Chapter 2.5.6 : The Role of Statistical Thinking in Environmental Microbiology

Abstract:

Environmental microbiologists collect data, lots of data. However these data are not always the appropriate data to answer the questions the researchers are interested in. In this chapter we discuss what needs to be done to ensure that what you think you are discovering is in fact what the data are saying. We encourage a large amount of statistical thinking prior to the first data point being collected or before the first sample is obtained. Statistical thinking is not generally taught in introductory statistics classes. The nuts and bolts of what will be discussed in the chapter do find their way into statistic classes however not necessarily in a way that prepares scientists for the task of doing science. Here we discuss concepts such as defining the problem, experimental design, the weight of evidence, statistical power, and sources of variation, scope of inference, measurement scale, scale transformation and other topics. We show that environmental microbiology done without careful thinking before, during and after data collection rarely can answer any important question - regardless of how big the spreadsheet is. After reviewing countless papers for over three decades it is our experience that many studies in environmental microbiology are statistically weak and more importantly statistically flawed and that one need read no farther than the methods to decide whether to continue to forward.