Nominal scales of measurement are used for labeling variables that do not have quantitative values. In other words, nominal scales are ‘labels’ for mutually exclusive variables that do not have numerical significance (Whitley, 1992). The following are some of the examples of nominal scales:
What is your gender?
o Male
o Female
What is the color of your hair?
o Brown
o Gray
o Black
o Other
Ordinal Scale
In an ordinal scale, it is the order of variables/values that matter, but the differences between the variables/values are unknown. In other words, it is challenging to quantify the difference between the variables. Therefore, ordinal scales are measures of various non-numeric concepts, such as happiness, satisfaction, and discomfort among others (Dahl " Dodgson, 1986). The following is an example of an ordinal scale:
How satisfied are you with our mode of service?
o Very unsatisfied
o Fairly unsatisfied
o Neutral
o Fairly satisfied
o Very satisfied
Interval Scale
Interval scales refer to numeric scales in which both the order and the exact differences between the values are known. Examples of interval scales include time and the Celsius temperature scale. The differences between the values are known, measurable, and consistent. For instance, the difference between 30 and 40 degrees Celsius is 10 degrees Celsius, which is similar to the difference between 60 and 70 degrees Celsius. However, interval scales have no ‘true zero,’ which makes it impossible to compute ratios (Whitley, 1992; Dahl " Dodgson, 1986).
Ratio Scale
Ratio scales form the ultimate solution when it comes to measurement scales since they show both the order of values/units and the exact value between units, and they also have true/absolute zero, which allows for the application of a broad range of both inferential and descriptive statistics. Examples of ratio variables include weight and height (Rossi " Crenna, 2013).
Difference between Positive and Negative and Positive Correlation
A negative correlation is where variables move in opposite or inverse directions. In other words, one variable decreases as the other increases. On the other hand, a positive correlation in where the variables move in a similar direction. In other words, one variable increases as the other increases and vise vasa (Hess " Hess, 2016).
How Regression can be used to assess the Relationship between Multiple Variables
Regression can be used in examining and predicting the relationship between dependent and independent variables through the use of a mathematical formula or equation (y = a + bx + e), where y represents the dependent variable and x represents the independent variable (Hess " Hess, 2016). For example, if a company needs to know the relationship between its sales volume (dependent variable) and its advertising expenditures (independent variable), the company can collect data for a given period and determine if there exists a significant relationship or connection between the two variables using the regression equation.
References
Dahl, H., " Dodgson, J. (1986). On Scales of Measurement. Teaching Statistics, 8(3), 92-93. doi: 10.1111/j.1467-9639.1986.tb00642.x
Hess, A., " Hess, J. (2016). Linear regression and correlation. Transfusion, 57(1), 9-11. doi: 10.1111/trf.13928
Rossi, G., " Crenna, F. (2013). On ratio scales. Measurement, 46(8), 2913-2920. doi: 10.1016/j.measurement.2013.04.042
Whitley, B. (1992). Units of Analysis, Measurement Scales, and Statistics: A Comment on Kerwin and Shaffer. Personality and Social Psychology Bulletin, 18(6), 680-684. doi: 10.1177/0146167292186003
