Levels of Measurement
Classification that describes the nature of the
information within the values assigned to variables
4 Levels of Measurement
Nominal, Ordinal, Interval & Ratio
Nominal
Categorial variable; values cannot be meaningfully ordered. Assign numbers to objects that are categorized
Nominal Characteristics
1. Categories must be mutually exclusive
2. All categories used to represent the data must be equivalent.
3. The list must be exhaustive.
Examples of Nominal Variables
1. Biological sex (male=1, female =2)
2. Political party affiliation ( 1=democrat,
2=republican, 3=other, 4=not registered)
Nominal Data
Descriptive Statistics & Frequency Distributions
Ordinal
Categorical; Numbers have meaning & are ranked with regard to more or less of a characteristic
Ordinal Characteristics
1. Do not have equal intervals
2. Do not have an absolute zero
3. Number tells the order
4. Distances between attributes have no meaning
Ordinal Examples
1. Socioeconomic Status (1=lower, 2=middle, 3=upper)
2. College Class (1= Freshman, 2= Sophomore, 3= Junior, 4=Senior)
Ordinal Data
Numerical values are assigned to each category with higher numbers representing higher values & the numbers are then related to the variable
Interval Data
Continuous; items are ordered and the size of the intervals between two items on the scale are equal. The distance between attributes has meaning
Interval Examples
1. Temperature measured in Fahrenheit
2. Likert Scales (assuming equal intervals)
Interval Data
Numerical values are assigned to each value and a measure of central tendency is calculated (mean, median, mode)
Ratio
Continuous; items are ordered in an interval scale except there IS a TRUE ZERO & intervals have to be equal
Ratio Examples
1. Number of times you drank soda this week, 0= no soda
2. Number of clients in past 5 months
Ratio Data
Numerical values that are related to the variable are assigned to each category. Higher numbers = "more" of the variable