Revision › Sampling › Worksheet

Sampling Hard

GCSE Maths · Statistics · Name ____________________ Date ________

Builds on: Charts and tables, Averages and spread

Do now

Quick questions on earlier topics - to get your brain going.

  1. A frequency table lists the number of tickets sold in three sections of a theatre: Section A 30, Section B 45, Section C ?. The total tickets sold were 110. How many tickets were sold in Section C? [1]

  2. A café sold 60 coffees, 40 teas and 20 scones on a Sunday. How many customers bought tea? [1]

  3. A student incorrectly calculated the median as 4 for the data set {2, 3, 3, 5, 7}. What is the correct median? [1]

  4. What is the mean of the numbers 4, 8 and 12? [1]

Which is right?

Choose the right answer - then say what mistake each wrong one makes.

  1. A university with **1200** students wants to estimate how many have travelled abroad last year. Two independent simple‑random samples are taken: - Sample 1: **200** students, **68 %** say they have travelled. - Sample 2: **300** students, **65 %** say they have travelled. The researcher decides to use a weighted average of the two sample proportions, weighting each by its sample size. a) What is this combined estimate? b) Which statement about that estimate is correct (assume large‑population normal approximation)?

    1. 66.7 % and it should not exceed either of the individual sample proportions
    2. 67 % – we need to check for bias because the samples were taken at different times
    3. 66 % – we can combine the samples by a simple (unweighted) average
    4. 66.2 % and its variance can be approximated as p(1‑p)/n where n is the total sample size 500
  2. A research team wants an estimate of how many students in a school with **400** pupils like chocolate ice‑cream. They randomly select **50** pupils and find that **30** of them say they do. Which of the following gives the best estimate for the whole school's percentage?

    1. 30 %
    2. 60 %
    3. 70 %
    4. 8 %

Practice

  1. In a simple random sample of 25 apples, 10 are bruised. Estimate the total number of bruised apples in a batch of 200 apples. [2]

  2. From a simple random sample of 28 people, 10 own a pet. Estimate the population proportion to three decimal places. [1]

  3. In a random sample of 80 students, 24 answered "yes" to liking football. The estimated proportion in the whole student body is therefore 0.30. If you were to take a new simple random sample of 100 students with the same true proportion, how many would be expected to answer "yes"? [1]

  4. What is the minimum sample size required to estimate a proportion within ±5 % at 95 % confidence, assuming the worst‑case proportion of 0.5? (Use Z = 1.96) [2]

  5. A population of 250 students has an unknown number who own pets. You take a simple random sample of 30 students and find that 12 own a pet. Estimate the total number of students in the whole school who own a pet. [2]

  6. A coffee shop records the waiting times (in minutes) for 20 randomly selected customers; the mean waiting time is 5 min 30 s. Estimate the total waiting time, in minutes, for all 200 customers who visited on that day. [2]

  7. A random sample of 16 people from a community found that 5 had previously travelled abroad. What is the estimated number of people in a population of 88 who have travelled abroad? Round your answer to the nearest whole number. [1]

  8. A random sample of 40 students gave an estimated proportion of 0.25 for those who have pets. How many students with pets were there in that sample? [1]

  9. A quality inspector at a factory randomly picks 8 battery packs from a shipment of 80 and finds that 5 are defective. What is the best estimate of how many defective packs are in the entire shipment? Give your answer as a whole number. [1]

  10. From a random sample of 25 pupils, the average height was \(160\text{ cm}\) and \(6\) of them were taller than \(170\text{ cm}\). If the whole school has 200 pupils, what is the best estimate of how many pupils are taller than 170 cm? Give your answer as a whole number. [1]

  11. A teacher randomly selects 20 pupils from a class of 60 and finds that 12 of them have passed the maths test. What is the best estimate of how many pupils in the whole class would have passed if all 60 took the test? Give your answer as a whole number. [1]

  12. A scientist samples 20 water bottles from a factory of 500 to test for contamination. The mean contamination level found was 0.15 mg/L. Assuming each bottle holds exactly 1 L, what is the estimated total amount of contaminant in all bottles? Give your answer in milligrams. [1]

Exam practice

Real past-paper questions on sampling, with their mark schemes, are in 4 Triangles - free to practise.

Mixed practice

Questions on topics this one links to - mixing them helps you remember both.

  1. The five numbers 14, 16, 18, x and 20 have a mean of 17. Find x. [2]

    (Describing a population)
  2. The following five temperatures (in °C) were recorded: –5, –2, 0, 3, 8. What is the range of these temperatures? [2]

    (Describing a population)
  3. Scores in a maths test are shown with their frequencies. Score | Frequency ---|--- 10 | 2 20 | 3 30 | 1 40 | 4 What is the mean score? Express your answer to one decimal place. [2]

    (Charts and tables)
  4. A pie chart shows that a slice is 72° for one option, and there were 25 respondents in total. Each person represents an equal angle. How many people chose that option? [1]

    (Charts and tables)
  5. In the data set –10, 0, 5, 7, 8, 9 and 30 use the IQR rule to identify outliers. What is the largest observation that is not an outlier? [3]

    (Averages and spread)
  6. Find the median of the data set 23, 18, 27, 30, 20 and 22 to 1 decimal place. [2]

    (Averages and spread)

💼 Where it leads: Stock control assistant · Sales administrator · Market researcher · Market research data analyst - more

Stuck, or finished? Scan for help on sampling - an explanation, hints and unlimited practice that marks itself.
Free at 4triangles.co.uk

Answers

  1. 35
  2. 40
  3. 3
  4. 8
  5. D 66.2 % and its variance can be approximated as p(1‑p)/n where n is the total sample size 500
    Weighted estimate: e = (p₁·n₁ + p₂·n₂) ÷ (n₁ + n₂) = (0.68×200 + 0.65×300)/500 = 331/500 = **0.662** → **66.2 %**. With the combined sample treated as a single random sample of size 500, the variance is p(1‑p)/n. All other options either ignore weighting or misinterpret bias. A: Using a simple average (unweighted) then rounding, plus confusion about bounds. …
  6. B 60 %
    The estimate comes from dividing the number who say ‘yes’ by the sample size and then multiplying by 100: e = (30/50)×100 = **60 %**. Using the whole school population or adding arbitrary adjustments gives an incorrect figure. A: Halving the proportion due to perceived bias – believing that only half of the sample actually counts. …
  7. 80
  8. 0.357
  9. 30
  10. 385
  11. 100
  12. 1100
  13. 28
  14. 10
  15. 50
  16. 48
  17. 36
  18. 75
  19. 17
  20. 13
  21. 27
  22. 5
  23. 9
    Q1=0, Q3=9, IQR=9; lower bound=-13.5, upper bound=22.5 => outlier 30; largest non‑outlier =9
  24. 22.5
    sorted list:18,20,22,23,27,30 median=(22+23)/2=22.5

Some key word definitions from Oak National Academy (Open Government Licence).

Stuck, or finished? Scan for help on sampling - an explanation, hints and unlimited practice that marks itself.
Free at 4triangles.co.uk