Step 10 of 18 · Worked example · TEACH

← Statistics & Probability

0 of 3 done

Work through the example to CHECK to open the problems.

Reach CHECK to open the problems.Higher Level · Statistics · Step 10

Name each method by what is picked, then judge it by who could be left out or not heard.

A sports club has 1,250 members: 500 juniors, 625 adults and 125 seniors. The committee wants a sample of 50 to gauge opinion on a fee change. (a) Design a stratified sample. (b) Name the method and main weakness of each plan: A, pick 2 of the 50 training groups at random and ask all 25 members of each; B, survey every 25th name on the membership list and use the replies; C, ask the first 50 members into the clubhouse on Saturday.

A sports club has 1,250 members: 500 juniors, 625 adults and 125 seniors.

The committee wants a sample of 50 members to gauge opinion on a fee change.

(a) Design a stratified sample.

(b) Name the method and the main weakness of each plan:

A: pick 2 of the club's 50 training groups at random and ask all 25 members of each.

B: send the survey to every 25th name on the membership list, and use the replies.

C: ask the first 50 members into the clubhouse on Saturday.

Read it once, slowly. Then copy it onto your page.
  • What is going on? The population is all 1,250 members, in three age groups. 500 + 625 + 125 = 1,250: each member is in one group.
  • (a) wants: 50 members, chosen as a stratified sample. One calculation.
  • (b) wants: for each plan, the method's name and its main weakness. Plan A picks groups, B uses a list, C uses whoever turns up.
Céim

Higher Level Maths · Statistics & Probability · Step 10 of 18

Sampling Methods, Bias & Representativeness

Judge a sample by how it was chosen: name the method by what is picked, say who could be left out or not reply and which way that pushes the result, and give each group its share in a stratified sample.

Why it matters

Bias cannot be repaired by a bigger sample or better arithmetic; it has to be prevented in the design.

The worked example

A sports club has 1,250 members: 500 juniors, 625 adults and 125 seniors. The committee wants a sample of 50 members to gauge opinion on a fee change. (a) Design a stratified sample. (b) Name the method and the main weakness of each plan: A: pick 2 of the club's 50 training groups at random and ask all 25 members of each. B: send the survey to every 25th name on the membership list, and use the replies. C: ask the first 50 members into the clubhouse on Saturday.

How the worked example goes

  1. Read

    • What is going on? The population is all 1,250 members, in three age groups. 500 + 625 + 125 = 1,250: each member is in one group.
    • (a) wants: 50 members, chosen as a stratified sample. One calculation.
    • (b) wants: for each plan, the method's name and its main weakness. Plan A picks groups, B uses a list, C uses whoever turns up.
  2. Understand

    • (a) Why stratify?
    • Juniors, adults and seniors may feel differently about fees. A good sample has the same age mix as the club.
    • Stratified: split the population into groups (strata), then pick at random inside each, in proportion to its size.
  3. Reduce

    • How is the sample of 50 split? 20 J, 25 A, 5 S: Each group's share of the club, applied to 50: 0.4, 0.5 and 0.1 of 50.
    • (a) Where do I start?
    • With each group's share of the club. Then take that share of 50.
  4. Solve

    • (a) Work out each share.
    • Juniors: 500 ÷ 1250 = 0.4; 0.4 × 50 = 20
    • Adults: 625 ÷ 1250 = 0.5; 0.5 × 50 = 25
  5. Check

    • (a) Is it right?
    • 20 + 25 + 5 = 50 ✓
    • Same mix as the club: 20 ÷ 50 = 0.4, 25 ÷ 50 = 0.5, 5 ÷ 50 = 0.1 ✓

Common mistakes to avoid

Key terms

simple random · stratified · cluster · quota · bias · representative · generalisability