Non-random sampling (AQA GCSE Statistics): Revision Notes
Non-random sampling
Non-random sampling methods are techniques where not every member of the population has an equal chance of being selected. These methods are often used when random sampling is impractical, expensive, or impossible to carry out. Understanding these different approaches is essential for GCSE statistics.
Non-random sampling methods are particularly valuable in real-world research situations where practical constraints make pure random sampling difficult to achieve. While they may introduce some bias, they often provide more feasible approaches to data collection.
Types of non-random sampling
There are five main types of non-random sampling methods you need to know about:
1. Judgement sampling
Judgement sampling relies on the researcher's expertise and knowledge to choose a sample that they believe will be representative of the entire population. The researcher uses their professional judgement to select participants who they think will provide the most useful and accurate information.
This method is particularly useful when you need expert knowledge to identify the most suitable participants for your study.
The key advantage of judgement sampling is that it leverages expert knowledge, but it's also its main weakness - the sample quality depends entirely on the researcher's expertise and potential biases.
2. Opportunity sampling
Opportunity sampling involves selecting participants who are easily available and accessible at the time of data collection. This is often called "convenience sampling" because researchers choose whoever happens to be convenient to study.
For example, a researcher might survey people walking past a particular location, or students who are available in a classroom at a specific time.
Opportunity sampling is the quickest and cheapest sampling method, making it popular for student projects and preliminary research. However, it may not provide a truly representative sample of the population.
3. Cluster sampling
Cluster sampling works when your population is naturally divided into groups or "clusters". Instead of sampling individuals directly, you first randomly select some of these clusters, then include every single member from the chosen clusters in your sample.
This method is particularly effective when dealing with geographically spread populations, as it's more practical to study complete groups rather than scattered individuals.
4. Quota sampling
Quota sampling involves dividing the population into different groups based on specific characteristics such as age, gender, income level, or occupation. You then decide how many people you want from each group and collect data until you reach these predetermined quotas.
For instance, if you want to survey people about a new product, you might decide to interview exactly 10 adults and 10 children to get perspectives from different age groups.
Quota sampling ensures representation from different population subgroups, but the selection within each quota is often non-random, which can introduce bias.
5. Systematic sampling
Systematic sampling means selecting items from the population at regular, predetermined intervals. This could be intervals in time (every 5 minutes) or space (every 3rd house on a street, or every 10th car that passes a checkpoint).
The key feature is the regularity - you follow a consistent pattern throughout your data collection process.
Be careful with systematic sampling - if there's a hidden pattern in your population that matches your sampling interval, you could get a biassed sample. Always check for potential periodicities in your data.
Worked example: Understanding sampling methods
Worked Example: Sampling Methods for Dental Research
Question: A city has 250 dental surgeries employing over 600 dentists in total. A researcher wants to carry out face-to-face interviews with a sample of 70 dentists.
Part (a): Describe the difference between an opportunity sample and a cluster sample.
Step 1: Define opportunity sampling in context
- Opportunity sample: The researcher would interview dentists who happen to be available at the time of data collection - essentially whoever is most convenient to reach.
Step 2: Define cluster sampling in context
- Cluster sample: The researcher would randomly select a number of dental surgeries (the clusters), then interview all the dentists working at those selected surgeries.
Part (b): Explain why a cluster sample would be suitable for carrying out the interviews.
Answer: Since the dental surgeries are likely spread across different geographical areas of the city, it would be much more efficient and cost-effective to interview all dentists at a smaller number of surgeries rather than travelling to many different locations to interview individual dentists.
Part (c): Describe what the sampling frame would be in this case.
Answer: The sampling frame would be a complete list of all dental surgeries in the city, arranged in a systematic order such as alphabetical order or by postcode. This list would make it possible to take a proper random sample of surgeries.
Key concepts to remember
Essential Definitions:
Sampling frame: This is the complete list of all items in the population from which your sample will be drawn. It needs to be comprehensive and well-organised to ensure fair sampling.
Representative sample: A sample that accurately reflects the characteristics of the entire population you're studying.
Key Exam Tip: When describing sampling methods in exam questions, always explain your answer in context rather than just giving a general definition. Show how the method applies to the specific situation described in the question.
Remember!
Quick Reference - Non-Random Sampling Methods:
- Judgement sampling uses expert knowledge to select representative participants
- Opportunity sampling chooses whoever is conveniently available at the time
- Cluster sampling randomly selects groups, then includes everyone from chosen groups
- Quota sampling sets specific numbers to collect from different population subgroups
- Systematic sampling follows regular intervals in time or space when selecting participants