1. WHAT IS STATISTICS, AND WHAT ARE STATISTICS?

"Statistics" (singular) is the body of techniques for dealing with "statistics" plural. Each technique has its own special capabilities and unique instances of applicability. The family of techniques can assist in the accomplishment of description, explanation, prediction and control.

• Description -- Some of the techniques have rather amazing abilities to describe phenomena. They can find patterns of relationships where the human eye may fail to do so.
• Explanation -- Other techniques have special capabilities to explain or to help explain relationships. For instance, they can isolate the impact of one variable on another; show relative differences between the magnitude of impact for two or more variables; or even reveal how one set of variables impinges on another set.
• Prediction -- Certainly every decision maker desires the ability to predict, and statistical techniques afford that promise. Certain statistical analyses are explicitly designed to yield predictive models with specific levels of accuracy.
• Control -- Finally, certain statistical techniques can be brought to bear on the age-old desire to control. Of course, the techniques themselves will not institute control, but their application can help decision makers develop cause-and-effect relationships and models which will ultimately enhance their ability to control events to some degree.

• How do you efficiently sample a population?
• How is the population grouped into life-style/buying-habit clusters, and how do marketers reach them with media?
• How can probability and statistics assist in the evaluation of alternative projects or investments?
• How do you continuously improve a production or service process?
• How do you evaluate credit applicants?
• How do you determine what characteristics potential consumers want in a new product being developed?
• To what extent does compensation/salary determine employee job satisfaction? What other job attributes contribute to satisfaction, and to what extent?
• How do you forecast production costs?
• How do you forecast demand for a product? How does this relate to inventory planning?

2. DATASETS

It is commonly the case that several variables are observed for a sample of individuals. The result is often called a "data set." It is customary to envision a data set as being comprised of rows and columns. The rows pertain to each observation, such as each person or each completed questionnaire in a large survey. The columns pertain to each variable, such as a response or an observed characteristic for each person.

Columns: fields; variables; characteristics; responses; etc.

3. HOW ARE STATISTICAL TECHNIQUES IMPLEMENTED?

4. CLASSIFICATION OF TYPES OF DATA

4.1. Classification by Scales

4.1.1. Metric vs nonmetric

Metric: quantitative/numerical

Nonmetric: qualitative/categorical

4.1.2. Scaled ordinal data

Likert scales: Very low/low/medium/high/very high scaled as 1,2,3,4,5

Semantic differential: agree strongly/agree somewhat/ neutral/disagree somewhat/strongly disagree, scaled as 1,2,3,4,5

Summated rating scale: a variable defined by summing over several Likert or semantic-differential variables presumed to measure the same thing

4.2. Classification by Modes and Ways

The most common type of data is a table of rows and columns, the rows corresponding to individuals; the columns, to variables. This is two-way, two-mode data. "Two-way" refers to its being in a two-way array (matrix). The two "modes" are individuals and variables.

Three-mode data. Persons by variables by occasions-of-measurement (time) would be 3-way, 3-mode data.

Distance between planets by date would be three-way, two-mode data (modes = planets and dates; ways = planets x planets x dates), as would be paired comparisons done on say, three separate occasions.