ABSTRACT
Statistical problem solving provides a set of powerful tools which can be used to maximize the efficiency and productivity of empirical problem solving. Research in biotechnology generates great quantities of data. The recent spread of computers in the research environment has greatly increased our ability to collect and manage this data. Data is collected to solve empirical problems. Data serve as a basis both for understanding and for action. Every experiment has a design. Some designs are better than others. Since data must be collected anyway, the use of statistically designed experiments for data collection adds only incrementally to its cost. An experimental design is a collection of predetermined settings of the process variables. A response variable is a measure of process performance. Empirical problem solving is difficult because the problems are complex and progress along the learning curve is often slow and difficult.
