Measurement

Measurement refers to the process of assigning meaningful numbers (labels) to

individual persons based on the degree to which they possess a particular
characteristic. In biostatistics, the levels of measurement are very important since
different statistical methods commonly used are depending on the levels of
measurement. The four levels of measurement, namely nominal, ordinal, interval, and
ratio. Both nominal and ordinal levels of measurement use nonparametric statistics
while interval and ratio levels of measurement use parametric statistics.

1. Nominal scale consists of a finite set of possible values or categories that have
unordered scales. Examples include cause of death (cancer, heart attack, accident,
etc.) gender (male, female), blood type, nationality, occupation, civil status, etc. In this
scale of measurement, there is no natural order of categories.
For analysis in the computer, a nominal scale might be stored using numbers
rather than text. For example, cause of death. (1-cancer, 2-heart attack, 3-accident,
etc.). The number assigned to the observation serves only as a “name” for the category
to which the observation belongs. The numbers have no meaning. They are only labels.
Note that on a nominal scale’ the order of listing of the categories is irrelevant to the
statistical analysis. This scale of measurement is the lowest or least precise scale.

2. Ordinal scale consists of a finite set of possible values or categories that have
ordered scales. Examples include pain level (none, mild, moderate, etc.), social status,
attitude toward abortion, cancer stages, socio-economic status, etc. Unlike a nominal
scale, an ordinal scale is a natural ordering of the categories.
For analysis in the computer, like a nominal scale, the ordinal scale might be
stored using numbers rather than text, for example, pain level (1-none, 2-mild, 3-
moderate, etc.). The ordinal scale ranks the categories clearly, but the absolute
distance between categories is unknown. The numbers have limited meaning. The real
differences between adjacent ranks may not be equal.

3. Interval scale is generally measured on a continuum and differences between any

two numbers on the scale that are of known size. The name derives from the fact that it
is based on an interval that is accepted as a common standard and that yields identical
results in repeated applications. Other examples are tons of garbage, the number of
arrests, income, age, etc. An important property of the interval scale is that there are no
true zero points. That is, the value “0” is arbitrary and does not reflect the absence of
the attribute.

4. Ratio scale like the interval scale is also measured on a meaningful continuum. The
distinction is that the ratio scale has a meaningful zero point. Weight in pounds is a
good example of a ratio scale. Other examples are height and age. The ratio scale is
used when not only the order and interval size are important, but also the ratio between
two measurements is meaningful. This scale of measurement is the highest or most
precise scale.

Nominal Ordinal Interval Ratio

Zip Code Grade (A, B, C, D) SAT Score Height
Sex (Male, Female) Judging (1st, 2nd, 3rd) IQ Weight in pounds
Eye Color (Blue, Rating scale (Poor, Temperature Time
Brown, Green, Hazel) Good, Excellent)
Political Affiliation Ranking of Players in Tons of Salary
such events garbage
Religious Affiliation Pain Level (none, mild, Number of Age
moderate, severe) arrests
Major Field Social Status Income
(Teaching)
Nationality Attitude toward abortion
Cause of Death Cancer stages
Blood Type
Occupation
Civil Status

Variables
Variables refer to the characteristic of a person or objects which can take on
different values or labels for different persons or objects under consideration. For
example, the undergraduate major is a variable that can take on values such as
epidemiology, biostatistics, mathematics, sciences, and the like. Another example of
variables includes smoking habit, attitude toward the boss, height, faculty ranks, etc.

2 Types of Variables

1. Response variable is a variable that is affected by the value of some other variable.
This may be continuous, ordinal, or nominal. In a regression setting, they are called
dependent variables or Y variables.

2. Explanatory variable is a variable that is thought to affect the values of the response
variable. It is sometimes called the independent variable or X variable in a regression
setting.
In this case, the explanatory variable, like the response variable, may be continuous,
ordinal, or nominal.

2 Classifications of variable

1. Qualitative Variables are ones whose categories are simply used as labels to
distinguish one group from another, rather than as the basis for saying that one group is
greater or less, higher or lower, or better or worse than another. His variable has
intrinsically non-numeric values. Qualitative variables generally have either nominal or
ordinal scales. Examples, Cause of death, Nationality, Race, Gender, Severity of pain,
etc.
Qualitative variables can be with reassigned numeric values but they are still
intrinsically qualitative. For instance, SEX (1-Male, 2-Female). Some variables are
always qualitative in nature. For example Occupation, Sex, Diseases Status, Cause of
Death, Race, etc.
2. Quantitative Variables is one whose categories can be measured and ordered
according to quantity. These variables are intrinsically numeric values. The number of
children in a family and age are good examples of quantitative variables. Both interval
and ratio scales belong to this classification.

Two groups of Quantitative Variables

1. Discreet variables refer to each element of a set of possible values that is either
finite or countably infinite that can appear only as whole numbers. Examples are
numbers of missing teeth, number of household members, and the number of patients
at hospital X, etc.
- assume values that can be counted.
- can be assigned values such as 0, 1, 2, 3 and are said to be countable.

2. Continuous variables refer to each element of a set of possible values including all
values in an interval of the real line that can be expressed with fractions or digits after a
decimal point. Body mass index, blood pressure, cholesterol levels, and height are just
some examples of continuous variables.

Since CONTINUOUS DATA must be measured, answers must be rounded because of

the limits of the measuring device. Usually, answers are rounded to the nearest given
unit.
i.e. height in nearest inch
weight in ounce

Data

Qualitative Quantitative

Discreet Continuous

Summation Notation (“∑”)

- greek capital letter Sigma and is used to denote sum.
- define the limits of summation.
i =1 means that we start with the first value of X
n = means that we end with the nth value of x1 to xn

RULES CONCERNING SUMMATIONS:

Example 1: Let X1 = 3, X2 = 5, X3 = 2 and X4 = 6
Evaluate:

Solution:

= 16

1. The sum of a constant from 1 to n equals the product of the constant and n. In
symbol,

Example 2: Evaluate

Solution:

= 36

2. The sum of a variable and a constant equals the sum of the variable plus the product
of the constant and n. In notation,

Example 3: Let X1 = 3, X2 = 5, X3 = 2, X4 = 6 and X5 = 3

Evaluate

Solution:

= 3 + 5 + 2 + 6 + 3 + 5(7)
= 19 + 35
= 54
3. The difference between a variable and a constant equals the difference of the
variable minus the product of the constant and n. In notation,

Example 4: If X1 = 3, X2 = 5, X3 = 2, X4 = 6 and X5 = 3
Evaluate
 Solution:

= 3 + 5 + 2 + 6 + 3 - 5(3)
= 19 - 15
=4
4. The sum of the products of a variable and a constant equals the product of the
constant and the sum of the variable. In notation

Example 5: Let X1 = 3, X2 = 5, X3 = 2, X4 = 6 and X5 = 3

Evaluate

= 7)
= 7(3 + 5 + 2 + 6 + 3)
= 7(19)
= 133