Chapter 2: Levels of Measurement & Their Importance

Levels of Measurement

refers to the way numbers are assigned to represent
variables (how much information is being provided by the outcome measure)

Variable

anything that can take on more than one value

Measurement

the assignment of labels to a variable or outcome

Types of Variables

• Nominal

• Ordinal

• Interval

• Ratio

Nominal Variables

It
Categorizes data without any order or ranking. The categories are mutually exclusive and cannot be arranged in a meaningful sequence

• qualitative and categorical in nature

• measured only by the nature of the group to which they belong

• Nominal-level variables are always qualitative

Examples:

• Ethnicity

• Eye-color

• Political affiliation

• Marital status

• Religious Affiliation

Ordinal Variables

Allows for data categorization in a meaningful order or ranking of the categories.

• Variables can imply severity, rank, and lesser to greater

• The limitation of ordinal variables is that we can know where an outcome ranks in
  comparison to others, but we do not know the true nature of that outcome

Examples:

• ES (low, middle, high)

• Income Level (less than 50k, 50-110k, 110+k)

• Letter Grades (not GPA- that’s interval)

• Level of Education

Interval Variables

Data that not only provides order but also equal intervals between values (without a true zero point)

• Interval scores tell us that each score is different from others in some way

• Interval level gives us the most information for most variables

Examples:

• Temperature (celsius & Fahrenheit)

• pH

• Credit scores

• SAT scores

• Dates (on the Roman Calendar)

Ratio Intervals

Variables are on a quantitative scale but include an absolute zero indicating an absence of the measured variable.

• Ratio is the most informative variable but the least likely to be seen in social/behavioral sciences

Examples:

• Weight

• Rainfall

• Number of students in a classroom

• Speed of travel

• Temperature (Kelvin scale)

• Income

• Crime Rate