Revision Library

revisionlibrary.co.uk

GCSE · Statistics

S10 Designing Questionnaires and Surveys

EDEXCEL 1ST0 · Calculator allowed · about 75 minutes

Total Marks

Name: _______________________________ Date: ____ / ____ / ______

Answer ALL questions. Show all your working.

For each statement about questionnaires, write down whether it is True or False.

(i)A questionnaire is a set of written questions used to collect data from people.(1)

(ii)Every question in a questionnaire must give the respondent a fixed set of answers to choose from.(1)

(iii)A pilot survey is a small-scale trial of a questionnaire, carried out before it is used on the full sample.(1)

(Total for Question 1 is 3 marks)

A closed question gives the respondent a fixed set of response options to choose from, such as tick boxes. An open question lets the respondent answer freely, in their own words, with no fixed options given.

(i)State whether the question 'What is your favourite subject at school?', with no options given, is open or closed.(1)

(ii)State whether the question 'Do you walk to school? Yes / No', with these two options given, is open or closed.(1)

(iii)State whether the question 'How do you feel about the new school canteen menu?', with no options given, is open or closed.(1)

(iv)Give one advantage of using closed questions, rather than open questions, in a questionnaire.(1)

(Total for Question 2 is 4 marks)

A leading question is a question that is worded in a way that pushes the respondent towards a particular answer, rather than allowing them to give their own honest opinion.

(a)Explain, in your own words, what is meant by the term 'leading question'.(1)

(b)A questionnaire asks: 'Most people think the new bus timetable is a big improvement. Do you agree?' Explain why this is an example of a leading question.(2)

(Total for Question 3 is 3 marks)

A researcher is investigating how long teenagers spend on their phone each day. She asks: 'How many hours do you spend on your phone each day?' using the response boxes shown below.

(a)Identify the fault with these response boxes.(1)

(b)Give one value, in hours, that could be ticked in two different boxes, and explain why.(2)

(Total for Question 4 is 3 marks)

A researcher asks the question 'How much do you exercise?' in her questionnaire.

(a)Give two reasons why this question would not produce clear, useful data.(2)

(b)Rewrite the question so that it clearly states both a time frame and a unit.(2)

(Total for Question 5 is 4 marks)

Using the question in Question 4, design an improved set of response boxes for 'How many hours do you spend on your phone each day?' Your response boxes should not overlap and should cover every possible amount of time.

(Total for Question 6 is 3 marks)

A school survey asks: 'Everybody loves the new school uniform, don't you think it looks smart?' Tick one box: Yes / No.

(a)Identify why this is a leading question.(1)

(b)Rewrite the question so that it is a fair, unbiased question about the new school uniform.(2)

(Total for Question 7 is 3 marks)

A survey about favourite drinks gives respondents the response boxes shown below.

(a)Explain why these response boxes are not exhaustive.(1)

(b)Give one extra response box that could be added to make the list of options exhaustive.(1)

(c)State one problem that could occur if a respondent's true answer is not included in the response boxes.(1)

(Total for Question 8 is 3 marks)

A researcher wants to find out how much exercise, on average, people in a town do each week. She stands outside a gym at 7 am and asks the first 20 people who leave the gym.

(a)Give one reason why this sample is likely to produce biased results.(1)

(b)Suggest a more appropriate way she could select a sample to get a fair picture of exercise levels across the whole town.(2)

(Total for Question 9 is 3 marks)

A college has 1500 students. A researcher wants to find out the opinions of all 1500 students about the canteen menu, so she asks the 8 students in her own tutor group.

(a)Give one reason why a sample of 8 students from one tutor group is unlikely to give a reliable picture of the opinions of all 1500 students.(1)

(b)Suggest a better way she could select a sample of students, so that it better represents the whole college.(2)

(Total for Question 10 is 3 marks)

A town council wants to find out the opinions of all residents about a new leisure centre. It shares an online survey link only on its social media page.

