This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
In sampling a population, all elements of the population must be available to the sampling procedure. For example, if one wanted to identify the prevalence of Avian flu in India, the sampling must be throughout all of India, which is probably impossible from a practical viewpoint. If a microbiologist sampled from Calcutta[0], Delhi, and Mumbai (formerly Bombay) and stated that the sample "represented India," this statement would be erroneous. The most one could conclude would be the prevalence in these three cities. Even then, more than likely, certain destitute people would not be available to the sampling schema, so the study could not be generalized to "all individuals" in these three cities. These kinds of potential sampling bias and restriction must be evaluated before one can generalize sampled data to a larger population.
- 2.
This is the average sample value. However, if one wants to assure that the average can be no less than 8.0 log10/mL, the H 0 hypothesis is rejected, because 7.935 is lower than 8.0. It is important to identify what one means in statistical testing.
- 3.
To be completely correct in sample size determination, the t distribution can be used, but it requires that one perform a series of iterations. A more straight-forward approach based on the z distribution is presented here that, in practice, works satisfactorily for the vast majority of microbiological applications. For the moment, we will use Table 2.5 to represent the z normal distribution table. We will look at the use of the t distribution later in this chapter.
- 4.
Here, a practical problem exists in determining the sample size prior to running the experiment. Before the sample size can be estimated, one must know the variability (variance [s2]) of the data. But one most often cannot know that until the experiment is run :(. The variance must be estimated based on other similar experiments, if possible. If none exist, guess. But be conservative, and use your experience in microbiology. After the experiment has been completed, recalculate the sample size based on the known variance (s2) for future reference.
Rights and permissions
Copyright information
© 2008 Springer Science+Business Media, LLC
About this chapter
Cite this chapter
(2008). One-Sample Tests. In: Biostatistics and Microbiology: A Survival Manual. Springer, New York, NY. https://doi.org/10.1007/978-0-387-77282-0_2
Download citation
DOI: https://doi.org/10.1007/978-0-387-77282-0_2
Publisher Name: Springer, New York, NY
Print ISBN: 978-0-387-77281-3
Online ISBN: 978-0-387-77282-0
eBook Packages: Biomedical and Life SciencesBiomedical and Life Sciences (R0)