1.6
Interval Level of Measurement
For effective statistical analysis, data are classified into four levels of measurement—nominal, ordinal, interval, and ratio.
Data measured using the…
Data can be divided into four levels of measurement.
The interval level of measurement includes data that can be ordered, and the difference between the values is meaningful.
For example, the temperature in a desert is measured in degrees Celsius and can be ordered from low to high.
The difference between the highest and the lowest values indicates how much the temperature changes in a day.
But, a nighttime temperature of zero degrees Celsius does not mean there is no temperature; on the contrary, it is quite cold. So, the interval level of measurement does not have a natural zero point.
On the contrary, consider a catalog of computer prices. Zero dollars means no cost. So this data set has a natural zero point and cannot be considered for interval level of measurement.
View the full transcript and gain access to JoVE Core videos
Q1: What makes interval level of measurement different from ordinal level?
Interval level data can be ordered like ordinal level data, but the key difference is that the differences between values are meaningful and measurable. For example, the difference between 30°C and 20°C equals 10 degrees, and this difference is the same as between 40°C and 30°C. Ordinal data only shows ranking without meaningful numerical differences between values.
Q2: Why does interval level data not have a natural zero point?
Interval level data lacks a natural zero point because zero does not represent the absence of the measured property. For temperature, zero degrees Celsius does not mean there is no temperature; it is simply cold. Negative temperatures like −10°F exist and are colder than zero, demonstrating that zero is arbitrary rather than a true starting point.
Q3: Can you compare ratios of values in interval level measurement?
No, ratios have no meaning in interval level measurement. For example, 80°C is not four times as hot as 20°C, even though 80 is mathematically four times 20. This is because the arbitrary zero point prevents meaningful ratio comparisons. Interval data can be used in calculations like addition and subtraction, but not for ratio-based comparisons.
Q4: What is an example of interval level data in real-world applications?
Temperature measurement is the primary example of interval level data. Temperatures measured in degrees Celsius or Fahrenheit can be ordered from low to high, differences between values are meaningful, but zero degrees does not indicate the absence of temperature. Other examples include calendar years and standardized test scores.
Q5: How does interval level measurement differ from ratio level measurement?
Interval and ratio levels both have meaningful differences between values, but ratio level data has a natural zero point representing the absence of the property. For example, computer prices use ratio level measurement because zero dollars means no cost. Temperature uses interval level because zero degrees does not mean no temperature, making ratio level data suitable for meaningful ratio comparisons.
Q6: What calculations are appropriate for interval level data?
Interval level data can be used in addition and subtraction calculations. For instance, you can calculate that the temperature difference between 40°C and 20°C is 20 degrees, or find the average temperature across multiple days. However, multiplication and division for ratio comparisons are not meaningful because of the arbitrary zero point.
Q7: Where does interval level fit among the four levels of measurement?
Interval level is the third of four measurement levels: nominal, ordinal, interval, and ratio. It represents data that is ordered with meaningful differences between values, but lacks a natural zero point. This positions it between ordinal level, which only ranks data, and ratio level, which includes both meaningful differences and a true zero point.