A population is everyone of interest. A sample is the part actually measured, because measuring everyone is usually impossible.
Step 1: Let's Learn
Read it, or press Listen and follow the words.
Randomness is the safeguard
A simple random sample gives every individual an equal chance of selection. That is what makes the sample representative.
Other good methods
Stratified sampling splits the population into groups and samples within each, guaranteeing every group appears.
Convenience sampling
Asking whoever is nearby is fast and unreliable. The people nearby are rarely typical of everyone.
Voluntary response
Online polls attract people with strong opinions, which skews the result. This bias does not shrink with a bigger sample.
Size does not fix bias
A large biased sample is a confidently wrong sample. Method matters more than size.
Why sample at all
Asking everyone is usually impossible. A sample lets you learn about a population from a part of it, provided the part is representative — and that is the entire difficulty.
Random samples are the defensible ones
In a random sample every member has a known chance of selection. Randomness is what protects against bias; careful hand-picking almost always introduces more bias than it removes.
How samples go wrong
Voluntary response attracts people with strong opinions. Convenience samples reach whoever is nearby. Undercoverage misses part of the population entirely. Each produces systematic, directional error.
Size does not cure bias
A large biased sample is a precise wrong answer. The 1936 Literary Digest poll surveyed millions and called the election wrongly, because of who it surveyed. Method beats size, always.
Step 2: Try It Yourself
Tap and try it out.
Group A
Group B
Group C
Step 3: Watch an Example
One step at a time.
Watch Yusuf Spot the Bias
A survey about library funding is conducted inside the library.
- Step 1
The population of interest is all residents, not only library users.
Step 4: Your Turn
Practice makes it stick.
The Poll
Problem 1 of 2
An online poll where readers choose to respond. Is this voluntary response bias? 1 yes, 0 no.
The Size
Problem 2 of 2
Does a larger sample fix bias? 1 for yes, 0 for no.
Judge the Method
1 of 8
Every individual has an equal chance. Is it a simple random sample? 1 yes, 0 no.
2 of 8
Asking friends in the corridor. Convenience sample? 1 yes, 0 no.
3 of 8
Sampling within each year group separately. Stratified? 1 yes, 0 no.
4 of 8
A survey in a library about libraries. Biased? 1 yes, 0 no.
5 of 8
Which matters more for accuracy? 1 method, 2 size.
6 of 8
Is the population the group actually measured? 1 yes, 0 no.
7 of 8
Which are sources of bias?
8 of 8
A phone poll at 10am on a weekday. Does it miss working people? 1 yes, 0 no.
Step 5: Quick Check
Show what you know.
Question 1 of 2
A magazine prints a survey and readers mail it back. Biased? 1 yes, 0 no.
Question 2 of 2
Why does a larger sample not fix bias?
What You Learned
- A sample stands in for a population, so it must represent it.
- Randomness is what protects against bias.
- A larger sample makes a biased result more precise, not more correct.