Pilot surveys and random response method Revision Notes for Edexcel GCSE Statistics

Pilot surveys and random response method

What are pilot surveys?

A pilot survey is essentially a "practice run" of your questionnaire before you conduct the full survey. Think of it like a dress rehearsal before the main performance. The main purpose is to test your questions on a smaller group of people to make sure they work properly.

infoNote

Pilot surveys help you check that:

People understand what you're asking
The questions are clear and not confusing
You can collect all the data you actually need
Your results will be valid and useful

You typically carry out a pilot survey on a small portion of your total sample population - just enough people to spot any problems before it's too late to fix them.

Why should you use a pilot survey?

Understanding the practical benefits of pilot surveys is essential for conducting reliable research. Let's examine a real-world scenario to see these benefits in action.

lightbulbExample

Worked Example: Abbi's Insurance Questionnaire

Abbi wants to investigate insurance claims using a questionnaire. She should carry out a pilot survey for two main reasons:

1. To spot problems with wording or response options - She might discover that people are confused by certain questions or that the answer boxes don't cover all possible responses.

2. To ensure the questionnaire collects the information needed - She might find that her questions don't actually gather the data required for her investigation.

The random response method

When do you need this method?

chatImportant

Sometimes people don't answer questions truthfully, especially when the questions are about sensitive or embarrassing topics. This means your survey results won't be accurate. The random response method is designed to solve this problem by using a clever technique involving chance.

How does the random response method work?

The method uses a random event (like tossing a coin) to help people give truthful answers without feeling embarrassed. The key principle is providing plausible deniability so participants feel safe to respond honestly.

Basic setup:

You give participants a sensitive question
You also give them a random instruction (like "toss a coin")
Depending on the random outcome, they either answer truthfully or give a predetermined response

Worked example: Breaking and not owning up

Let's examine how this method works in practice with a step-by-step calculation.

lightbulbExample

Worked Example: Breaking and Not Owning Up

You want to find out how many people have broken something and not owned up to it. You give people this instruction:

"Roll a fair dice. If you get a 6, tick box A. If you get 1, 2, 3, 4, or 5, answer this question truthfully: Have you ever broken something and not owned up to it? If yes, tick box A. If no, tick box B."

The results:

432 people ticked box A
468 people ticked box B
Total sample: 432 + 468 = 900 people

Step-by-step calculation:

Step 1: Work out how many people rolled a 6

Probability of rolling a 6 = $\frac{1}{6}$
Estimated number who rolled a 6 = $\frac{1}{6} \times 900 = 150$ people

Step 2: Find the number who answered truthfully

People who ticked A and were truthful = $432 - 150 = 282$ people

Step 3: Calculate the proportion

People who answered the question truthfully = $900 - 150 = 750$ people
Estimated proportion who had broken something = $\frac{282}{750} \approx 0.376$ (or about 37.6%)

Another worked example: Pretending to be someone else

lightbulbExample

Worked Example: Pretending to be Someone Else

Question: "Have you ever pretended to be someone else?" Instructions: "Flip a coin. If you get heads, answer 'Yes'. If you get tails, answer truthfully."

Results:

520 people answered Yes
480 people answered No
Total: 1000 people

Step-by-step calculation:

Step 1: Expected number who got heads: $1000 \div 2 = 500$ people

Step 2: Number who answered 'Yes' truthfully: $520 - 500 = 20$ people

Step 3: Number who answered the question: $1000 - 500 = 500$ people

Step 4: Proportion who pretended to be someone else: $\frac{20}{500} = 0.04$ (or 4%)

Key points about the random response method

infoNote

Benefits of the Random Response Method:

It protects privacy - Individual responses can't be traced back to specific people
It encourages honesty - People feel safer answering sensitive questions
It uses probability - The random element allows you to calculate estimates
It requires larger samples - Because some responses are random, you need more participants to get reliable results

Common exam tips

chatImportant

Essential Exam Strategies:

Always show your working - Break down calculations into clear steps
Remember the formula structure - (Total ticking sensitive box) - (Expected from random event) = (Truthful responses)
Check your denominator - When calculating proportions, use only the people who answered truthfully, not the total sample
Understand why it works - The method gives people "plausible deniability" so they can answer honestly

bookmarkSummary

Key Points to Remember:

Pilot surveys are like dress rehearsals - they help you test your questionnaire before the main survey to spot problems and ensure you collect valid data
Random response method tackles sensitive questions - it uses random events like coin tosses to help people answer truthfully without embarrassment
The maths involves three steps - calculate random responses, subtract from total responses, then find the proportion of truthful answers
Always subtract the random element - don't include people who followed random instructions when calculating your final proportions
Larger samples are needed - because some responses are random, you need more participants to get reliable estimates

Pilot surveys and random response method (Edexcel GCSE Statistics): Revision Notes