Sampling methods worksheet with answer key
8 problems on sampling methods, 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
Name the sampling method used in each design. (a) A principal numbers all 1,450 students in the school directory and uses a random number generator to draw 100 different numbers. (b) A line inspector draws a random number between 1 and 25, starts at that bottle, and then pulls every 25th bottle. (c) A pollster splits 8,000 registered voters into four age groups and takes an SRS of 50 voters from each group. (d) A district numbers its 60 bus routes, picks 6 routes at random, and surveys every student who rides those 6 routes. (e) A reporter stands outside one grocery store and interviews the first 30 shoppers who walk out. (f) A magazine prints a survey and asks readers who wish to reply to mail it in; 900 readers do.
Problem 2
A warehouse has 1,200 employees on a numbered payroll list, and an auditor wants a sample of 60. (a) Name the design and find the interval . (b) If the random start is employee 7, give the first three employee numbers selected and the last one. (c) What is each employee's chance of being selected? (d) Is this a simple random sample? Explain.
Problem 3
A district has 4,000 students: 1,800 in elementary schools, 1,200 in middle schools, and 1,000 in high schools. The superintendent wants a stratified random sample of 200 students, allocated in proportion to the size of each level, to estimate the mean time students spend on homework. (a) How many students come from each level? (b) Describe how to select them. (c) Give one reason to stratify by school level here rather than take an SRS of 200 from the whole district. (d) What population does the estimate describe?
Problem 4
A minor league baseball team posts a poll on its own fan page asking whether the city should fund a new stadium with a hotel tax. About 50,000 people see the post and 4,200 answer it, of whom 2,856 say yes. The team's press release states that 68% of city residents support the stadium. (a) Name the sampling method. (b) Verify the 68% and find the share of the people who saw the post who answered it. (c) Name the bias and the direction it pushes the estimate. (d) What population does the 68% actually describe? (e) Would 42,000 responses fix the problem?
Problem 5
A university houses 7,200 students on 240 dorm floors of about 30 residents each. The housing office numbers the floors 1 to 240, uses a random number generator to choose 8 floors, and surveys every resident of those floors about dining hall hours. (a) Name the method and give the approximate sample size. (b) Explain why this is not a stratified random sample. (c) The office wants to report the result for all 12,000 undergraduates at the university. Name the problem this creates and state what the sample can actually support.
Problem 6
A city wants to estimate the proportion of its adults who have a primary care doctor. Design A dials 600 randomly chosen listings from the residential landline directory and 540 people answer. Design B mails a survey to 800 addresses drawn at random from the city address list and 168 are returned. (a) Find each response rate. (b) Name the main bias in each design and give a likely direction. (c) Which design is easier to repair, and why?
Problem 7
A town takes an SRS of 600 registered voters and splits it at random into two groups of 300. Group A is asked, "Do you support a small sales tax increase so the public library can stay open seven days a week?" and 219 say yes. Group B is asked, "Do you support raising taxes to pay for longer library hours?" and 129 say yes. (a) Find both percentages and the gap between them. (b) The selection was random, so name what went wrong. (c) Say which direction each wording pushes its estimate. (d) Would asking 6,000 voters close the gap? (e) What should the town report?
Problem 8
A high school with 1,600 students wants to estimate the mean number of hours its students sleep on a school night. A statistics teacher gives the survey to the 120 students in her four sections, an elective taken almost entirely by juniors and seniors, and 96 return it, with a mean of 6.4 hours. (a) Name the sampling method and give the response rate. (b) Name two ways this design biases the estimate of the school mean, each with a direction. (c) The principal instead wants a stratified random sample of 160 students allocated in proportion to grade size, where the grades hold 480, 420, 380, and 320 students. Give the number from each grade and describe the selection. (d) The new sample gives a mean of 6.9 hours. State the population this describes, and say whether the school can conclude that more sleep causes higher test scores.
From statslearn.org/worksheets/sampling-methods-worksheet. Free to print and copy for classroom use.
Sampling methods: answer key
(a) simple random sample; (b) systematic random sample; (c) stratified random sample; (d) cluster random sample; (e) convenience sample; (f) voluntary response sample.
(a) Systematic random sample, . (b) 7, 27, 47, and the last is 1187. (c) for each employee. (d) Not an SRS: only 20 samples are possible, so groups such as employees 1 through 60 cannot occur.
(a) 90 elementary, 60 middle, 50 high, from a sampling fraction of . (b) Number the students within each level and use a random number generator to draw 90, 60, and 50 distinct numbers, then combine. (c) Levels differ on homework time, so stratifying guarantees representation and lowers the variability of the estimate. (d) All 4,000 students in this district.
(a) Voluntary response sample. (b) , or 68%, and , or 8.4%, of those who saw the post answered it, a participation rate rather than a response rate since nobody was selected. (c) Voluntary response bias plus undercoverage of non-fans, both pushing the estimate of support upward. (d) Only the 4,200 who answered. (e) No, a larger self-selected sample keeps the same lean.
(a) Cluster random sample of about residents. (b) Cluster sampling takes everyone from some groups, while stratified sampling takes some people from every group; 232 floors contribute no one. (c) Reporting for all 12,000 undergraduates is undercoverage bias; the sample supports a conclusion about the 7,200 students in university housing only.
(a) Design A 90%, Design B 21%. (b) A has undercoverage bias from the landline frame, likely overstating the proportion because landline households skew older; B has nonresponse bias from the 79% who did not reply, likely overstating it because responders are more engaged with health care. (c) Design B, because its frame covers the population and nonresponse can be reduced with follow-up, while A's frame permanently excludes cell-only households.
(a) 73% and 43%, a gap of 30 percentage points. (b) Selection was random, so this is response bias from question wording, not sampling bias; because the wordings were randomly assigned, the wording is what caused the gap. (c) Version A leans high at 73%, Version B leans low at 43%. (d) No, a larger sample reduces variability but not bias. (e) Neither figure; re-ask with neutral wording.
(a) Convenience sample, response rate . (b) Undercoverage of everyone outside her junior and senior sections pushes 6.4 hours downward as an estimate of the school mean, and nonresponse from the 24 who did not return the form pushes it upward; the two leans oppose, so the net direction is unknown. (c) 48, 42, 38, and 32 students from the four grades, selected with a random number generator inside each grade and combined. (d) It describes the mean sleep of all 1,600 students at this school; no causal conclusion is possible, since random selection allows generalization but only random assignment supports cause and effect.
Every step is worked at statslearn.org/practice/sampling-methods-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.
More worksheets from this unit
- Identifying bias worksheet with answer key
- Boxplot and five-number summary worksheet with answer key
- Categorical data worksheet with answer key
- Comparing two distributions worksheet with answer key
- Describing and comparing distributions worksheet
- Experimental design worksheet with answer key
- Investigative question worksheet with answer key
- Dotplot, stemplot, and histogram worksheet with answer key
- Quartiles, IQR, and outliers worksheet with answer key
- Scope of inference worksheet with answer key
- Standard deviation and variance worksheet with answer key
- Choosing mean vs median and SD vs IQR worksheet with answers
- Classifying variables worksheet with answer key