Evidence-Based Practice for Nurses

Levels of Measurement

Measurement is the process of assigning numbers using a set of rules. Four
categories are used to describe measurements: nominal, ordinal, interval, and
ratio. These categories are more commonly known as levels of measurement.
When researchers collect data, they must know which level of measurement is
used to select appropriate statistical tests for analyzing data (Trochim, 2000).

Nurses must be able to identify levels of measurement correctly to appraise
evidence. Nominal measurement is the weakest level of measurement. The word
nominal is derived from the word name. Researchers use nominal measure-
ment to classify or categorize variables, also referred to as categorical data.
The numbers assigned to each category are just labels and do not indicate any
value. For example, responses of “yes” and “no” on a survey are often assigned
numbers 1 and 2. The value of these numbers has no meaning because one
cannot claim that a “no” response is higher than a “yes” response. The arbitrary
numbers are assigned for the purpose of coding, recording, and entering data
for collection and analysis. Questionnaires often utilize nominal measures to
record a variety of categorical data and/or closed-ended questions with fixed
responses such as gender, race, and diagnosis (see Figure 10-4). Dichotomous

Gender: Male = 1

Female = 2

Race:

Asian = 2

Caucasian = 3

Hispanic = 4

Native American = 5

Other = 6

Type of dementia: Alzheimer’s disease = 1

Lewy body dementia = 2

Vascular dementia = 3

Frontotemporal

dementia = 4

Unknown = 5

African American = 1

FIGURE 10-4

Examples of Nominal Level

Measurement

KEY TERMS

ratio: The

highest level of

measurement that

involves numeric

values that begin

with an absolute

zero and have

equal intervals;

in epidemiology

a mathematical

relationship

between two

numbers

levels of

measurement: A

system of classifying

measurements

according to

a hierarchy of

measurement

and the type of

statistical tests

that is appropriate;

levels are nominal,

ordinal, interval,

and ratio

categorical data:

Lowest level of

measurement

whereby data

are categorized

simply into groups;

nominal data

dichotomous:

Nominal

measurement when

only two possible

fixed responses

exist such as yes

or no

10.2 Collecting Quantitative Data 261