Your cart is currently empty!
M09.02.10 Scales
In statistics, scales of measurement are crucial for converting qualitative and quantitative observations into numerical data. This data can then be analyzed effectively to draw insights. The four primary scales of measurement are Nominal, Ordinal, Interval, and Ratio. Each scale provides a different level of information and precision, with unique characteristics and applications.
Types of Measurement Scales
| Type of Scale | Description | Key Characteristics | Examples |
|---|---|---|---|
| Nominal | Categorizes without order | “This or that” classification | Gender, intervention comparisons |
| Ordinal | Ordered groups | Rank or sequence | Olympic medals, class rank in med school |
| Interval | Equal intervals, no true zero | Mean and standard deviation | Height, blood pressure, drug dosage |
| Ratio | Interval + true zero point | True zero value | Temperature in Kelvin |
Overview of Each Scale
The four scales arranged hierarchically from least to most informative, provide unique types of data interpretation. Notably, higher scales can be “downgraded” to lower scales as needed for analysis.
1. Nominal Scale
- Purpose: Classifies observations into distinct categories without specifying order.
- Characteristics:
- No inherent ranking or order.
- Simply identifies “types” or “groups” without further detail.
- Examples:
- Gender (male or female)
- Treatment group (drug vs. control)
2. Ordinal Scale
- Purpose: Sorts observations into ordered groups but does not quantify the difference between them.
- Characteristics:
- Establishes a sequence or hierarchy.
- Does not indicate the degree of difference between categories.
- Examples:
- Ranking individuals by height, e.g., “Ben is taller than Fred.”
- Class ranks in medical school, Olympic medals (gold, silver, bronze).
3. Interval Scale
- Purpose: Measures the exact difference between observations using equal intervals.
- Characteristics:
- Equal increments across the scale.
- Does not include a true zero point.
- Often includes a mean and standard deviation for data.
- Examples:
- Blood pressure measurements
- Scale of length, where each unit is equivalent, e.g., inches
4. Ratio Scale
- Purpose: Provides both an ordered scale and a true zero, allowing for meaningful ratio comparisons.
- Characteristics:
- True zero point (absolute absence of the variable).
- Zero serves as a natural lower boundary; negative values are impossible.
- Examples:
- Temperature measured in Kelvin (where zero represents absolute zero)
- Height, weight, and dosage, as these can be measured in exact quantities from zero.
Key Points to Remember
- Nominal and interval scales are frequently tested and essential for foundational understanding.
- Each higher level of measurement includes the properties of the previous levels but adds new, distinctive qualities.
- Interval data can be treated as ordinal if needed, offering flexibility in data interpretation.