Data Measurement Scales
These scales describe how data can be categorized, ranked, or measured with varying degrees of precision. Below are four common data types—Nominal, Ordinal, Interval, and Ratio—along with examples and references.
1. Nominal
Definition
- Data can only be categorized.
- No intrinsic ordering exists between different categories.
Examples
- City of birth: (Delhi, Ranchi, Kolkata, Mysore)
- Gender: (Male, Female)
- Car brands: (Hyundai, Toyota, Mercedes)
- Marital status: (Unmarried, Married, Divorced)
2. Ordinal
Definition
- Data can be categorized and ranked.
- The exact difference between rankings is not necessarily equal.
Examples
- Language ability: (Beginner < Intermediate < Fluent)
- Spiciness levels: (Mild < Medium < Hot < Extra Hot)
- Medal placements: (Gold > Silver > Bronze)
3. Interval
Definition
- Data can be categorized, ranked, and evenly spaced.
- No true zero point (0 does not mean an absence of the quantity).
Examples
- Temperature (Fahrenheit/Celsius): The difference between 10°C and 20°C is the same as between 20°C and 30°C, but 0°C does not mean “no temperature.”
- Calendar years: The interval of five years is always the same length, but year 0 doesn’t represent the absence of time.
- Time of day: From 1 p.m. to 2 p.m. is the same interval as from 2 p.m. to 3 p.m., but 0:00 hours is just a starting reference, not “no time.”
4. Ratio
Definition
- Data can be categorized, ranked, evenly spaced, and has a natural zero point.
- Zero means the complete absence of the quantity.
Examples
- Height: If you have 0 cm of cloth, that is no cloth.
- Age
- Weight: 10 kg is half of 20 kg.
- Temperature in Kelvin: 0 K is absolute zero, meaning no thermal energy.
Q1
How could you differentiate interval and ratio data in practical survey scenarios?
Q2
Are Likert scales and other similiar 5 point rating systems Ordinal or Interval scales?
Q3
How do interval and ratio scales affect the choice visualizations?