Most people have been asked to participate in a survey or questionnaire at some point in their lives. The question was probably in the form of a scale that the research team setup in the questionnaire. Scales help researchers by giving respondents a way of conveying their thoughts and feelings in a subjective and measurable way. The survey or questionnaire is a tool for the researcher to capture the respondent’s thoughts and feelings. Each scale has different characteristics, depending on the answers to be measured. The nominal, ordinal, interval, and ratio are four scales typically used in questionnaires.
If a questionnaire consists of a question with a “this or that” type of answer it is probably based a nominal scale. A nominal scale is used to assign objects that are, "mutually exclusive, (and) labeled categories" without creating "necessary relationships among the categories" (Aaker, Kumar, & Day, 2007, p. 288). An example of a nominal scale can be as easy as a yes/no or male/female style of answer. Researchers could use this scale as a simple count or create percentages such as 55% of the respondents are male.
Mode is used to find the central tendency of a nominal scale because the researcher would want to know the answer that had the highest frequency (Aaker, Kumar, & Day, 2007). The mode answers if people answer yes or no more in the questionnaire. In a questionnaire for a refrigeration company the researcher would want to know if the person answering the question was an installer. If the respondent answered yes it would skew how other answers would be analyzed for the research. The answer would also tell the researcher what percentage of the respondents were installers and if the questionnaire was selecting the sample target market of installers.
An ordinal scale ranks objects or is obtained, “by arranging them in order with regard to some common variable” (Aaker, Kumar, & Day, 2007, p. 288). An ordinal scale is measured by percentage, mode, rank order correlation, and median. A rank-order scale is an ordinal scale that requires "the respondent to arrange a set of objects with regards to a common criterion" (Aaker, Kumar, & Day, 2007, p. 292). This is useful to researchers to gauge how a respondent’s ranks the importance of some subject.
In the example of a questionnaire for a refrigeration company, the respondent would be asked to rank their preference, one ranked as the most preferred to five as the least preferred, from a list of the top five refrigeration competitors. The researcher would be able to show the client where their company stands among the top five competitors. A comparative scale would give respondent’s attitude toward comparisons to the preference of quality with specific competitors (Aaker, Kumar, & Day, 2007).
As defined in Marketing Research, "In an interval scale the numbers used to rank the objects also...