Sample size calculation in entomological studies using R language of statistical computing Part 1: comparison of means and proportions
Subject Areas : entomology and othea arthropods
1 - Department of Plant Protection Isfahan Research Center for Agriculture and Natural Resources
Keywords:
Abstract :
In many entomological studies, a smaller sample is taken than a statistician would advise. Clearly, such experiments would often fail to detect any significant effects, resulting in a complete waste of time and money. The present paper introduces statistical power of a test and its importance in determining the minimum sample size required for detecting significant effects. A number of experiments fail simply because issues of statistical power were not confronted at the planning stage. A fundamental question that is needed to address at the planning stage concerns the number of samples required to test the hypothesis of interest. The answer depends on the variance of response variable, the size of difference in the response variable that would be detectable as significant, the risk of a Type I error and the risk of a Type II error. Here using R language of statistical computing, the specific methods of the calculation of optimum sample size for laboratory and field studies are presented for two common scenarios.
_||_
