Research Methods in Tourism, Hospitality and Events Management

Simple and systematic random sampling

Simple random sampling is a method which is similar to a lottery. In the case of a

known list of customers of, say, 100,000 names, a sample of 1,000 could be drawn

by placing all the names on paper, placing them in a very large hat and drawing

1,000 from it. Such a method in practice would be somewhat cumbersome and

computers are able to generate random numbers to achieve the same goal. In a

survey, it would be necessary to draw 1,000 different names, thus once a name had

---

been selected it would not be eligible for selection again. In theoretical terms,

statisticians would start to complain about this as the second name drawn from that

hat would have a 1 in 99,999 chance of being drawn unless the first name was

replaced in the hat. However, in surveys it is not practical or desirable to interview

the same person more than once. Obviously, some common sense and judgement are

necessary to be fair yet pragmatic. If random numbers are used, then the next random

number selected should suffice if duplication occurs.

To overcome some of the potential pitfalls with simple random sampling, the

selection of respondents can be carried out in a more systematic way. Taking the

same scenario, where the population is known, if a sample of 1,000 is required from

the list of 100,000 names, then every 100th name could be taken. The starting point

could be determined randomly and then every 100th name selected from this point.

Simple and systematic samples are used where the members of a population are

similar or homogeneous. No account is made of any particular characteristic of the

population and every member of the population is given an equal chance of inclusion

in the survey. As we have already seen, in tourism, hospitality and events, full

knowledge of the population may not be possible or it may not be homogenous. If

the latter aspect is the case, then the following method of sampling may be more

appropriate.