1.5 The soup analogy: two ways to sample
The different orientations extend to how researchers select their data – what we call sampling.
Imagine you want to learn about a pot of soup.
- Quantitative sampling is like testing whether the soup is too salty. You stir the pot well and take a small, random spoonful. That spoonful represents the whole pot. If it is too salty the whole pot is too salty. The key is that the sample is random and representative – any spoonful should taste the same as any other.
- Qualitative sampling is like understanding what ingredients are in the soup. You might specifically hunt for a piece of carrot or onion, or a noodle. Those individual pieces represent the different textures and flavours present, even if they are not the ‘average’ bite. The key is that the sample is purposive – you select specific elements because of what they can reveal.
Both are valid approaches to knowing about the soup. They simply answer different questions.
In research terms:
- random sampling (quantitative) aims for representativeness – a randomly selected sample should reflect the characteristics of the wider population, allowing findings to be generalised
- purposive sampling (qualitative) aims for richness and diversity of perspectives – participants are selected because they can illuminate the phenomenon of interest, not because they are statistically representative.
When you mix methods you typically use both sampling strategies: usually, random or representative sampling for the quantitative strand and purposive sampling for the qualitative strand. These pairings are conventions rather than rules. Quantitative studies can use purposive sampling when generalisation is not the goal, and qualitative studies sometimes use convenience or snowball sampling. Understanding the conventional distinction is the foundation; understanding when to deviate from it is the next step.
Activity 1.3: Soup or spoonful?
a.
Random sampling
b.
Purposive sampling
The correct answer is a.
Answer
The correct answer is random sampling. To calculate a meaningful percentage that represents all students, you need a sample that reflects the wider population. A random sample allows you to generalise your findings.
a.
Random sampling
b.
Purposive sampling
The correct answer is b.
Answer
The correct answer is purposive sampling. Withdrawn students are a specific group whose experiences you want to understand in depth. You would deliberately seek out these individuals rather than hoping they appear in a random sample.
a.
Random sampling
b.
Purposive sampling
The correct answer is a.
Answer
The correct answer is random sampling. Comparing averages across groups requires data that represents each group fairly. Random sampling within each demographic category would support valid comparisons.
a.
Random sampling
b.
Purposive sampling
The correct answer is b.
Answer
The correct answer is purposive sampling. You are interested in the particular experiences of a specific group. You would deliberately recruit students from non-traditional backgrounds to explore their perspectives in depth.
1.4 Quantitative and qualitative: a brief introduction


