Population is everyone you want to learn about. A sample is the smaller group you actually measure. Research usually studies a sample and speaks about the population.
Why it matters
You can rarely measure everyone. A well-chosen sample gives reliable answers at a fraction of the time and cost.
Key ideas
- Population = the full group of interest (e.g. all adults in a district)
- Sample = the subset you actually collect data from
- Parameter = a number describing the population (μ)
- Statistic = a number computed from the sample (x̄)
- Random selection makes a sample representative
- Bigger samples give estimates that wobble less
Real-life research example
A researcher wants the average hemoglobin of pregnant women in a district. She measures 200 randomly chosen women. The 200 women are the sample; all pregnant women in the district are the population.
When to use
- The population is large, spread out or costly to reach
- You want to generalise, so select participants randomly
When NOT to use
- Don't generalise from a convenience sample (e.g. only your friends) to a whole population
- Don't treat a sample mean as the true population mean — it is only an estimate
Mini example
Population: 1, 2, 3, … 100 (mean = 50.5) Random sample of 5: 12, 47, 63, 80, 91 → sample mean = 58.6 Another draw gives another mean. That wobble is called sampling variation.
Quick takeaway
A sample is a window into the population.
Parameter = the truth. Statistic = your estimate of it.
Try it yourself
Draw several random samples below. Watch how the sample mean changes compared with the population mean (50.5). What happens when you increase the sample size?
Each sample gives a different mean but they cluster around 50.5. With larger samples the means cluster more tightly — that is why bigger samples give more precise estimates.