Planning experiments hypothesis testing

As we may remember from Sections 2.3 and 2.4.10, the ANOVA technique is useful in cases where the number of results in each cell is different (but see below ). This may happen sometimes when single experiments fail or, in environmental analysis, when some samples are exhausted more quickly than others or when sampling fails. We also recognize ANOVA to be a valuable technique for the evaluation of data from planned (designed) environmental analysis. In this context the principle of ANOVA is to subdivide the total variation of the data of all cells, or factor combinations, into meaningful component parts associated with specific sources of variation for the purpose of testing some hypothesis on the parameters of the model or estimating variance components (ISO 3534/3 in [ISO STANDARDS HANDBOOK, 1989]). 

To test the validity of your hypothesis, your next step is to plan your experiments. You must begin by identifying as many factors as possible that could account for your observations. A factor that could affect the results of an experiment is called a variable. A scientist changes variables one at a time to see which variable affects the outcome of an experiment. 

Experiments A hypothesis means nothing unless there are data to support it. Thus, forming a hypothesis helps the scientist focus on the next step in a scientific method, the experiment. An experiment is a set of controlled observations that test the hypothesis. The scientist must carefully plan and set up one or more laboratory experiments in order to change and test one variable at a time. A variable is a quantity or condition that can have more than one value. 

Devise a sampling plan for one of the following investigations or experiments. In each case, the investigation can be stated in terms of a hypothesis that is to be tested. This helps to define the problems and objectives clearly. Ask yourself the types of questions put in the opening paragraph of this chapter. 

Design an Experiment Based on your observations and results, would temperature variations affect reaction rates Plan an experiment to test your hypothesis. 

Microarrays are observed as tools for descriptive research and not for hypothesis-driven research, which is primarily driven to answer specihc questions related to a known set of molecules. However, most microarray-based studies do have clear objectives and are designed to answer well-defined questions. The plan for specimen selection should follow from the objectives of the microarray study. Studies may either be exploratory, whose results should be confirmed, or designed to test various models in which case, the experiments have to be designed in a focused manner [126]. 

The basis for the scientific method is this from previous observations, a hypothesis is formed. An experiment is then planned to test this hypothesis. The test may either be one to substantiate the hypothesis or it may be a test to refute the hypothesis. The latter usually forms the more compelling evidence. Sometimes, for particularly strong proof, especially if the subject of the experiment is controversial, both experiments are conducted (see Heilman (2001) or de Kruif (1926) for examples). In the conduct of experiments, new observations are made that lead to additional hypotheses, so the pattern... 

As an extension activity, students could plan their own experiments to see whether (a) copper and (b) zinc oxide change mass on heating. To make these genuine experiments, students should work in small groups to first make a prediction and then try to support it with their reasoning - in other words, they should develop a hypothesis to test. On no account allow students to mix these two together and heat them. 

At a first glance these studies appear to provide merely a quantitative framework for common notions. However, because these experiments were planned to test a specific hypothesis they became part of a systematic approach to the elucidation of a problem. The purpose of briefly describing these experiments was not to substitute a working hypothesis whose merit must obviously be weighed against less subjective and better controlled experimentation, but to illustrate the complexity of the variables influencing the incidence of obesity. 

---