Decisions about sample size is an important issue because samples that are too large tend to waste time, resources and money whereas samples that are too small may lead to inaccurate results. According to Neuman, (2006) a researcher’s decision about the best samples size is guided by; (a) the degree of accuracy required (b) the degree of variability or diversity in the population and (c) the number of different variables examined simultaneously in data analysis. He also relates that everything being equal, large samples is needed if one wants high accuracy and is needed when the population has a greater deal of variability and there is the need to simultaneously analyse many data variables. In contrast to the former, smaller samples is sufficient when less emphasis is placed on accuracy and also when there is homogeneity in the population.
The issue of sample size can be addressed from two perspectives: one is to make assumptions about the population and use statistical equations about random sampling processes. A second method is a rule of thumb – a conventional or a commonly accepted amount (Neuman, 2006).
In this research a rule of thumb was used to select a suitable sample size because we rarely have the information required by the statistical method and the rule of thumb method gives sample sizes relatively close to those derived from statistical method. According to Neuman, (2006) for small populations (under 1,000), a researcher needs a large sampling ratio (about 30 percent) to give a higher degree of accuracy. Since our final sampling frame consisted of 206 hotels and falls under 1000, we chose a large sampling ratio of 50 percent representing a sample size of 103. We also believe that, with this sample size a higher degree of accuracy or representation will be achieved and the results could be generalized as in the case of most probability sampling. Moreover sample size of one hundred and three (103) respondents was selected because of cost and time constraints. Using all the 206 members in our sampling frame in this survey would require large financial resources which we could not afford. Again, the time limit within which the research was to be completed would not permit the use of larger sample size.