Scaling is a technique used in measuring responses such as feelings, perception, likes, dislikes, interests, and preferences. Scales are used to measure objective responses and rank them in a given spectrum determinate of the type of scale used to gain the information.
The nominal level scale is a very simple scale consisting of an assignment of choices that tend to be mutually exclusive. In a nominal scale, the choices cannot be ranked because all the categories are different from each other. A good example of a nominal scale is gender, where males are put into Group 1 and females into Group 2. It makes no sense to rank male and female because neither is greater than the other. Neither is a better answer, and the numbers merely organize the data into numerical categories. These scales are the least restrictive of all scales and really represent a list of categories.
Ordinal scales are the simplest of attitude-measuring scales used in marketing research. While a nominal scale may contain numbers arbitrarily, each number in an ordinal scale represents a rank of order. In an ordinal scale, products or objects are rated based on their importance within a given category. For instance, an ordinal scale of beers might ask you to rank your preference from 1 to 5, where 1 is the kind you like best and 5 is the one you like least. Such a scale makes no attempt to rank a favorite in any one given product, but rather rates it on a spectrum against competing products.
Interval scales are also known as ranking scales, because, unlike the ordinal scale, you are asked to rank each object or product on its own scale. An example of an interval scale is if you were asked to rank how well you had enjoyed a particular movie on a scale of 1 to 5, where 1 is not at all and 5 is very much.
A ratio scale is similar to an interval scale, except the answers to these questions have a simple unambiguous starting point, typically zero. Ratio scales are not commonly used in marketing research but are used to describe a physical scale. Ratio scales often measure things like money, miles, height and weight where the answers describe how far the respondent is from zero. A ratio scale might ask you to fill in your annual income or square footage of your house, where instead of choosing an arbitrary measurement, you are filling in a blank. It becomes a scale when the data is all compiled and your answers are put on a spectrum with other respondents.