Non-random sampling (Edexcel GCSE Statistics): Revision Notes
Non-random sampling
Non-random sampling methods are essential techniques in data collection where researchers don't use random selection to choose their samples. Instead, they use specific strategies based on convenience, judgement, or systematic approaches. Understanding these methods is crucial for GCSE statistics as they each have distinct characteristics and applications.
Unlike random sampling where every member of the population has an equal chance of being selected, non-random sampling uses deliberate strategies to choose participants based on specific criteria or practical constraints.
The five main types of non-random sampling
Judgement sampling
This method relies on the researcher's expertise and knowledge to handpick a sample they believe will best represent the entire population. The researcher uses their professional judgement to select individuals or items that they think will provide the most accurate representation of the group being studied.
Example: School Survey Selection
A headteacher selecting specific classes to survey about school meals because they know these classes represent a good mix of year groups and backgrounds.
Opportunity sampling
Also known as convenience sampling, this approach involves selecting whoever or whatever is readily available at the time of data collection. It's the most straightforward method but can introduce significant bias since it only captures those who happen to be present.
Example: Shopping Centre Survey
A researcher standing outside a shopping centre and asking people walking by to complete a questionnaire about shopping habits.
Cluster sampling
This technique is particularly useful when the population naturally exists in groups or clusters. The researcher randomly selects some of these clusters and then includes every single member from the chosen clusters in their sample. This differs from other methods because once a cluster is selected, everyone in that cluster participates.
The key distinction of cluster sampling is that it involves two stages: first randomly selecting clusters, then including ALL members from the chosen clusters.
Example: Student Attitude Survey
A researcher studying student attitudes might randomly select 3 schools from a list of 20 schools, then survey every student in those 3 selected schools.
Quota sampling
This method involves dividing the population into different subgroups based on specific characteristics (such as age, gender, income level, or location), then selecting a predetermined number of participants from each subgroup. The researcher ensures they meet their quota for each category.
Example: Market Research Quotas
A market research company might need to interview exactly 50 teenagers and 50 adults about a new product, ensuring they get equal representation from both age groups.
Systematic sampling
In this approach, researchers select items or people from the population at regular, predetermined intervals. This could be based on time intervals (every 10th minute) or spatial intervals (every 5th house on a street). The key is maintaining consistent spacing between selections.
Example: Regular Interval Sampling
Surveying every 7th person entering a train station, or testing every 20th product coming off a production line.
Worked example: Understanding the differences
Worked Example: Comparing Opportunity and Cluster Sampling
Scenario: A city contains 250 dental surgeries employing over 600 dentists in total. A researcher wants to conduct face-to-face interviews with a sample of 70 dentists.
Opportunity sampling approach The researcher would interview dentists who are readily available at the time of the study. This might mean visiting a few conveniently located surgeries and speaking to whoever is free.
Cluster sampling approach
The researcher would randomly select a small number of dental surgeries from the complete list, then interview all dentists working at those selected surgeries.
Why cluster sampling works better here: Since the dental surgeries are spread across the city geographically, cluster sampling is more efficient. Rather than travelling to many different locations to find individual dentists, the researcher can visit fewer locations but interview more people at each one.
Sampling frame: For this study, the sampling frame would be a complete list of all dental surgeries in the city, arranged in alphabetical order or by postcode. This organised list makes it easier to select a proper random sample of surgeries.
Practice scenarios
Practice Example: Systematic Sampling in Action
Scenario: A street has 160 houses (80 on each side), arranged in 20 blocks of 4 houses each. A researcher wants to systematically sample 40 houses.
Simple approach: Number all houses from 1 to 160, then select every 4th house (since ).
Potential disadvantage: If there's a pattern in the street layout (for example, every 4th house is a corner house with different characteristics), this could bias the results.
Improvement: Rather than using a fixed interval, the researcher could randomly select 2 houses from each of the 20 blocks, ensuring better representation across the entire street.
Practice Example: Quota Sampling for Market Research
Scenario: A marketing company needs quick results about a new perfume. They have budget to interview 40 people and know that about 70% of expensive perfume is bought by women and 30% by men.
Why quota sampling works: The company can set quotas of 28 women and 12 men to match the known purchasing patterns. This ensures their sample reflects the target market's characteristics.
How to implement: Researchers could position themselves in shopping areas and interview people until they reach their quota for each gender group.
Common exam tips and traps
Essential Exam Strategies
When answering questions about non-random sampling methods:
-
Always explain in context: Don't just give generic definitions. Relate your answer to the specific scenario presented in the question.
-
Consider practical constraints: Think about factors like time, cost, and accessibility when explaining why a particular method might be chosen.
-
Identify advantages and disadvantages: Each method has trade-offs between convenience and potential bias.
-
Understand sampling frames: This is simply an organised list of the population from which the sample is drawn. Be specific about what this would look like for each scenario.
Key Points to Remember:
-
Non-random sampling uses specific strategies rather than random selection to choose samples
-
Judgement sampling relies on expert knowledge to select representative participants
-
Opportunity sampling uses whoever is conveniently available at the time
-
Cluster sampling randomly selects groups, then includes everyone in those chosen groups
-
Quota sampling ensures specific numbers from different subgroups are represented
-
Systematic sampling selects participants at regular intervals in time or space