Measurement scales
#inferentialstatistics #researchlandscape
Scales for measurement matter in data because humans make assumptions
We also need to conceptualize and operationalize our measurement scales, which means to define what we are measuring and how we will measure it. Conceptualization involves defining the concept or idea that we want to measure, while operationalization involves defining the methods or procedures that we will use to collect and measure data related to that concept. Essentially, it transforms abstract concepts into observable and measurable variables.
-
Nominal Scale: This is the simplest type of scale that assigns names or labels to distinguish different groups, categories or objects. It is used for categorical data and does not imply any order or numerical value. For example, classifying people by their nationality or by the color of their hair.
-
Categorical Scale: This is similar to the nominal scale as it divides data into distinct categories where no order or priority is implied. The only difference between them is that categorical variables can be ordinal as well. For example, survey responses such as “yes”, “no” and “maybe”.
-
Ordinal Scale: Unlike nominal scale, ordinal scale not only categorizes the variables but also orders them in a certain way. It represents relative order of items, but the intervals between values may not be equal. For example, rating satisfaction on a scale from 1 (very dissatisfied) to 5 (very satisfied).
-
Interval Scale: This scale has all the features of ordinal scale plus equal intervals between adjacent values. However, it does not have a true zero point (a point at which none of the quantity is present). Temperature scales like Celsius and Fahrenheit are examples.
-
Ratio Scale: This is the highest level of measurement which has all the properties of an interval scale along with a definitive zero point (no negative numbers). It allows comparisons about how many times greater one score is than another. Examples include age, height, weight and income.