Sampling Methods Explaineds: Random, Stratified, Systematic & Quotas

Sampling Methods Explained: Random, Stratified, Systematic & Quotas

🎧 Sampling Methods… the ones everyone thinks they remember

Right—sampling. This topic seems easy until you actually have to explain it to someone out loud and suddenly realise your brain has quietly merged stratified and quota sampling into the same fuzzy blob. Happens every year.

But let’s slow it down and walk through them like you’re sitting in class and I’m waving a whiteboard marker around trying to keep everyone awake.

And yes, getting comfortable with these ideas is a big part of cracking A Level Maths topics explained. These sampling questions appear in stats papers constantly.

Alright, enough scene-setting—let’s get into the actual stuff.

🔙 Previous topic:

Before carrying out hypothesis testing using normal distributions and z-scores, it is essential to understand how data is collected, which is why we now look at sampling methods.

📘 Where examiners use this

Examiners use sampling methods to test whether you really understand how data is collected—not just how it’s analysed.

A lot of students think it’s “just definitions”, but nope, examiners want:

reasons for using each method
strengths and weaknesses
why some methods give bias
examples that actually make sense

And the moment someone writes “stratified is random” without explaining how, examiners start sharpening their marking pens.

📏 What we’ve got (Problem Setup)

Imagine a school wants to survey students about revision habits. There are year groups, genders, subjects… all sorts of subgroups floating around.

We want a sample that represents the school properly—something like:

For example, $N = 1200$ students overall.

Now we need to pick a method for selecting our sample. That’s where the four main sampling types come in.

🖼️ Required Diagram

Let me drop in a simple visual—just a “shape of the idea” representation of how samples might be selected. Doesn’t matter that your exam won’t give you one; it helps just to picture it.

🧠 Let’s break this apart (Key Ideas Explained)

🎲 1. Simple Random Sampling

This is the one people think of as the default. Everyone in the population has an equal chance of being chosen.

Two main ways to do it:

Number everyone and use a random number generator
Pick names from a hat (well… metaphorically, unless you like papercuts)

For example, if 1200 students are numbered 1–1200, you could use a random generator to pick 50 unique numbers.

Why examiners like it:
It’s the cleanest in theory—no bias if done correctly.

Real teacher note:
Students often forget that random ≠ convenient. Taking “the first 50 students I meet in the corridor” is not random, even if it feels accidental.

🧊 2. Stratified Sampling (the organised sibling)

This method is basically:
“Split the population into categories beforehand and sample within each category in proportion.”

For example, if the school has:

40% Year 12
60% Year 13

…and you want a sample of 50, then:

For example, $20 \text{ from Year 12},;30 \text{ from Year 13}$ .

Inside each stratum, you choose randomly.

Why this is good:
It represents subgroups accurately. Examiners LOVE when you state that:
“It produces a more representative sample when subgroups differ.”

Where people slip up:
They forget the proportional bit and just grab the same number from each stratum. Nope—that’s quota sampling. Different thing entirely.

📏 3. Systematic Sampling (the every-kth method)

Ah, systematic sampling—the method that sounds suspiciously easy.

Steps:

Pick your sample size.
Calculate the step size (k).
Choose a random starting point.
Take every (k)th person.

For example, $k = \frac{1200}{50} = 24$ .

So you pick a random starting student—say number 17—and then sample 17, 41, 65, 89… until you’ve got your 50.

Where this fails:
Patterns. If the population list itself has a cycle (like being sorted by ability or alphabetically by tutor group), systematic sampling accidentally amplifies the pattern.

I once had a student sample every 20th person from a list sorted by height. Shockingly, he ended up with a sample full of taller students. Who knew?

🧮 4. Quota Sampling (the non-random one)

This is the one everyone quietly forgets until the night before the exam. Quota sampling sets target numbers for categories, like:

10 first-years
10 second-years
10 third-years

But the researcher chooses participants non-randomly—often whoever is easiest to find.

So it’s proportional like stratified, but not random like convenience sampling.

Good:
It’s quick. Cheap. Practical in the real world.

Bad:
Potentially biased. Heavily.

Examiners jump on the line: “Quota sampling is not random and may introduce selection bias.”

Say that and you’ll pick up marks.

💭 Quick pause

Let me drop the mid-section anchor right where it naturally fits: sorting out which method suits which scenario becomes much easier when you practise with mixed examples inside A Level Maths revision essentials . It forces you to spot the structural differences rather than memorise definitions.

Okay—back to the flow.

❗ Mistakes people make (Common Errors & Exam Traps)

Mixing up stratified and quota sampling (“they both split into groups” is not enough).
Forgetting that systematic needs a random start.
Assuming convenience = random (it’s not… ever).
Saying simple random sampling is “always the best” (sometimes impractical).
Not explaining why a method reduces bias.
One optional LaTeX note: For example, $k = N/n$ gives the step size for systematic sampling.

🌍 Why this isn’t abstract (Real-World Link)

Every survey you’ve ever seen—political polling, market research, medical studies—uses some variation of these methods.

Pollsters almost never take pure random samples because it’s too slow; they use stratified or quota.
Factories often use systematic sampling on production lines.
Universities often use stratified methods for student surveys.

This topic is basically “how real data gets collected”, simplified for A Level.

🚀 Next step forward

If you want a bunch of worked, exam-style sampling questions (including the subtle ones where stratified and quota look identical until line three), the A Level Maths Revision Course with guided practice is the fastest way to make these feel automatic.

📏 Optional Table

Simple random: everyone equal chance.
Stratified: split into categories and sample proportionally.
Systematic: pick every (k)th; watch for patterns.
Quota: set group targets; non-random; fast but biased.
For example, $k = \frac{N}{n}$ for systematic sampling.

Author Bio – S. Mahandru

I’m a maths teacher who has spent a suspicious amount of time drawing little boxes representing sample strata on the whiteboard. If you like stats explained in a way that feels like someone talking through real examples rather than reciting definitions, stick around.

🧭 Next topic:

Once data has been collected using appropriate sampling methods, it can then be analysed to identify relationships between variables using correlation and regression.

❓ FAQ

How do I know which sampling method to choose in an exam?

Honestly, read the scenario. If different subgroups matter (like year groups, genders, locations), stratified is usually the right answer. If time or practicality is a problem, quota shows up. If the population is listed physically (like seats on a bus), systematic makes total sense. And if the question literally says “simple random sample”, do what it says and pick randomly.

Why is systematic sampling risky sometimes?

Because any underlying pattern can distort your sample. Imagine people are listed alphabetically by form group, and ability levels cluster by form. If you take every 30th student, you might accidentally sample mostly high-achievers or low-achievers depending on that pattern. Always mention “may be biased if the list has a pattern”.

How big should a sample be??

A Level questions rarely require a calculation here — they usually just want you to comment. Bigger samples reduce variability but cost more time and money. Small samples are quick but less reliable. If the question says “justify your sample size”, mention practicality and accuracy.