#
What is Statistical Science ?

The process of posing
questions and then seeking to answer them by collecting and
analyzing suitable data is an essential component of
research in a remarkably diverse array of fields. This includes
agriculture, medical research, industrial research, forensic science,
market research, environmental science, political science and quality
assurance. This mode of inquiry also features prominently in the
decision making processes of both commerce and government.
The role of the Statistician is to determine, for a
given question,
**the type of data that is needed, the way it should be collected
and how it should be analyzed in order to best answer that question.**

The data may result from a planned experiment
designed to investigate
certain specific things. This sort of data, **experimental data**,
is
common in such areas as agricultural research. The concern of the
Statistician is not just the analysis of the data from someone
else's experiment, but is also about **designing the experiment**
in the
first place, to ensure that resources are used efficiently and that the
questions being asked can be answered by the experiment.

Other types of data arise from observational
studies: investigators go out
and see what is actually there. Censuses of the population, hospital
data
bases, Gallup polls, traffic data and consumer data bases
are examples of these. Survey and questionnaire design
are important issues in many of these examples.

- to understand the nature of statistical
inference; that is, its scope and
limitations and its proper role in the process of scientific
investigation,
- to be able to express a generally posed
scientific question as a
statistical question,
- to be familiar with a variety of commonly used
techniques and the models
underlying them,
- to be able to recognize the nature of, and to
model, the random
variation underlying given data,
- to be conversant with the mathematical
underpinning of often-used
statistical techniques to the extent of being able to make simple
modifications in appropriate situations,
- in preparing to make inferences about a
population based on a sample from
it, to be able to decide how to obtain a suitable sample,
- to be able to provide advice on the design of
experiments,
- to be able to use standard statistical packages
to perform
statistical calculations,
- to be able to interpret the the results
obtained from standard
statistical packages,
- to be equipped with a variety of graphical
techniques for displaying
statistical data,
- to be able to know where to find recently
developed statistical
methods.

Statisticians are employed as biometricians
(statistical scientists
specializing in biology related applications) in government
agricultural
departments, as biostatisticians (statistical scientists specilizing
in medical related applications) in government health related
departments
(e.g., NIH, CDC,...) and as consultants in a number of government,
quasi-government
and private research firms. Many manufacturing companies also employ
statisticians to perform quality control and process control of their
products. The research and development units of some larger
corporations,
larger hospitals and health departments employ statistical
consultants. The National Bureau of Standards employs
many statisticians in the
area of civil statistics, and various government departments in finance
and
industrial relations areas have statisticians working with economic
data.
There are opportunities for statisticians to work in quality
improvement initiatives within industrial and commercial organisations.

Wei-Min Huang

Last update 01-Jan-96