Transcript Categorical variable

Different types
of
variables
Issues to cover – in a flash
1. Quantitative variable (numeric variables)
–
–
1.1 Discrete variable
1.2 Continuous variable
2. Qualitative variable (Categorical variable)
–
–
–
2.1 Nominal variable
2.2 Binary variable
2.3 Ordinal variable
TYPES OF VARIABLES
• QUALITATIVE VERSUS QUANTITATIVE
• CONTINUOUS VERSUS DISCRETE
• NOMINAL, ORDINAL, INTERVAL, RATIO
• DEPENDENT VERSUS INDEPENDENT
• EXPOSURE VERSUS OUTCOME
• HIGHLY CONFUSING!!!
VARIABLES
QUALITATIVE
QUANTITATIVE
CONTINUOUS
DISCRETE
1. Quantitative variable
1.1 Discrete variable: A variable which can only
take certain values, usually whole numbers (a
count)
– Examples number of visits to a doctor, number of
asthma episodes in a year, number of children in a
household etc.
1.2 Continuous variable: a variable which can
theoretically take any value within a given range
– Examples: height, body mass index etc.
2. Qualitative variable
(Categorical variable)
• A variable whose values are represented
by names or labels of the same
characteristic. Examples are ethnicity,
blood group, eye colour, nationality.
• All qualitative variables are discrete
2.1 Nominal variable
(Categorical variable)
• Categorical variables where the categories
are not ordered in increasing or decreasing
quality or intensity.
E.g. Hair colour
Other examples?
2.2 Binary variable
• A categorical variable which takes only
two categories or possible values.
– Examples: Yes/No, Dead/Alive, Positive/Negative
Other examples?
2.3. Ordinal variable
• A categorical variable in which the different
levels of the variable are ordered. Successive
levels represent “increases” or “decreases” in the
characteristic conveyed by the variable.
Examples:
– Severity of pain: no pain, minimal pain, moderate pain,
severe pain
– Grades of malnutrition categorized by BMI (Mild, moderate
and severe)
Ordinal variables
• It is important to analyse this type of data
with special methods
• E.g. chi square for trend, Kruskal-Wallis test
(non-parametric significance test) and not with
methods used for nominal variables