Level:
Correct: 0
Incorrect: 0
Marks: 0/0
Exam Practice
Course:
Topics:
Edexcel GCSE Statistics (Higher Paper 2) Smart practice A mixed round aimed at your weak spots
Unit 1
The statistical enquiry cycle
Statistical enquiry cycle stages 1 and 2; (a) Planning: 1a.01, 1a.02, 1a.03
Statistical enquiry cycle: planning and collecting
Statistical enquiry cycle stages 3, 4 and 5
Statistical enquiry cycle: interpreting and evaluating
Unit 2
The collection of data
(b) Types of data: 1b.01, 1b.02, 1b.03
Types of data and variables
(b) Types of data: 1b.04; (d) Collecting data: 1d.01, 1d.02
Primary and secondary data, reliability and validity
(c) Population and sampling: 1c.01, 1c.02, 1c.03
Populations, sampling frames and non-random sampling
(c) Population and sampling: 1c.04, 1c.05
Random and systematic sampling
(c) Population and sampling: 1c.06
Stratified sampling
(d) Collecting data: 1d.05, 1d.06
Problems with collected data and cleaning data
Unit 3
Processing, representing and analysing data
(a) Tabulation, diagrams and representation: 2a.01
Tabulation, tallies, pictograms and two-way tables
(a) Tabulation, diagrams and representation: 2a.01, 2a.02
Pie charts and comparative pie charts
(a) Tabulation, diagrams and representation: 2a.01
Stem and leaf diagrams
(a) Tabulation, diagrams and representation: 2a.02
Population pyramids, choropleth maps and 2D and 3D diagrams
(a) Tabulation, diagrams and representation: 2a.04
Histograms and frequency density
(a) Tabulation, diagrams and representation: 2a.05, 2a.06, 2a.07, 2a.08
Choosing representations and misleading diagrams
(a) Tabulation, diagrams and representation: 2a.09, 2a.10
Skewness
(b) Measures of central tendency: 2b.01, 2b.02, 2b.03
Mode, median and mean
(b) Measures of central tendency: 2b.01
Weighted mean and geometric mean
(c) Measures of dispersion: 2c.01, 2c.06
Standard deviation and standardised scores
(a) Tabulation, diagrams and representation: 2a.03; (c) Measures of dispersion: 2c.02, 2c.03
Box plots and outliers
(b) Measures of central tendency: 2b.03; (c) Measures of dispersion: 2c.04, 2c.05
Comparing distributions
(d) Further summary statistics: 2d.01
Index numbers
(d) Further summary statistics: 2d.02
Rates of change
(e) Scatter diagrams and correlation: 2e.01, 2e.02, 2e.03
Scatter diagrams, correlation and causation
(e) Scatter diagrams and correlation: 2e.04
Lines of best fit and regression lines
(e) Scatter diagrams and correlation: 2e.05, 2e.06, 2e.07, 2e.08
Spearman's rank and product moment correlation coefficients
(f) Time series: 2f.01, 2f.02; (b) 2b.01 mean seasonal variation
Time series, moving averages and seasonal variation
(g) Quality assurance: 2g.01, 2g.02
Quality assurance and control charts
(h) Estimation: 2h.01, 2h.02, 2h.03
Estimation, sample size and capture-recapture
Unit 4
Probability
3. Probability: 3p.01, 3p.02, 3p.03, 3p.04
Probability, expected frequency and risk
3. Probability: 3p.05, 3p.06, 3p.10
Experimental and theoretical probability and bias
3. Probability: 3p.11
Binomial distribution
3. Probability: 3p.12, 3p.13
Normal distribution
Avg
score:
–
Topics
done:
0/40

Bias, sensitive questions and the random response technique

Practise 13 exam-style questions on Bias, sensitive questions and the random response technique for Edexcel GCSE Statistics (Higher Paper 2). Write your answer as you would in the exam and get it marked in seconds against the mark scheme, point by point, with the Study Helper on hand if you get stuck.

  • 13 questions
  • 1–5 marks each
  • Instant marking
  • No sign-up needed
13 questions, marked as you go

Sample questions with model answers

Here are 4 of the questions in this topic, each with a model answer, an explanation of why it earns the marks and the mark scheme an examiner would use.

Question 1

1 mark

A cycling charity wants to estimate the proportion of cyclists in a town who have cycled at night without lights in the last month, which is against the law. It will use the random response technique. Each cyclist will be given a random outcome, then either tick Yes or answer the question honestly, depending on the outcome.

Which of these is the most suitable way for each cyclist to get the random outcome?

A   Think of a whole number from 1 to 6.
B   Roll a fair dice where nobody else can see it.
C   Roll a fair dice and show the result to the interviewer.
D   Use the last digit of their house number.

