Descriptive Statistics

From a vast collection of statistical methods, one may choose those that ensure a better presentation of raw data. Those methods belong to what is generally know as Descriptive Statistics. When analyzing a collection of raw data about certain entity or process, by simply looking at the data, one may not be able to clearly see what is the highest (lowest) value, what is the most probable value, what values tend to appear more (less) frequently, what is the spread of data, etc. Descriptive Statistics supplies everything one would need for an in-depth random data presentation. It simply turns the data into information by means of summary measures   and tabular  as well as graphical  data representation.

Before generalizations (inferences) about the population can be made reliably based on a sample, one must be able to organize and present data in a convenient and understandable way. The Descriptive Statistics provides efficient tools for such an organization and presentation. Figure 2 shows a typical statistical process. First a sample is selected from the population. Next the sample is described by means of summary measures and/or frequency distribution. Finally, using the descriptors, inferential procedures are employed to reason about the population.

Figure 2. Typical Statistical Process

Examples:

Describing a qualitative sample (color preferences) [colors.xlsm]:

  

Figure 3. Typical descriptors of a catgorical (qualitative) sample.

Describing a quantitative sample (hours of travel time from Boston to New York City) [travelTime.xlsm]:

  

Figure 4. Typical descriptors of a quantitative (numeric) sample.