(a)Give one disadvantage of only sharing the survey on social media for this aim.(1)

(b)The council also considers posting a paper questionnaire to every house in the town. Evaluate whether this would be a sensible method, giving one advantage and one limitation.(2)

(Total for Question 11 is 3 marks)

A filter question directs different respondents to different follow-up questions, depending on their answer, so that people are not asked questions which do not apply to them. For example, a questionnaire about cars begins: 'Question 1: Do you own a car? Yes / No. If Yes, go to Question 2. If No, go to Question 4.'

(a)Explain why a questionnaire might use a filter question rather than asking every respondent every question.(1)

(b)Question 2 of this questionnaire asks: 'How many miles do you drive per week?' Explain why this question should only be asked to respondents who answered Yes to Question 1.(1)

(c)Give one other reason why the order in which questions are asked in a questionnaire is important.(1)

(Total for Question 12 is 3 marks)

A student writes a questionnaire for a project. Before giving it to her full sample of 200 people, she first gives it to 10 people from her class.

(a)State the name given to this small-scale trial of a questionnaire.(1)

(b)Give two reasons why it is useful to pilot a questionnaire before using it on the full sample.(2)

(Total for Question 13 is 3 marks)

A school wants to find out how its pupils usually travel to school, in order to help it decide whether to build a new bike shed. Design one closed question, with response boxes, that the school could use to collect this information.

(Total for Question 14 is 4 marks)

A parish council questionnaire about a local park includes the question below, which has not yet been checked for problems: 'Most residents agree the park needs more bins, don't you? How often do you visit the park? Rarely / Sometimes / Often' (tick one box).

(a)Identify the fault in the first sentence of this question.(1)

(b)Identify the fault in the response boxes for 'How often do you visit the park?'.(2)

(c)Rewrite this as one fair, clear question with well-defined response boxes.(3)

(Total for Question 15 is 6 marks)

A gym manager wants to find out whether members would use a new spin class, in order to decide whether to run it. She plans to use the statistical enquiry cycle: plan, collect, process, interpret.

(a)At the plan stage, state one thing the manager should decide before she writes her questionnaire.(1)

(b)At the collect stage, give one reason why she should pilot her questionnaire on a small number of members first.(1)

(c)The manager only gives her questionnaire to members who attend the gym on Monday mornings. Explain, at the interpret stage, why her conclusions about all members might not be reliable.(2)

(Total for Question 16 is 4 marks)

A university wants to find out how satisfied all 12000 students are with the library. Method A: a researcher stands outside the library and surveys the first 50 students who enter. Method B: the university uses a computer to select 50 students at random from its full list of 12000 students, and emails them the questionnaire.

(a)Give one reason why Method A is likely to produce biased results.(1)

(b)Explain why Method B is likely to give a more representative sample than Method A.(2)

(c)Give one limitation of Method B.(1)

(Total for Question 17 is 4 marks)

A national newspaper wants to find out the opinions of adults across the whole country on a proposed new law. It posts a link to its questionnaire only on its own website.

(a)Evaluate whether this is a sensible way to distribute the questionnaire to get a fair picture of the opinions of all adults in the country. Give two reasons in your answer.(2)

(b)Suggest a more appropriate method of selecting a sample for this national survey.(1)

(Total for Question 18 is 3 marks)

A supermarket wants to find out how satisfied customers are with its new self-service checkouts, and how long customers typically spend using them, in order to decide whether to install more machines. Design two suitable closed questions, with response boxes, that the supermarket could use in its questionnaire.

(Total for Question 19 is 4 marks)

A parish council's draft questionnaire about a proposed new playground has not yet been piloted. It contains the three questions below.

