Edexcel A-Level Maths
Population, sample and sampling techniques
Total 27 marks
Name
Class
Date
- 1A college has 900 students: 540 in Year 12 and 360 in Year 13. The principal wants to find out opinion on a new timetable by asking a sample of 60 students.(a)Which of the following is the population in this investigation?[1 mark]
- AThe 60 students asked
- BAll 900 students at the college
- CThe 540 Year 12 students
- DAll students in the town
Show the mark scheme
Mark scheme
Answer: B(b)The principal takes a stratified sample of 60 students using year group as the strata. How many Year 13 students are in the sample?[1 mark]- A20
- B30
- C24
- D36
Show the mark scheme
Mark scheme
Answer: C(c)State one advantage and one disadvantage of using a sample rather than a census in this investigation.[2 marks]Show the mark scheme
Mark scheme
B1: advantage, e.g. cheaper, quicker, or less data to process than asking all 900 students
B1: disadvantage, e.g. may not be representative of the whole population, or the results are subject to sampling variability so are less accurate than a census
Accept: a census would be time-consuming; a sample gives only an estimate
Do not accept: a sample is easier, with no reasonTotal for question 1: 4 marks
- 2A factory produces 2000 bottles each day, numbered 1 to 2000 in order of production. A quality controller takes a systematic sample of 80 bottles, choosing the starting bottle at random from the first interval.(a)What is the sampling interval?[1 mark]
- A25
- B80
- C2000
- D40
Show the mark scheme
Mark scheme
Answer: A(b)The first bottle selected is number 12. Which of these bottles is also in the sample?[1 mark]- A1000
- B1250
- C1500
- D987
Show the mark scheme
Mark scheme
Answer: D(c)The bottling machine has 25 filling nozzles used in turn, and the 7th nozzle is faulty. Explain why the controller's sample may give a misleading picture of the proportion of faulty bottles.[2 marks]Show the mark scheme
Mark scheme
B1: the sampling interval (25) matches the pattern of the nozzles
B1: so the sample contains either all faulty bottles or none (every selected bottle uses the same nozzle), so it is biased
Accept: the sample is not random, so it can be periodic with the fault
Do not accept: the sample is too smallTotal for question 2: 4 marks
- 3A market researcher wants to find out how adults in a town feel about a new bus service. She stands outside the railway station on a Monday morning and interviews the first 20 men and the first 20 women who pass and agree to answer.(a)Name the sampling method used and give two reasons why the sample may not be representative of adults in the town.[3 marks]
Show the mark scheme
Mark scheme
B1: quota sampling (with opportunity sampling)
B1: only people at the station on a Monday morning are included, so some groups are missed (e.g. people who do not use trains or who are at work)
B1: the researcher chooses who to ask, which is non-random and may introduce bias
Accept: non-response, as only those who agree are interviewed
Do not accept: the sample is too small, with no reason(b)The researcher has access to the town's electoral register of 12 000 names. Describe how she could select a simple random sample of 50 adults, and state one advantage of this over her method.[4 marks]Show the mark scheme
Mark scheme
B1: number (label) the names 1 to 12 000
B1: generate random numbers between 1 and 12 000 (calculator or random number generator)
B1: ignore repeats and continue until 50 different numbers, then select those people
B1: advantage, e.g. every adult on the register has an equal chance of selection, so there is no researcher bias
Accept: draw numbers from a hat
Do not accept: choose 50 names without numbering themTotal for question 3: 7 marks
- 4A school has 900 students: 450 in Years 7 to 9, 300 in Years 10 and 11, and 150 in the sixth form. The head teacher wants to estimate the mean number of hours of sleep students get on a school night, using a sample of 60 students.(a)(i) Calculate how many students from each group should be in a stratified sample of 60.[6 marks]
(ii) Explain why a stratified sample is more appropriate than asking the 30 students in her own tutor group.
(iii) A second teacher takes another stratified sample of 60 and finds a different mean. Explain why this can happen.Show the mark scheme
Mark scheme
M1: (i) or equivalent proportion
A1: (i) 30, 20 and 10 students
B1: (ii) a stratified sample reflects the proportions of the age groups, so is representative of all years
B1: (ii) a tutor group is from one year group (and chosen for convenience), so is unlikely to represent the whole school
B1: (iii) different samples contain different students, so sample means vary
R1: (iii) one sample gives only an estimate of the population mean, so conclusions can differ
Accept: (iii) sampling variability(b)Evaluate the use of a census, instead of a stratified sample, for this investigation, and state which you would recommend.[6 marks]Show the mark scheme
Mark scheme
B1: a census gives the true population mean, with no sampling error
B1: every student is included, so no sample bias from the choice of sample
B1: but it is time-consuming and costly to collect and process data from 900 students
B1: students who do not respond could still bias the results
B1: a well-chosen stratified sample is quicker and gives a good estimate
R1: a justified recommendation, e.g. stratified sample, as the estimate is reliable enough and the cost of a census is not justified
Accept: a census, if justified by the need for accuracyTotal for question 4: 12 marks
End of questions