Worked example · Music · Year 7

Designing a survey of music listening habits

LNF: Numeracy → Statistics and probability → Collecting data

What this is

A secure Year 7 response, annotated with the success criteria each part meets. It is also the answer key. Key answers: Step 1, “most” of 30 is at least 16. Step 3 faults: 2 O and G, 3 G, 4 T, 5 V, 6 V, 7 P, 8 G. Step 7: Q1 A 5, B 11, C 9, D 5; Q2 streaming app 15, video website 7, radio 4, CD or vinyl 1, other 3; Q5 Yes 13, Maybe 10, No 7. Every table totals 30. Step 8: 9 + 5 = 14 pupils listen for more than 1 hour, which is not most, so the hypothesis is not supported. Pupils’ own questions and samples will differ: accept any that follow the rules.

Step 1: question or hypothesis?

Hypotheses: the 2nd and 4th statements. The 1st is a question and the 3rd is an opinion that data cannot prove.

(a) At least 16, because half of 30 is 15 and “most” means more than half.

(b) Pupils who play an instrument are more likely to want to come to a lunchtime music club than pupils who do not.

Each hypothesis is a statement with something countable in it, and “most” is turned into a number before any data is collected. Evidences criterion 1.

Step 2: what type of data?

QualitativeQuantitative: discreteQuantitative: continuous
(a) (e) (h)(b) (d) (g)(c) (f) (i)

(j) Pupils cannot time their listening exactly, so a box they can tick quickly is more honest, and 4 boxes are much easier to tally than 30 different times.

Counted things (songs, instruments, concerts) are discrete; timed things are continuous; words are qualitative. Evidences criterion 2.

Step 3: fix the faulty questions

No.FaultBetter version
2O, GOn a school day, how long do you usually spend listening to music? 0 to 30 minutes · more than 30 minutes, up to 1 hour · more than 1 hour, up to 2 hours · more than 2 hours. (1 hour and 2 hours were in 2 boxes, and 4 hours fitted none.)
3GHow many songs did you listen to yesterday? 0 · 1 to 10 · 11 to 20 · 21 to 30 · more than 30. (0, 11 to 14 and 26 to 29 had no box.)
4TSplit it: Do you play a musical instrument? Yes · No. Would you come to a lunchtime music club? Yes · Maybe · No.
5VIn the last 7 days, on how many days did you listen to music? 0 · 1 to 2 · 3 to 4 · 5 to 6 · 7
6VHow many concerts or gigs have you been to in the last 12 months? 0 · 1 · 2 to 3 · 4 or more
7PDelete it: the survey does not need to know who anyone is. If forms matter, ask “Which form are you in?” with one box per form.
8GWhich style of music do you listen to most? Pop · Hip hop or rap · Rock · Welsh-language · Film or classical · Other

Every fault is named, and every new set of boxes is checked: each possible answer fits exactly one box. “Most” in question 8 stops pupils ticking several. Evidences criteria 3 and 4.

Step 4: design your questionnaire

No.QuestionAnswer boxesType of data
1On a school day, how long do you usually spend listening to music?0 to 30 minutes · more than 30 minutes, up to 1 hour · more than 1 hour, up to 2 hours · more than 2 hourscontinuous
2On a Saturday, how long do you usually spend listening to music?the same 4 boxes as question 1continuous
3How do you listen to music most often?Streaming app · Video website · Radio · CD or vinyl · Otherqualitative
4How many musical instruments can you play?0 · 1 · 2 · 3 or morediscrete
5Which lunchtime would suit you best for a music club?Monday · Tuesday · Wednesday · Thursday · Friday · Nonequalitative
6What would you most like to do at a lunchtime music club?(write your answer)qualitative

Box test: 0 minutes: 0 to 30 · 30 minutes: 0 to 30 · 1 hour: more than 30 minutes, up to 1 hour · 61 minutes: more than 1 hour, up to 2 hours · 3 hours: more than 2 hours.

Question 1 tests the hypothesis; question 2 covers the weekend, which question 1 misses. “Up to” and “more than” make the edges exact, so 1 hour can only go in one box. Evidences criteria 2 and 4.

Step 5: who will you ask?

(a) 30 pupils. 30 out of 120 = 1/4 = 25%.

(b) List all 120 Year 7 pupils in alphabetical order and ask every 4th pupil, or pull 30 names from a hat. That gives pupils from all 4 forms an equal chance.

(c) My friends probably like the same music as me, so the results would not represent the whole year group. They might also give the answers they think I want.

The sample size is a sensible fraction of Year 7, the method gives everyone the same chance, and the bias from asking friends is explained. Evidences criterion 5.

Steps 6 and 7: tally table and collecting the data

Time on a school day (Q1)TallyFrequency
A: 0 to 30 minutes||||5
B: more than 30 minutes, up to 1 hour|||| |||| |11
C: more than 1 hour, up to 2 hours|||| ||||9
D: more than 2 hours||||5
Total30
Q2TallyFreq.Q5TallyFreq.
Streaming app|||| |||| ||||15Yes|||| |||| |||13
Video website|||| ||7Maybe|||| ||||10
Radio||||4No|||| ||7
CD or vinyl|1Total30
Other|||3
Total30

The table is ruled, has headings, writes each box in full and has a Total row. Tallies are in bundles of 5 and every total is 30, so no questionnaire was missed or counted twice. Evidences criterion 6.

Step 8: is the hypothesis supported?

(a) Boxes C and D: 9 + 5 = 14 pupils.

(b) 14/30 = 7/15. Half of 30 is 15, so 14 is less than half.

(c) The hypothesis is not supported. Only 14 of the 30 pupils listen for more than 1 hour on a school day, while 16 listen for 1 hour or less, and “most” would need at least 16. However, Q1 only asks about school days, so pupils may listen for much longer at weekends, and the times are estimates.

(d) 13 × 4 = 52 pupils, about 52 of the 120 (plus some of the “Maybe” pupils).

The decision uses the frequencies, compares them with the 16 worked out in Step 1, and names a limit of the data. Evidences criterion 7.

If you finish early: answers

Why this response is secure