Mixed inference worksheet with answer key
8 problems on mixed inference, ready to print for class or homework. The answer key prints on its own page, so hand out everything before it.
Name: ______________________________ Date: ______________ Period: ______
Problem 1
A campus transportation office wants to estimate the proportion of undergraduates who commute by city bus. It takes a simple random sample of 250 undergraduates from the enrollment list of 9,400 and finds that 90 of them commute by bus.
(a) Name the procedure and justify it from the data structure. (b) Check the conditions. (c) Build a 95% confidence interval and interpret it.
Problem 2
A coffee roaster advertises that its retail bags hold a mean of 340 grams. An inspector weighs a random sample of 25 bags from one week's production and gets a sample mean of 336.4 grams with a sample standard deviation of 8.0 grams. A dotplot of the 25 weights is roughly symmetric with no outliers.
(a) Name the procedure and justify it. (b) State the hypotheses and check the conditions. (c) Compute the test statistic and bound the p-value from a t table. (d) Conclude at .
Problem 3
An online retailer randomly assigns 1,000 site visitors to one of two checkout pages, 500 to a redesigned page and 500 to the current page. Of the redesigned group, 145 complete a purchase; of the current group, 115 do.
(a) Name the procedure and justify it. (b) State the hypotheses. (c) Check the conditions and compute the test statistic using the pooled proportion. (d) Conclude at and say what kind of conclusion this design supports.
Problem 4
A greenhouse randomly assigns 20 basil seedlings to two feeding schedules, 10 seedlings to each, and measures height in centimeters after six weeks. Schedule A: , , . Schedule B: , , . Neither sample shows strong skewness or outliers.
(a) Name the procedure and justify it, including why it is not paired. (b) Compute the test statistic. (c) Bound the p-value with conservative degrees of freedom, then give the technology value. (d) Conclude at .
Problem 5
A hand-therapy clinic measures grip strength in kilograms for 8 randomly selected patients before and after a four-week program.
| Patient | Before | After |
|---|---|---|
| 1 | 28 | 31 |
| 2 | 31 | 33 |
| 3 | 25 | 29 |
| 4 | 34 | 36 |
| 5 | 29 | 30 |
| 6 | 27 | 31 |
| 7 | 33 | 34 |
| 8 | 30 | 34 |
(a) Name the procedure and justify it. (b) Compute the differences with their mean and standard deviation. (c) Test : against : at , then say what the design does not let you claim.
Problem 6
A grocery chain takes one random sample of 200 shoppers at a single store and records each shopper's payment method and whether that shopper used a loyalty account.
| Payment | Loyalty | No loyalty | Total |
|---|---|---|---|
| Card | 66 | 34 | 100 |
| Cash | 18 | 22 | 40 |
| Mobile wallet | 36 | 24 | 60 |
| Total | 120 | 80 | 200 |
(a) Name the procedure and explain why a two-proportion z-test will not do. (b) Compute the statistic and the degrees of freedom. (c) Conclude at and comment on how close the call is.
Problem 7
A sensor manufacturer's spec sheet claims that no more than 8% of shipped units fail a calibration check. An independent lab pulls a random sample of 300 units from a shipment of 12,000 and finds 33 failures.
(a) Name the procedure and explain which proportion belongs in the standard error. (b) State the hypotheses and check the conditions. (c) Compute the test statistic and the p-value. (d) Conclude at .
Problem 8
A cycling club tests two tire compounds. Each of 8 riders rides the same 15 km course twice, once on each compound, with the order randomized. The times come back in minutes as two columns.
| Rider | Tire A | Tire B |
|---|---|---|
| 1 | 42.1 | 41.3 |
| 2 | 38.6 | 38.0 |
| 3 | 45.0 | 44.6 |
| 4 | 40.2 | 39.4 |
| 5 | 43.5 | 43.1 |
| 6 | 39.8 | 39.6 |
| 7 | 41.7 | 40.5 |
| 8 | 44.3 | 43.9 |
A club member sees two columns of 8 numbers, runs a two-sample t-test, and gets with a p-value near 0.62. (a) Explain why that procedure is wrong here. (b) Run the correct test at . (c) Explain why the two answers land so far apart.
From statslearn.org/worksheets/mixed-review-inference-worksheet. Free to print and copy for classroom use.
Mixed inference: answer key
One-proportion z-interval (one sample, categorical response, estimation). , , , interval .
One-sample t-test (one sample, quantitative, unknown). , , , two-sided p between 0.02 and 0.05 (about 0.034). Reject at .
Two-proportion z-test (two independent groups, categorical response). , , , two-sided p about 0.031. Reject , and random assignment supports a causal reading.
Two-sample t-test (independent groups, quantitative response). , ; conservative gives p between 0.02 and 0.05, technology gives p about 0.016. Reject .
Paired t-test (same patients measured twice). , , , , , one-sided p under 0.0005. Reject ; grip strength rose, but with no control group the program itself is not established as the cause.
Chi-square test for independence (one sample, two categorical variables, three payment categories). , ; since (p about 0.072), fail to reject .
One-proportion z-test (one sample, categorical, claimed value). Use in the standard error: . With , and the one-sided p-value is 0.0277, so reject .
Paired t-test on the 8 differences: , , , , , two-sided p between 0.001 and 0.002 (about 0.0011), so reject . The two-sample test fails because it keeps the rider-to-rider variation that pairing removes.
Every step is worked at statslearn.org/practice/mixed-review-inference-practice. Free to print and copy for classroom use.
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.