Biostatistics

is the science of collecting, processing, analyzing and interpreting data in such a way that conclusions can be drawn in an objective manner.

pertains to statistics as applied to the biological and health sciences

number of students receiveing an A

State & Federal statistics (pregnancy rate among women under the age of 16, number of cases of AIDS reported yearly, unemployment rate)

Proportion of study subjects receiving a new experimental drug who are cured of their disease

Determining the factors that increase the risk of an individual developing coronary heart disease

Establishing whether a laparoscopic surgical procedure results in a shorter recovery period than a more invasive, traditional alternative

compare the effectiveness of two drugs that are often prescribed to patients who suffer from arthritis

asessing the reduction of child mortality in MVA resulting from the use of seat belts.

the science of characterizing the behavior of random phenomena

Parameter

the complete set of objects of interest

subset of a population

Nominal or ordinal

Discrete or continuous

values are unordered categories or classes for which there is not a natural ordering

ex: political affiliation, gender, marital status

There is a natural ordering among the category values

ex: injury severity scale, activity level, severity of disease

data values are counts. Both order and magnitude are important. They tell how many were observed

ex: # of live births, # of deaths attributed to a given cause, # of relapses of a disease observed in a given time frame

whole #

Data that can take on any decimal value within a given range

ex: height or weight, distance, time, temperature

are a common method of displaying data, especially for categorical or discrete data

number of units in a given category

proportion of the total count

percentage of the total count (total number of observations) that have a value less than or equal to the upper limit of the interval

• Graphical
• Arithmetic

very common and are used to show the frequency/relative frequency observed in differnet (nominal or ordinal) categories

is a graphical representation of a frequency or relative frequency distribution

a graphical representation of the quartiles of a data set. They also indicate unusual data values. Atypical values in a data set are outliers.

the left side is a mirror image of the right side

if the majority of the area of the box plot is on the right, and the area gradually trails off on the left.

If the majority of the area of the box is on the left and the area gradually trails off to the right.