Before Designing Samples

METHODOLOGY samples provide guidance in selecting a small group of subjects / objects that can be representative of the population. The essence of the sample is representative of the population.

Before performing the sampling Eriyanto (2007) mentions a few things that should be taken into consideration before mendesai sample of the survey objectives, survey timing, approach, budget (budget).

Design samples for research aimed at exploration, in contrast to the description of research or analysis. Exploratory research is a kind of exploratory research. In this study there was no specific problem that we wanted to answer. From the name envisaged that this kind of research just started mapping the subject / object to be studied. The study illustrates the objective description of certain aspects of the subject / object of research. While research in-depth analysis of two previous studies. This study has attempted to search or examine the relationship antarvaribel or figure out the background to the phenomenon. Samples were chosen to be adapted to the purpose of the survey.

There are at least three times the survey: snapshot, panels, and tracking. In a snapshot, the sampling is done only once. While the panel, selected samples in a given time series. Misalanya, after a year of working president, we wanted to examine how society menilaian on performance. Similar survey we did again when the second, third, fourth, and fifth reign. They were selected as the sample is the same from year to year. Similar to the panel, tracking is also done in a certain period of time are patterned. The difference, in tracking samples taken not to be the same from time to time. So they were sampled in the first year, may not necessarily be sampled in the next year.

Before performing the sampling, we also have to consider whether the results obtained from these studies will be generalized or not. If you want to generalize, then we must choose the method of probability which allows all members of the population to be selected as a sample. If you do not want to generalizing the results of the survey, we can use non-probability approach. In this approach, not all members of the population have a chance of being sampled elected. They were chosen deliberately by researchers with specific scientific considerations.

Finally, do not forget to consider the suitability of the research budget with the sampling method we choose. The method uses a lot of the cost is not necessarily the best method. Quality Sampling (once again) lies in its reliability in representing the population