Choose your answer.

  • A
  • B
  • C
  • D

Model answer

B

A fair dice gives every outcome the same chance, and rolling it where nobody else can see it means that nobody knows whether a cyclist's Yes was forced by the dice or is a true answer.

Explanation

The random outcome must be both random and private. A number that someone thinks of is not random, because people can choose the outcome that suits them. If the interviewer sees the dice, a Yes from a cyclist who was told to answer the question is a real admission, so the protection is lost. A house number is fixed and can be known by others.

Mark scheme
  1. 1 mark

    B1 for B

    Accept the method written out in words. More than one letter given scores B0. A is not random: people choose their own number, and someone who wants to avoid the question can pick a number that tells them to tick Yes. C is not private: the interviewer knows which cyclists had to answer the question, so their Yes answers are not hidden. D is not random, and other people may know the house number.

Question 2

2 marks

Volunteers from an environmental campaign group carry out a street survey about plastic. Each volunteer wears a T-shirt with the group's logo and the slogan 'Say no to plastic'. They ask each person they stop,

'How many drinks in single-use plastic bottles did you buy last week?'

Explain why the way the survey is carried out may bias the answers to this question.
Suggest how the survey could be carried out to reduce this bias.

Model answer

People can see from the T-shirts that the volunteers are against plastic. Many people will not want to be judged, so they may say they bought fewer drinks in plastic bottles than they really did. The results are likely to be too low.

To reduce this bias, the interviewers should not show any link to the campaign group (plain clothes and a neutral introduction), or people could answer anonymously on a form that they post in a sealed box.

Explanation

Bias can come from who asks a question as well as from the question itself: people tend to give the answer they think the interviewer wants to hear. Here the bias has a direction, so say that the answers are likely to be too low. Neutral interviewers or anonymous answers remove the pressure; a bigger sample would simply repeat the same bias.

Mark scheme
  1. 1 mark

    B1 for people can see that the volunteers are against plastic, so they may say they bought fewer drinks in plastic bottles than they really did (to avoid being judged / to give the answer the volunteers want), so the answers are likely to be too low

    oe. Must link the bias to the volunteers or their T-shirts (who is asking, face to face), or say that the answers are likely to be too low with a reason. Accept 'people may be embarrassed to admit buying plastic bottles in front of the campaigners'. B0 for 'people may lie' with no reason. B0 for 'people may forget' (not caused by the way the survey is carried out). B0 for comments about which people are stopped (sampling).

  2. 1 mark

    B1 for a way to reduce the bias, e.g. the interviewers should not show any link to the campaign (plain clothes, a neutral introduction) / use an independent survey company / let people answer anonymously, e.g. on a form they post in a sealed box or online

    oe. This mark is independent of the first. B0 for 'ask more people' or 'use a random sample' (these do not remove the pressure on people to give a low answer). B0 for 'use the random response technique' (the answer is a number, not Yes or No).

Question 3

3 marks

The rangers of a national park want to estimate how many of the 3000 people who stayed at the park's campsites last summer lit a fire where fires are not allowed. They choose 200 of these campers at random. Each camper spins, in private, a fair spinner with 8 equal sections. 3 of the sections say 'Yes' and the other 5 say 'Answer'.

If the spinner lands on Yes, tick Yes.
If the spinner lands on Answer, answer this question honestly.
'Last summer, did you light a fire in the park where fires are not allowed?'

95 of the 200 campers tick Yes.

Work out an estimate for the number of the 3000 campers who lit a fire where fires are not allowed.

Model answer

Expected number of forced Yes answers: \(\dfrac{3}{8} \times 200 = 75\)
Expected number who answer the question: \(\dfrac{5}{8} \times 200 = 125\)
Yes answers to the question: \(95 - 75 = 20\)
Estimated proportion: \(\dfrac{20}{125} = 0.16\)
Estimate for all the campers: \(0.16 \times 3000 = 480\)

Explanation

Only the campers whose spinner lands on Answer, about \(\dfrac{5}{8}\) of the 200, answer the question. Remove the 75 expected forced Yes answers, then divide the remaining 20 by the 125 who answered, not by 200. The proportion from the random sample is then used to estimate the number among all 3000 campers.

Mark scheme
  1. 1 mark

    M1 for \(\dfrac{3}{8} \times 200\) (= 75), the expected number of forced Yes answers, or \(\dfrac{5}{8} \times 200\) (= 125), the expected number who answer the question, or \(\dfrac{3}{8} + \dfrac{5}{8}p = \dfrac{95}{200}\) oe

    oe, e.g. 0.375 for \(\dfrac{3}{8}\) and 0.625 for \(\dfrac{5}{8}\). Answer-only rule for this question: A correct answer with no working scores full marks.

