1 - 2 Biostatistics
1 - 2 Biostatistics
• describing;
• organizing;
• analyzing, data
• interpreting
Statistics
• Descriptive statistics
• Inferential statistics
4
• Descriptive statistics are used to organize and describe the
characteristics of a collection of data. The collection is sometimes
called a data set or just data.
• Inferential statistics are often (but not always) carried out after
descriptive statistics. They are used to make inferences from a smaller
group of data to a larger one.
• An example is using results from one kindergarten classroom to infer,
or generalize, about a population of a whole kindergarten grade.
Descriptive statistics
6
Descriptive statistics
• Descriptive statistics organizes and summarizes observations, allowing
an overview of the general characteristics of a series of data.
Qualitative Quantitative
12
Data are measured on 4 different
scales
• They are: nominal, ordinal, interval, and ratio.
• When data are measured using the nominal scale, numbers merely represent
which category an observation belongs to.
• When data are measured on the ordinal scale, the numbers represent an
ordering or ranking of the data.
• Nominal and ordinal data are inherently categorical (individuals fall within
categories).
• When data are measured on the interval scale, the numbers represent equal
units from one value to another on the scale.
E. Petrela
Data (cont)
• And when data are measured on the ratio scale, the data have a
meaningful zero value representing the absence of a quantity being
measured.
• Interval and ratio data is quantitative.
• Qualitative data is divided into data with discrete or a finite or
countable number of outcomes or continuous/infinite outcomes.
E. Petrela
Quantitative data
• they are used to collect the bare facts, the numbers.
• These are statistical and structured data, which are used to support
the drawing of general conclusions from the research.
• healthy - sick,
• alive - dead,
• therapy - placebo.
Ordinal variables
• When the order between categories is important, the data is called
ordinal data.
• Many types of medical information can be characterized by more
than two values and have a clear direction (e.g., best to worst), but
these data are not measured by a rigorous measurement scale.
• Examples:
• stages of a disease: 1,2,3,4
• NYHA classification: I, II, III, and IV
• nutrition level: underweight, normal, overweight, obese
Discrete variables
• For discrete data, both order and magnitude are important.
• Numbers represent actual measurable quantities rather than mere
symbols.
• They can take on specific values – often whole numbers – NOT
intermediate values. They take on a finite or countable amount of
values.