Nominal, ordinal and scale in SPSS
Nominal, ordinal and scale is a way to label data for analysis. While nominal and ordinal are types of categorical labels, the scale is different. In SPSS, we can specify the level of measurement as:
- scale (numeric data on an interval or ratio scale)
Nominal and ordinal data can be either string alphanumeric or numeric. Each of these has been explained below in detail.
In the primary research, a questionnaire contains questions pertaining to different variables. Some of those variables cannot be ranked, and some can be ranked but cannot be quantified by any unit of measurement. While some can be ranked as well as can be quantified. Upon importing the data for any variable into the SPSS input file, it takes it as a scale variable by default since the data essentially contains numeric values. It is important to change it to either nominal or ordinal or keep it as scale depending on the variable the data represents.
Difference between nominal, ordinal, and scale in SPSS
In the SPSS input file, it is required to define the variables on the basis of nominal, ordinal or scale. At the same time, it needs to code the variables according to the categories those variables are divided into.
A variable can be treated as nominal when its values represent categories with no intrinsic ranking. For example the department of the company in which an employee works. Examples of nominal variables include region, zip code, or gender of individual or religious affiliation. The nominal scale can also be coded by the researcher in order to ease out the analysis process, for example; M=Female, F= Female.
A variable can be treated as ordinal when its values represent categories with some intrinsic ranking. For example, levels of service satisfaction from highly dissatisfied to highly satisfied. Examples of ordinal variables include a degree of satisfaction among the consumers, preference degree from very high to very low, and degree of concern towards a certain issue. Generally, it is preferable to assign numeric codes to represent the degree of something among respondents. For example 1=Highly satisfied, 2=satisfied, 3= neutral, 4= dissatisfied, 5= highly dissatisfied.
A variable can be treated as a scale when its values represent ordered categories with a meaningful metric, so that distance comparisons between values are appropriate. Examples of scale variables include age in years, income in thousands of Rupees, or the score of a student in the GRE exam. For example in a classroom of 60 students, each one would have been given GRE entrance test, and therefore Scale is used to determine the average score for the class, the highest and lowest score in the class so on and so forth.
How to choose between the three in a study
Generally, for a study that involves primary data collection, close-ended survey questionnaires are used. It is important to select the type of measurement properly while framing the questionnaire to avoid gaps in your study. The table below shows how to choose the correct one.
|Meaning||Categories have no meaningful order or rank but just record the perception of different things||Categories with meaningful order or rank to them||The data is not grouped based on any linkage but just has the numerical values|
|Feature||General perception recording for different things though are of the same field but completely unrelated to each other||Have a level of agreement or record satisfaction level||The data has numerical values with no associated order or rank with open response questions|
|Applicability||Just want to record perception for some specific things that have meaningful ranking||Things wherein no specific difference could be depicted but just an order represent the variation in perception||Differences in responses could be measured and each category defines the different level|
|Example||Marital status, political party, region, eye colour, or yes/no questions||Perception recorded via Likert scale (3-point, 5-point, or 7-point)||Height, weight, income, time|