Question 1: 'Everyone agrees a new playground would be great for the village, don't you think so too?' Yes / No

Question 2: 'How old are your children?' 0 - 5 / 5 - 10 / 10 - 15 / 15 or over

Question 3: 'How often do you visit the village green?' Never / Sometimes / A lot

(a)Identify the fault in Question 1 and explain why it should be avoided.(2)

(b)Identify the fault in the response boxes for Question 2.(2)

(c)Identify two faults with Question 3 (one about the response boxes, one about the wording).(2)

(d)Explain why the parish council should pilot this questionnaire before giving it to the whole village.(2)

(Total for Question 20 is 8 marks)

Mark scheme · S10 Designing Questionnaires and Surveys

Question 1

(i) B1 True cao
(i) Answer: True
(ii) B1 False cao
(ii) Answer: False
(iii) B1 True cao
(iii) Answer: True

Question 2

(i) B1 Open cao
(i) Answer: Open
(ii) B1 Closed cao
(ii) Answer: Closed
(iii) B1 Open cao
(iii) Answer: Open
(iv) B1 any valid advantage, e.g. closed responses are quicker and easier to sort and count than open answers written in someone's own words
(iv) Answer: Closed responses are much quicker to sort and count than open answers, which are all written differently.

Question 3

(a) B1 definition referring to wording that pushes/influences the respondent towards a particular answer oe
(a) Answer: A question worded so that it suggests, or pushes the respondent towards, one particular answer.
(b) B1 identifies that the statement 'most people think it is a big improvement' suggests the expected/popular answer before the respondent is asked
(b) B1 explains that this makes the respondent more likely to agree, rather than give their own honest opinion
(b) Answer: The question tells the respondent what most people already think before asking their opinion, making them more likely to agree rather than answer honestly.

Question 4

(a) B1 the response boxes overlap oe (a boundary value could be ticked in more than one box)
(a) Answer: The response boxes overlap at 2, 4 and 6 hours.
(b) B1 correctly gives 2, 4 or 6 (hours)
(b) B1 explains that this value appears at the boundary of two boxes, e.g. 2 hours fits in both '0 - 2 hours' and '2 - 4 hours'
(b) Answer: 2 hours, because it appears in both the '0 - 2 hours' box and the '2 - 4 hours' box.

Question 5

(a) B1 no time frame is stated (e.g. per day, per week)
(a) B1 no unit is given (e.g. minutes, hours, number of sessions)
(a) Answer: It does not state a time frame (such as per day or per week), and it does not give a unit for the answer.
(b) B1 includes a clear time frame, e.g. per day or per week
(b) B1 includes a clear unit, e.g. minutes or hours
(b) Answer: On average, how many minutes do you spend exercising per day?

Question 6

B1 response boxes do not overlap, e.g. using 'up to' or 'over' so each boundary value belongs to exactly one box
B1 response boxes are exhaustive, covering every amount of time from 0 hours up to an unlimited top box (e.g. '6 hours or more')
B1 boxes are clearly and consistently labelled with the same unit (hours)
Answer: 0 hours up to 2 hours; over 2 hours up to 4 hours; over 4 hours up to 6 hours; over 6 hours.

Question 7

(a) B1 identifies that 'everybody loves...looks smart' suggests/pushes the respondent towards the expected answer (Yes)
(a) Answer: It suggests that everybody already agrees the uniform looks smart, pushing the respondent towards answering Yes.
(b) B1 removes the biased/opinion-suggesting wording
(b) B1 gives a neutral closed question with suitable, non-overlapping, exhaustive response boxes
(b) Answer: What is your opinion of the new school uniform? Very positive / Positive / Neutral / Negative / Very negative (tick one box)

Question 8

(a) B1 they do not cover every possible answer, e.g. water, hot chocolate or 'none of these' are missing
(a) Answer: They do not cover every possible favourite drink, for example water or hot chocolate.
(b) B1 any suitable catch-all box, e.g. 'Other' or 'None of these'
(b) Answer: Other
(c) B1 any valid problem, e.g. the respondent may leave the question blank or tick an incorrect box, giving inaccurate data
(c) Answer: The respondent might tick a box that is not their true answer, or leave the question blank, giving inaccurate data.

