Applied Statistics: Random Sampling
Statistical techniques are primarily used to manage data. For example, statistical studies might try to make sound generalizations about the characteristics of a given population. Rather than analyze the entire group, however, such conclusions are typically drawn from more manageable population samples. In order for the generalizations to be valid, these samples must be random.
Random sampling can be defined as "a method of selecting a sample in such a way that every possible sample has the same probability of being selected." Populations may be either finite or infinite. Finite populations consist of a fixed number of elements; whereas, infinite populations have a limitless number of values. With finite populations, random samples can be determined through the use of random numbers. Such numbers may be either obtained from tables or generated by computer. Individual items within a population can then be selected from numbered lists. With infinite populations, a sample is considered random if it has a corresponding distribution of independent random variables. This ensures that the sample represents the particular infinite population's overall characteristics.
Most any survey or poll will employ some form of random sampling. As an example, the popularity of television programs is based on "ratings." These ratings represent a survey of what shows people prefer to watch. The survey is based on a random sample of viewers.
Applied Statistics: Frequency Distribution
Frequency distributions are yet another technique by which statistics can simplify large quantities of data. This method typically groups or classifies data in order to more clearly convey certain information.
The construction of a frequency distribution is fairly simple. First, it is necessary to determine the number of classes that will be used to group the data. Second, the data must be so...