Water quality monitoring conducted within the WEST program uses the "standard scientific method" protocols. The goal is to produce valid results for landowners and program participants that can be repeated over time and reported with a high level of confidence.

The natural resource scientists who use statistical methods to reach conclusions all approach their problems by the same general procedure, commonly known as "the scientific method." The method is broken down into several stages:

1. State the problem.
2. Formulate the hypothesis.
3. Design the experiment or survey.
4. Make observations (collect data).
5. Interpret the data.
6. Draw conclusions.

Problem Statement

Formulation of a statement of the problem in written form helps solidify the idea to be studied and allows a scientist to proceed with the next steps in a focused manner.

"Hypothesis"

The hypothesis is the expected outcome if the experimenters speculations are true. A good hypothesis is comprehensive enough to explain a phenomenon and predict unknown facts and yet is stated in a simple way. A literature search will indicate whether the problem has been researched, whether a body of studies exists that yielded similar results, and whether or not additional study will make a worthwhile contribution.

Design the Experiment or Survey

An experiment is designed to test the hypothesis through controlled experimentation. It consists of a set of similar objects that receive a specific treatment and then a response is measured. Design considers questions such as; what variables should be measured? How will the measurements be taken and with how much precision? What treatments or conditions should be placed on the subjects to test the hypothesis? What samples will be measured out of the population of interest? In general samples meet the following criteria: random, representative of the population, sufficiently large, controlled for extraneous variables.

Random sampling is intended to avoid bias in the selection of plots on the ground. Such plots may not always fall in a convenient area close to a road or trail. A representative sample of the population refers to the subject being sampled. If tall and short grass species are present in an. area, then the plots sampled should each have tall and short grass present. Enough samples then must be taken to reach an adequate sample size to account for the variability in the plant community. Variability on a site may be due to a soi1 change, plant community variation, moisture difference within a site, climatic changes from season to season or year to year, etc.

1. Sampling must be carried out in a rigorous manner .
2. The study should be free from errors except for the variations that are due to the limitations of the equipment being used.
3. Bias should be avoided.
4. Experimental design and equipment should not be changed in the middle of an experiment.
5. Unusual values, outliers, should be checked to see whether or not it is due to a sampling error.
6. Good notes should be taken and kept until the study is completed. Summaries of the data are not sufficient for statistical analysis.
7. Conclusions should only be made after proper analysis is completed.

The following articles and books discuss ecological field experimental designs and treatment affects.

R..A. Fisher and J. Wishart. 1930. The arrangement of field experiments and the statistical reduction of the results. Imperial Bureau of Soil Science (London), Technical Communication Number 10: 1-23.

         "No one would now dream of testing the response to a treatment by comparing two plots, one treated and the other untreated".

Snedecor G. W. and William Cochran. 1967. Statistical methods. Iowa State University Press. Ames, IA.

         This text is a standard reference for researchers in designing and analyzing data collected on projects. The theories of math are used to provide an objective result in determining when numbers are different due to patterns occurring within a population that are not due to chance.

Hurlbert. Stuart H. 1984. Pseudoreplication and the design of ecological field experiments. Ecological Monographs. Ecological Society of America. 54(2).

          There are five components to an experiment: hypothesis experimental design, experimental execution, statistical analysis and interpretation. Clearly the hypothesis is of primary importance, for if it is not, by some criterion, "good," even a well-conducted experiment will be of little value.

          It is clear that experimental design and experimental execution bear equal responsibility for the validity an sensitivity of an experiment. In a practical sense execution is more critical than design. Errors in execution can and usually do intrude at more than one point in an experiment; come in greater numbers of forms, and are often subtler than design errors. The effects of undetected or undetectable errors make experimental execution critical. Statistical analysis and interpretation are the least critical aspects of experimentation, in that if purely statistical or interpretive errors are made, the data can be reanalyzed. On the other hand, the only complete remedy for design or execution errors is repetition of the experiment.