Question

A travel agency offers customers a range of ways to make holiday bookings – in store, online and through their call centres. Revenues are generated through different commission rates on holidays paid by tour operators (the actual suppliers of the holiday products). The company prides itself on delivering customer satisfaction. Therefore, management is keen to research customer satisfaction levels across the different booking methods and by tour operators (some tour operators offer higher commission margins). To determine the level of customer satisfaction, the company’s management has decided to use a survey of all types of customers and have asked you to devise an appropriate sampling scheme. Explain in detail how each of the following sampling methods could be applied to the overall sampling strategy for this study. Make sure you describe the merits and limitations of each as well as how each would be applied in practice.

i. Quota sampling

ii. Snowball sampling

Answer 1

Let us check how these sampling methods could be applied:-

A) Quota Sampling

Here, the sample representing the individual is chosen as per their characteristics belong to the population and is a non-sampling probability method. Market research samples can be used to collect data by dividing it into quotas so that it could represent the entire population.

Here the entire customer population could be segmentized into subgroups. There should be a recognition of how much weightage that sample belongs to the real population. The sampling frame must be decided as per the need of the survey. Then select the appropriate sample size and survey as desired.

Merits:-

• Quicker to carry out
• Easier to carry out
• Improved representation of groups belong to the population

Demerits:-

• Sampling error would be difficult to find
• The problem of generalization may intervene in the result

B) Snowball sampling

This sampling method is used when the research participants are hard to find. Here, in this scenario, we could use this sampling method to find out the dissatisfied customers due to some peculiar fault that occurred in the entity's service. We could find that by starting from a customer who has born the ill service and isn't willing to express towards the entity but is defaming it in public which cant be controlled by the entity. Thus, for not happening this event again in the future we would find out the samples though a chain where, on leads to another.

Merits:-

• Helps in finding the samples which are hard to trace,
• Helps in maintaining the privacy nature of the survey.

Demerits:-

• Making inferences and finding out the sampling error is so difficult.