Interpreting confidence intervals: worksheet with answer key

8 problems on interpreting confidence intervals:, ready to print for class or homework. The answer key prints on its own page, so hand out everything before it.

Problem 1

A county fair surveyed a random sample of 500 attendees and found that 100 arrived by public transit. The resulting 95% confidence interval for the proportion of all attendees who arrived by public transit is (0.165, 0.235)(0.165,\ 0.235). Write a full-credit interpretation of this interval.

Problem 2

A field ecologist tagged monarch butterflies and later checked a random sample of 200 of them, finding that 94 had survived the migration. Her 90% confidence interval for the survival proportion is (0.412, 0.528)(0.412,\ 0.528).

(a) Interpret the 90% confidence level, not the interval. (b) A colleague says the 90% means there is a 90% chance the survival rate is between 0.412 and 0.528. Explain why that is not what the confidence level says.

Problem 3

In a random sample of 900 seniors in a large district, 270 have a part-time job, which gives the 95% confidence interval (0.270, 0.330)(0.270,\ 0.330). Five students offer interpretations. Say which are correct, and for each wrong one name the specific error.

  1. About 95% of the sampled seniors gave answers between 0.270 and 0.330.
  2. There is a 95% probability that the true proportion of seniors with a part-time job is between 0.270 and 0.330.
  3. We are 95% confident that the interval from 0.270 to 0.330 contains the true proportion of all seniors in this district who have a part-time job.
  4. If we took many random samples of 900 seniors from this district and built a 95% interval from each, about 95% of those intervals would contain the true proportion.
  5. About 95% of all sample proportions from samples of 900 seniors fall between 0.270 and 0.330.

Problem 4

A snack company samples 600 random buyers and finds 210 are repeat customers, giving a 95% confidence interval for the proportion of all buyers who are repeat customers of (0.312, 0.388)(0.312,\ 0.388).

(a) The company's marketing page claims 35% of buyers are repeat customers. Does the interval give convincing evidence against that claim? (b) A competitor's analyst claims the real figure is only 25%. Does the interval give convincing evidence against that claim? (c) A colleague concludes from part (a) that the interval proves the true proportion equals 0.35. Correct him.

Problem 5

Three delivery services are studied with separate random samples. A 95% confidence interval for the proportion of Brand A packages arriving on time is (0.81, 0.89)(0.81,\ 0.89). For Brand B it is (0.86, 0.94)(0.86,\ 0.94). A third company, Brand C, gives (0.95, 0.99)(0.95,\ 0.99).

(a) A manager says Brand B is better than Brand A because its interval sits higher. Evaluate that reasoning. (b) What do the Brand A and Brand C intervals let you say? (c) What procedure actually answers the question 'is there a difference?'

Problem 6

A pharmacy tracks a random sample of 1600 prescriptions and finds that 320 were picked up late. Treat the sample as under 10% of all prescriptions the pharmacy fills.

(a) Build the 90% confidence interval for the proportion of all prescriptions picked up late. (b) Build the 99% interval from the same sample. (c) Explain in context what improved and what got worse when the confidence level rose.

Problem 7

A city bike-share program draws a random sample of 250 station-mornings from its season log and finds 60 on which the station was completely empty at 8 a.m. Treat the sample as under 10% of all station-mornings in the season.

(a) Construct a 95% confidence interval for the proportion of all station-mornings with an empty station. (b) Interpret the interval in context. (c) Interpret the confidence level in context.

Problem 8

A repair shop times a random sample of 16 tire rotations. The sample mean is 42 minutes with a sample standard deviation of 6 minutes, and a dotplot of the times shows no strong skew and no outliers, so a tt-interval is reasonable. Use t∗=2.131t^* = 2.131.

(a) Construct a 95% confidence interval for the true mean rotation time. (b) Interpret the interval in context. (c) The shop's website promises an average of 40 minutes. Does the interval give convincing evidence against that promise? (d) A trainee says the interval means 95% of tire rotations take between the two endpoints. Explain the error.

Interpreting confidence intervals:: answer key

  1. We are 95% confident that the interval from 0.165 to 0.235 contains the true proportion of all attendees at this county fair who arrived by public transit.

  2. (a) If she repeated this sampling many times with samples of 200 tagged monarchs and built a 90% interval from each, about 90% of those intervals would contain the true proportion of tagged monarchs that survived the migration. (b) The interval is already computed and the true proportion is a fixed number, so this interval either contains it or it does not; the 90% describes how often the method succeeds in the long run, not the chance for one finished interval.

  3. Statements 3 and 4 are correct. Statement 1 confuses the interval with the spread of individual responses, statement 2 attaches a probability to a finished interval whose endpoints are already fixed, and statement 5 centers the sampling distribution on p^\hat{p} instead of on pp.

  4. (a) No. 0.35 lies inside the interval, so it stays plausible. (b) Yes. 0.25 lies below the entire interval, so the data give convincing evidence the true proportion is higher than 25%. (c) An interval never proves one value; it lists every value the data leave plausible, and 0.35 is only one of them.

  5. (a) Not valid. The intervals overlap between 0.86 and 0.89, so a common true proportion is still plausible. (b) The Brand A and Brand C intervals do not overlap, which is strong evidence their true on-time proportions really differ. (c) Build an interval for p1−p2p_1 - p_2 or run a two-proportion zz-test; comparing two one-sample intervals settles nothing by itself.

  6. (a) About (0.184, 0.216)(0.184,\ 0.216). (b) About (0.174, 0.226)(0.174,\ 0.226). (c) Higher confidence buys a higher long-run capture rate at the cost of precision: the interval is about 1.57 times wider and says less about where the true proportion sits.

  7. (a) About (0.187, 0.293)(0.187,\ 0.293). (b) We are 95% confident that the interval from 0.187 to 0.293 contains the true proportion of all station-mornings this season on which a station was completely empty at 8 a.m. (c) If the program repeated this sampling many times with 250 station-mornings each and built a 95% interval from each, about 95% of those intervals would contain that true proportion.

  8. (a) About (38.80, 45.20)(38.80,\ 45.20) minutes. (b) We are 95% confident that the interval from 38.80 to 45.20 minutes contains the true mean tire rotation time at this shop. (c) No. 40 minutes lies inside the interval, so the promise stays plausible. (d) The interval estimates the mean of all rotation times, not the times of individual rotations, so it says nothing about where 95% of individual jobs fall.

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