Data can be classified according to
levels of measurement. The level of measurement determines how data should be
summarized and presented. There are four levels of measurement, nominal,
ordinal, interval and ratio.
The nominal level describes a characteristic that has no order and can be classified or counted examples include categories or color. The next level is the ordinal level which can be ranked however you cannot conclude how much better one rank is from another examples include good better best. Moving up the next level is the interval level which includes all characteristics of nominal and ordinal level data but now the intervals and differences between the levels are more meaningful examples include temperature in Fahrenheit. However one notable characteristic of the interval level is the value of 0 does not denote absence of data, rather it is just the point on the scale examples for this are shoe size ring size or dress size where there is no natural zero. With interval level data while the measurements between the intervals remain consistent their ratios are not. The final level is the ratio level, differences and ratios in this level are meaningful. Furthermore the value zero is also meaningful if you have zero of something there is none of it. Examples of ratio data often consist of counts.
Nominal
|
Ordinal
|
Interval
|
Ratio
|
· Hair
color
·
Sex
·
Religious affiliation
·
Gender
·
Eye color
·
Place of birth
·
Race
·
Department
·
Car Manufacturer
·
Demographic or qualitative data
·
Color
·
Shape
|
· Grades/Performance
at school, A/B/C/D/F
Severity level: low, medium, high
Rating systems centered around inferior, poor, good,
excellent, superior
· "Bad",
"Okay", "Good", and "Great"
·
Ranks and Ratings
|
·
Shoe size
· Fahrenheit and Celsius temperatures
· Clothing
sizes – ie pants size
· Ring
Size
· Bra
Size
|
·
Number of students in a class
·
Age
·
Disposable income
·
Kelvin temperatures
·
Distance to work
·
Number of patients seen by a
doctor
· Amount
of Money
|
No comments:
Post a Comment