Question 9

(a) B1 people leaving a gym already exercise regularly, so are not representative of the whole town
(a) Answer: People leaving a gym already exercise regularly, so they are not representative of everyone in the town.
(b) B1 a method that reaches the wider population rather than only gym-goers
(b) B1 named appropriately, e.g. a random sample selected from the electoral roll, or asking people in several different locations at different times of day
(b) Answer: Select a random sample of residents from the electoral roll, or ask people at several different locations and times, rather than only at the gym.

Question 10

(a) B1 the sample is too small and/or likely all from the same course or year group, so is not representative of the whole college
(a) Answer: 8 students from one tutor group is a very small sample, likely all the same age and course, so it does not represent the whole college.
(b) B1 a larger sample size
(b) B1 selected at random and spread across different courses/year groups, e.g. a stratified sample
(b) Answer: Take a much larger random sample, spread across different courses and year groups (a stratified sample), rather than just her own tutor group.

Question 11

(a) B1 excludes residents without internet/social media access (e.g. some older residents), so the sample may not represent all residents
(a) Answer: Residents without internet or social media access, such as some older residents, would be excluded from the sample.
(b) B1 advantage, e.g. it reaches residents without internet access, so is more representative
(b) B1 limitation, e.g. it is costly and slow to post and collect, or the response rate could still be low
(b) Answer: Advantage: it reaches residents who do not use the internet, so is more representative. Limitation: posting to every house is costly and slow, and many residents may still not respond.

Question 12

(a) B1 avoids asking respondents questions that do not apply to them, saving time and avoiding confusion
(a) Answer: It stops respondents from being asked questions that do not apply to them, saving time and avoiding confusion.
(b) B1 someone without a car does not drive, so cannot give a meaningful answer to the question
(b) Answer: Someone who does not own a car does not drive, so the question does not apply to them and they cannot give a meaningful answer.
(c) B1 any valid reason, e.g. simple/non-sensitive questions should be asked first to put the respondent at ease, with sensitive questions (such as age or income) asked later
(c) Answer: Simple, non-sensitive questions should be asked first to put the respondent at ease, leaving sensitive questions, such as income, until later.

Question 13

(a) B1 pilot survey oe (piloting/a pilot)
(a) Answer: A pilot survey
(b) B1 checks that the questions are clear and not ambiguous, and that response boxes are sufficient/exhaustive
(b) B1 identifies problems and allows them to be corrected before time and money are spent on the full-scale survey
(b) Answer: It checks that the questions are clear and the response boxes are suitable, so any problems can be fixed before the full survey is carried out, saving time and money.

Question 14

B1 question is clearly relevant to the stated aim (asks about the pupil's usual method of travel to school)
B1 response boxes are exhaustive, covering every method of travel, including an 'Other' option
B1 response boxes do not overlap and are clearly distinct categories
B1 question and boxes are fairly worded (not leading) and include a suitable time frame, e.g. 'usually'
Answer: How do you usually travel to school? Walk / Cycle / Bus / Car / Other (tick one box)

Question 15

(a) B1 it is a leading question, suggesting that most residents already agree
(a) Answer: It is a leading question: it suggests that most residents already agree the park needs more bins.
(b) B1 the terms Rarely/Sometimes/Often are vague and not clearly defined, so could mean different things to different respondents
(b) B1 no time frame is given, e.g. per week or per month
(b) Answer: The terms Rarely, Sometimes and Often are not clearly defined, and no time frame (such as per month) is given, so respondents could interpret them very differently.
(c) B1 removes the biased/leading wording
(c) B1 gives a clear time frame, e.g. per month
(c) B1 response boxes are non-overlapping and exhaustive with numeric ranges
(c) Answer: How many times do you visit the park per month? 0 times / 1 - 3 times / 4 - 6 times / 7 or more times (tick one box)

Question 16

(a) B1 any valid planning decision, e.g. the exact research aim/what she needs to find out, or who she is going to ask (the sample)
(a) Answer: She should decide exactly what she needs to find out, and who (which members) she is going to ask.
(b) B1 to check the questions are clear and the response boxes are suitable before using it on all members
(b) Answer: To check that the questions are clear and the response boxes are suitable, before giving it to all members.
(c) B1 Monday morning attendees may not represent all members, e.g. they may differ in age or availability from other members
(c) B1 this makes the sample biased, so conclusions cannot safely be generalised to every member
(c) Answer: Members who attend on Monday mornings may be a particular age group or have particular availability, so they may not represent all members, meaning her conclusions cannot safely be applied to everyone.

