Welcome to Statistical Sampling!
Hi there! Welcome to one of the most practical parts of your Maths course. Have you ever wondered how news channels predict election results before all the votes are counted? Or how scientists decide if a new medicine works without testing it on every single person on Earth? The answer is Sampling.
In this chapter, we’ll learn how to pick a small group of people or objects to represent a much larger group. Don’t worry if statistics feels a bit "wordy" compared to algebra—we’ll break it down step-by-step!
1. The Basics: Population vs. Sample
Before we can start picking groups, we need to know the difference between the "whole thing" and the "part we look at."
Key Terms:
Population: The entire group of people or things you are interested in.
Example: All the students in your school.
Sample: A smaller group selected from the population that you actually study.
Example: 20 students picked from the canteen at lunch.
Census: When you collect data from every single member of the population.
Example: The government’s national census every 10 years.
Sampling Frame: A list of everyone/everything in the population. You need this list to pick a random sample.
Example: The school register.
The "Soup" Analogy
Imagine you are cooking a massive pot of vegetable soup (the Population). You want to know if it needs more salt. You don't drink the whole pot (that’s a Census)! Instead, you stir it well and take one spoonful (the Sample). If that spoonful tastes salty enough, you infer that the whole pot is fine.
Quick Review Box:
• Population = The Whole Pot.
• Sample = The Spoonful.
• Census = Drinking the whole pot (expensive and time-consuming!).
Key Takeaway: We use samples because they are quicker and cheaper than a census. If our sample is good, we can make "informal inferences"—educated guesses—about the whole population.
2. Sampling Techniques You Need to Use
The OCR syllabus requires you to understand and be able to use two specific types of sampling: Simple Random and Opportunity.
A. Simple Random Sampling
In a Simple Random Sample, every member of the population has an equal chance of being selected. It’s like putting everyone’s name in a giant hat and pulling some out.
How to do it:
1. Give every member of the population a unique number.
2. Use a random number generator (on your calculator or a computer) to pick the numbers.
3. Match those numbers back to the names in your list.
Pros: It is totally fair and unbiased.
Cons: You need a full list of the population (the sampling frame), which isn't always available.
B. Opportunity Sampling (Convenience Sampling)
This is when you simply pick whoever is available at the time.
Example: Standing outside a supermarket and asking the first 10 people who walk past for their opinion.
Pros: Very easy, fast, and cheap.
Cons: Very likely to be biased. If you stand outside a gym, your sample will probably be fitter than the average person!
Key Takeaway: Simple Random is the "gold standard" for fairness, but Opportunity is the easiest to carry out.
3. Sampling Techniques You Need to Critique
For these next four methods, you don't need to know how to calculate them, but you do need to know what they are so you can discuss their pros and cons in exam questions.
A. Systematic Sampling
Choosing people at regular intervals.
Example: Picking every \(10^{th}\) person on a list.
Critique: It's simple to do, but if there’s a pattern in the list, it might lead to a biased sample.
B. Stratified Sampling
The population is divided into groups (called strata) like age or gender. You then take a random sample from each group so that the sample represents the population's proportions.
Example: If a school is 60% girls and 40% boys, your sample should also be 60% girls and 40% boys.
Critique: It's the most accurate representation of the population, but it's complex to organize.
C. Quota Sampling
Like stratified sampling, you have groups, but you don't pick them randomly. You just find people until you've filled your "quota."
Example: "I need 10 men and 10 women; I'll just stop the first ones I see."
Critique: No sampling frame is needed, but it can be biased by the researcher's choice of who to stop.
D. Cluster Sampling
You divide the population into small groups (clusters) that should each be representative of the whole, then you pick one whole cluster to study.
Example: To study UK students, you pick 3 specific schools and interview every student there.
Critique: Much cheaper than traveling all over the country, but if your chosen clusters aren't typical, your results will be wrong.
Did you know? The word "Strata" is Latin for "layers." Think of a stratified sample like a layered cake—you want a bit of every layer in your slice!
4. Bias and Sample Size
Even with the best intentions, things can go wrong. This is called Bias.
Common Mistakes to Avoid:
1. Small Sample Size: If you only ask 2 people their opinion, you can't possibly know what the whole country thinks. Larger samples generally give more reliable results.
2. Non-response: You send out 100 surveys, but only the people with strong (often negative) opinions reply.
3. Undercoverage: Your sampling frame (your list) is missing some people (e.g., using a phone book misses people without landlines).
Sampling Variability
It is important to understand that different samples will lead to different conclusions. If two students both take a random sample of 20 people from the same school, they will likely get slightly different average heights. This isn't a mistake; it's just the nature of sampling!
Don't worry if this seems tricky: In an exam, if you are asked to "critique" a method, look for why it might not represent everyone fairly. Is it missing a group? Is it just picking "easy" people? Is the sample size too small?
Key Takeaway: A sample is only "good" if it is representative. If it favors one group over another, it is biased.
Summary Checklist
Before moving on, make sure you can:
• Define Population, Sample, and Census.
• Explain how to take a Simple Random Sample using random numbers.
• Describe Opportunity Sampling and why it might be biased.
• Identify Systematic, Stratified, Quota, and Cluster sampling.
• Explain why a larger sample size is usually better.
Great job! You've just mastered the foundations of Statistical Sampling. Ready to look at some data presentation next?