  2. 1 mark

    M1 for \(\dfrac{95 - 75}{125}\) (= 0.16) oe, or a complete method to solve their equation for \(p\), or \(\dfrac{20}{125} \times 3000\) oe

    oe, e.g. \((0.475 - 0.375) \div 0.625\). 0.16 or 16% seen implies M1M1.

  3. 1 mark

    A1 for 480

    cao. Accept 'about 480'. A correct answer with no working scores full marks. 0.16 or 16% given as the answer (a proportion, not a number of campers) scores M1M1A0. 300 (\(\dfrac{20}{200} \times 3000\), dividing by all 200 campers) scores M1M0A0. 2280 (\(\dfrac{95}{125} \times 3000\), not removing the forced Yes answers) scores M1M0A0. 1425 (\(\dfrac{95}{200} \times 3000\), ignoring the spinner) scores M0M0A0. 20 (the honest Yes answers in the sample) scores M1M0A0.

Question 4

4 marks

A bus company sells cheaper tickets to young people aged under 16. It wants to estimate the proportion of its adult passengers who have said they were under 16 to get a cheaper ticket. It will use the random response technique.

Design a random response question that the bus company could use. Your design must use a fair six-sided dice, and about half of the passengers must be told to answer the question honestly.
Explain how the company could use the results to estimate the proportion.

Model answer

Roll the dice where nobody else can see it.
If you roll a 1, 2 or 3, tick Yes.
If you roll a 4, 5 or 6, answer this question honestly.
'In the last year, have you said that you were under 16 to get a cheaper bus ticket?'
Yes No

About half of the passengers (those who roll 1, 2 or 3) will tick Yes because of the dice, so the company subtracts half of the number of passengers asked from the number of Yes answers. This leaves the true Yes answers from the half who answered the question, so it divides by half of the number of passengers asked. For example, if 400 passengers are asked and 230 tick Yes, the estimate is \(\dfrac{230 - 200}{200} = 0.15\).

Explanation

A complete random response design has a Yes or No question, a private random outcome and an instruction for every outcome, with some outcomes forcing the answer Yes. With a dice, three outcomes give a probability of \(\dfrac{1}{2}\), so three outcomes must lead to the question. To analyse the results, remove the expected forced Yes answers and divide by the expected number who answered the question; forcing No instead of Yes would leave every Yes as a true admission.

Mark scheme
  1. 1 mark

    B1 for a suitable question relevant to the information required, e.g. 'In the last year, have you said that you were under 16 to get a cheaper bus ticket?'

    The question must have a Yes or No answer and be about claiming to be under 16 (or claiming a young person's ticket when not entitled to one). Condone a missing time frame for this mark. B0 for a leading question or a question with a numerical answer.

  2. 1 mark

    B1 for complete instructions for the dice: tick Yes for some outcomes and answer the question honestly for some outcomes, with an instruction for every one of the six outcomes

    A design that also has a forced No for some outcomes is fine (e.g. 1 or 2 tick Yes, 3 tick No, 4, 5 or 6 answer). Condone the omission of 'in private'. B0 for a design with only forced No answers (every Yes would then be a true admission). B0 if an outcome is left out (e.g. 1 to 5 assigned but not 6).

  3. 1 mark

    B1 for exactly three of the six outcomes leading to the question being answered honestly, e.g. 1, 2 or 3 tick Yes; 4, 5 or 6 answer the question

    Dependent on the previous mark. Any three outcomes are acceptable, e.g. even numbers answer the question and odd numbers tick Yes.

  4. 1 mark

    B1 for explaining the estimate: subtract the expected number of forced Yes answers (e.g. half the passengers asked) from the number of Yes answers, then divide by the expected number who answered the question (half the passengers asked)

    ft their design if it has forced Yes outcomes, e.g. for 1 or 2 Yes, 3 No, 4 to 6 answer: (number of Yes answers \(- \dfrac{1}{3}\) of the passengers) \(\div\) (\(\dfrac{1}{2}\) of the passengers). Accept an equation such as \(\dfrac{1}{2} + \dfrac{1}{2}p = \dfrac{\text{number of Yes answers}}{\text{number of passengers}}\) with a statement that it is solved for \(p\), or a correct worked example with invented numbers. B0 for dividing by all the passengers asked.

Ready to try the rest? Every answer you write is marked against the mark scheme, so you can see exactly where the marks are.

Study Helper Hints first, answers after you try

Stuck on this one? Ask for a hint and I’ll point you in the right direction, without giving the answer away. Once you’ve checked your answer, I’ll show you how the marks are awarded.