Question 17

(a) B1 only samples students who already use the library, so is not representative of all 12000 students
(a) Answer: It only samples students who already visit the library, so is not representative of all 12000 students.
(b) B1 every student has an equal/known chance of being selected, as it is a random sample
(b) B1 this means the sample is not restricted to students who already visit the library, so better reflects the whole student population
(b) Answer: Every student has an equal chance of being chosen, rather than only students who already visit the library, so the sample better reflects the opinions of all 12000 students.
(c) B1 any valid limitation, e.g. students who do not check their email or do not respond are missed, so non-response bias is still possible
(c) Answer: Students who do not check their email, or who simply do not reply, are missed, so the response rate could still be low.

Question 18

(a) B1 only reaches readers of that newspaper, who may share similar views or demographics, so is not a fair cross-section of the country
(a) B1 it is a voluntary-response (self-selecting) sample, so people with strong opinions are more likely to respond, and non-internet users are excluded
(a) Answer: It is not sensible: it only reaches readers of that newspaper, who may share similar views, and people choose whether to respond (a voluntary sample), so those with strong opinions are over-represented and non-internet users are excluded.
(b) B1 any valid method, e.g. a random sample drawn from across the whole adult population, stratified to reflect different regions/age groups
(b) Answer: Take a random sample of adults from across the whole country, stratified so that different regions and age groups are represented fairly.

Question 19

B1 Question 1 is a relevant closed question about satisfaction, with a suitable non-overlapping, exhaustive rating scale
B1 Question 2 is a relevant closed question about time spent, with a clear time frame/unit and non-overlapping, exhaustive response boxes
B1 neither question is leading or uses biased wording
B1 response boxes for both questions are clearly and consistently labelled
Answer: Q1: How satisfied are you with the self-service checkouts? Very satisfied / Satisfied / Neutral / Dissatisfied / Very dissatisfied. Q2: How many minutes did you spend using the self-service checkout today? 0 - 2 minutes / over 2 - 5 minutes / over 5 - 10 minutes / over 10 minutes.

Question 20

(a) B1 it is a leading question, suggesting that everyone already agrees
(a) B1 explains that this makes respondents more likely to answer in the suggested way, rather than give an honest view, so the results would be biased
(a) Answer: Question 1 is a leading question: it states that everyone already agrees, making respondents more likely to answer Yes rather than give their honest opinion, so the results would be biased.
(b) B1 the response boxes overlap at the boundaries (5 and 10 appear in two boxes each)
(b) B1 gives a specific example, e.g. a child aged 10 could be ticked in both '5 - 10' and '10 - 15'
(b) Answer: The response boxes overlap: a child aged 10, for example, could be ticked in both the '5 - 10' box and the '10 - 15' box.
(c) B1 the response boxes use vague/ambiguous terms (Never, Sometimes, A lot) that are not clearly defined and could mean different things to different respondents
(c) B1 the question does not state a time frame, e.g. per week or per month
(c) Answer: The response boxes (Never, Sometimes, A lot) are not clearly defined, and the question does not give a time frame, such as per week or per month.
(d) B1 piloting on a small group first would identify these problems, e.g. the leading wording, overlapping boxes and vague/undefined terms
(d) B1 correcting them before full distribution avoids collecting a large amount of unreliable/unusable data and wasting time and resources
(d) Answer: Piloting on a small group would reveal the leading wording, overlapping boxes and vague terms before the questionnaire is sent to the whole village, avoiding a large amount of unreliable data and wasted time and resources.

Designing Questionnaires and Surveys - Worksheets, Questions and Revision