# Tutorial on Introduction to biostatistics

**Variable classification according to their nature**

**a.
****Independent variables**

Variables that can either be manipulated by the researcher
or which are not outcomes of the study but still affect its results are called *independent
variables*. A good example is age of the subjects, which can be an independent
variable that may affect a study outcome variable such as subject survival.

**b.
****Dependent variables**

The outcome variables defined as part of the research
process are termed *dependent variables* and these will be affected by the
independent variables under study. An example of a dependant variable can be
the number of people who developed a particular disease in a cohort study.

**Variable classification according to their type**

**a.
****Quantitative variables**

Variables
that can be measured numerically are called *quantitative variables*.
These variables can be further classified as *continuous* and *discrete*
variables. A continuous variable could take any value in an interval. Examples
of continuous data are findings for measurements like body mass, height, blood
pressure or serum cholesterol.

Discrete variables will have whole integer values. Examples are the number of hospitalizations per patient in a year or the number of hypoglycemic events recorded in a diabetic patient over 6 months.

**b.
****Qualitative variables**

Variables, which cannot be measured numerically, are called qualitative variables. An example is gender.

**Variable classification according to the scale of
measurement**

**a.
****Nominal Scale variables**

**b.
****Ordinal Scale Variables**

**c. Interval Scale Variables**

In an interval measurement scale,
one unit on the scale represents the same magnitude of the characteristic being
measured across the whole range of the scale, i.e. the intervals between the
numbers are equal. However, the *ratio* between set of two numbers in the
scale are not equal because an interval scale lacks a true zero point.

** **Temperature in Fahrenheit would
be a perfect example for interval scales** **because though we can add and subtract degrees (70° is 10° warmer than
60°), we cannot multiply values or create ratios (70° is not twice as warm as
35°).

**d.**

**Ratio Scale Variables**

Ratio scale variables will have all the properties of interval variables with the ratio between two numbers in the ratio scales being identical. Ratio scales have an absolute or zero point. For example, a 100-year old person is indeed twice as old as a 50-year old one

# Tutorial on Introduction to biostatistics