Chapter: Populations, Sampling Frames and Samples
Welcome to Planning and Data Collection! Whenever you see a headline like "64% of teenagers prefer streaming to live TV", have you ever wondered how researchers found that out? Did they ask every single teenager in the country? Of course not—that would take years!
In this chapter, you will learn how statisticians decide who or what to study, how to choose a fair group to test, and the crucial differences between a census and a sample. Don't worry if these terms sound formal at first; we will break each concept down step by step with everyday examples.
1. The Core Building Blocks
Before we can collect any data, we need to understand the basic building blocks of sampling. Here are the five key definitions you must know for your CCEA GCSE Statistics exam.
1. Population
The population is the entire collection of individuals, items, animals, or events that you want to find out about.
Important Exam Warning: In everyday life, "population" means the number of humans living in a city or country. In Statistics, a population can be anything! For example:
• All the oak trees in a forest.
• All the batteries manufactured in a factory on a Tuesday.
• All students currently attending your school.
The target population is the specific group that you want your final conclusions to apply to (e.g. all Year 11 pupils in Northern Ireland).
2. Census
A census is an investigation or survey that collects data from every single member of the entire population.
Example: A teacher asking every single pupil in their classroom what their favourite subject is.
3. Sample
A sample is a smaller subset or group selected from the population from which data is actually gathered.
Example: Choosing \(30\) students out of a whole school of \(800\) students to complete a survey.
4. Sampling Frame
A sampling frame is a complete, physical list, register, or database containing every individual member or item within the population from which your sample will be drawn.
Analogy: If the population is all the students in your school, the sampling frame is the official school register or database listing every pupil's name.
5. Sampling Unit
A sampling unit is a single individual item or person identified within the sampling frame.
Example: One specific pupil on the register, one lightbulb on the factory conveyor belt, or one household on a street.
Key Takeaway:
• Population: The whole group.
• Sampling Frame: The actual list of the whole group.
• Sampling Unit: One item on the list.
• Sample: The smaller group picked from the list.
2. Census vs. Sample: Which One Should You Use?
Why don't statisticians always conduct a census if it looks at everyone? Let's compare the advantages and disadvantages of each approach.
The Census
Advantages:
• Completely accurate: There is no sampling error because every member provides data.
• Completely representative: No group or individual is left out.
Disadvantages:
• Expensive and time-consuming: It takes massive resources to track down every single member.
• Impractical for large or infinite populations: Counting every grain of sand or every fish in the ocean is impossible.
• Destructive testing makes a census impossible: If a factory wants to test the lifespan of its lightbulbs or test whether fireworks explode properly, testing every item means you destroy all your products and have nothing left to sell!
The Sample
Advantages:
• Cheaper and faster: It takes far less time and money to collect and process data from a small group.
• Practical when testing causes destruction: You only test a small batch of lightbulbs or matchsticks, leaving the rest to be sold.
• Easier to manage: Requires fewer staff and resources.
Disadvantages:
• Sampling Error: Because you are only asking a subset, there will always be a natural difference or uncertainty between the sample result and the true population value.
• Risk of Bias: If your sample is poorly chosen, it might not represent the whole population fairly.
Quick Memory Aid: Remember the "Soup Spoon Analogy"! When a chef tastes a single spoonful of soup to check the seasoning, they are taking a sample. They do not need to drink the entire pot (a census) to know how it tastes—and if they did, there would be no soup left for the customers (destructive testing)!
3. Sampling Frames: What Makes a Good One?
To take a fair, random sample, you usually need a reliable sampling frame. An ideal sampling frame must have three main qualities:
1. Up-to-date: It should not contain people who have moved away, left the school, or died.
2. Complete: It must include every single member of the target population with none missing.
3. Free from duplicates: No person or item should appear twice, as this gives them a double chance of being selected.
What If No Sampling Frame Exists?
In many real-world statistical investigations, a list simply does not exist. Examiners love to ask about these scenarios!
• Wild Animals: You cannot get an official register of all the trout in Lough Neagh or all the birds in a forest. (In these cases, ecologists use alternative methods such as capture-recapture).
• Transient or Moving Populations: There is no complete list of shoppers walking through a shopping centre on a Saturday afternoon or pedestrians crossing a bridge.
Key Takeaway: If you cannot write down a complete, numbered list of every single member of your population, you do not have a sampling frame.
4. Connecting to the Statistical Enquiry Cycle
In Unit 2 of your GCSE course, you will apply the Statistical Enquiry Cycle. When planning your investigation and designing your sample, you should follow this logical sequence:
Step 1: Define the hypothesis or question
Identify clearly what you are trying to investigate.
Step 2: Define the target population
Be precise about who or what your findings will apply to (e.g. "All registered car owners in Belfast").
Step 3: Establish the sampling frame
Find or create the physical database or list from which you can choose your units.
Step 4: Determine the sample size
Balance your available budget and time against the need to reduce sampling error. A larger sample gives greater reliability but costs more.
Step 5: Select the sampling technique
Choose an appropriate method to select members (such as random, stratified, or systematic sampling).
5. Common Pitfalls & How to Avoid Them in the Exam
Don't lose easy marks! Watch out for these frequent mistakes identified by examiners:
Mistake 1: Defining a Population as "People"
Wrong: "The population is all the people in the survey."
Right: "The population is the entire group of items, objects, or individuals being studied (e.g. all lightbulbs produced by a machine)."
Mistake 2: Confusing Sampling Frame with Sample or Population
Wrong: "The sampling frame is the \(50\) people we picked." (That is the sample!)
Wrong: "The sampling frame is everyone in the country." (That is the population!)
Right: "The sampling frame is the actual list or database containing all names/items in the population from which the sample is drawn."
Mistake 3: Giving Vague Advantages
Wrong: "A sample is better because it is easier."
Right: "A sample is advantageous because it is cheaper, less time-consuming, and practical when testing is destructive."
Quick Review Summary
• Population: The entire group of interest.
• Census: Observes 100% of the population. Accurate, but expensive, slow, and impossible for destructive testing.
• Sample: A selected subset of the population. Fast and economical, but carries sampling error.
• Sampling Frame: The master list/database used to select the sample.
• Sampling Unit: A single element on the sampling frame.