would produce a sample proportion at least as far above 0.450.45 as the one observed.

  2. All three are wrong: (a) and (b) either assign probability to a hypothesis or drop the null assumption, and (c) confuses the p-value with α\alpha. Correct: assuming p=0.30p = 0.30, about 2.8% of random samples this size would give a sample proportion at least this far above 0.300.30.

  3. (a) Reject H0H_0, since 0.043≤0.050.043 \le 0.05. (b) Fail to reject, since 0.043>0.010.043 > 0.01. (c) Reject, because the rule is to reject when the p-value is less than or equal to α\alpha. (d) The evidence is unchanged; only the cutoff changed.

  4. (a) p-value =1−0.9838=0.0162= 1 - 0.9838 = 0.0162. (b) If the true bicycle rate were 0.120.12, about 1.62% of samples this size would give z≥2.14z \ge 2.14. (c) Two-sided: 2(0.0162)=0.03242(0.0162) = 0.0324.

  5. (a) 0.086÷2=0.0430.086 \div 2 = 0.043. (b) 2(0.017)=0.0342(0.017) = 0.034. (c) P(Z>−0.90)=1−0.1841=0.8159P(Z > -0.90) = 1 - 0.1841 = 0.8159; the sample fell on the wrong side of 0.400.40 for Ha:p>0.40H_a: p > 0.40, so there is no evidence for the alternative.

  6. (a) p^=0.33\hat{p} = 0.33, z≈−0.84z \approx -0.84, two-sided p-value ≈0.401\approx 0.401. (b) Fail to reject H0H_0 at α=0.05\alpha = 0.05. (c) A large p-value never proves the null; it says only that 0.330.33 would not be surprising if the true proportion were 0.350.35.

  7. (a) p^=0.507\hat{p} = 0.507, z=2.80z = 2.80, two-sided p-value ≈0.005\approx 0.005. (b) Reject H0H_0 at α=0.05\alpha = 0.05. (c) The interval (0.5021,0.5119)(0.5021, 0.5119) puts the true rate at most about 1.2 percentage points above 0.5000.500, so the result is statistically significant but practically trivial.

  8. (a) H0:p=0.80H_0: p = 0.80, Ha:p>0.80H_a: p > 0.80; conditions met with np0=200np_0 = 200 and n(1−p0)=50n(1 - p_0) = 50. (b) p^=0.848\hat{p} = 0.848, z≈1.90z \approx 1.90, p-value ≈0.0287\approx 0.0287. (c) If the true carry-out rate were still 0.80, about 2.87% of random samples of 250 trail users would give a sample proportion at least as far above 0.80 as 0.848 is. (d) Reject at α=0.05\alpha = 0.05; the p-value is the probability of a result this extreme assuming the rate is 0.800.80, not the probability the signs did nothing. (e) At α=0.01\alpha = 0.01 you fail to reject, and only the cutoff changed.

The key gives final answers only. The full worked solution for each problem is on the matching practice set, and every other sheet is on the worksheets page.

More worksheets from this unit