Eliminating common misconceptions to enable intelligent use of biostatistics: How can a novice use statistics more intelligently?

Three types of numeric data commonly appear in the biomedical literature. These include nominal data, ordinal data, and numerical data, which can be expected to be normally distributed (parametric data). These three types of data are typically reported with descriptions of their central tendency and variability. More importantly, these three types of data are used for making statistical inferences. Specific biostatistical descriptors and specific inferential tests are appropriate to use with each type of data. Correct usage can be enhanced and errors of data reporting or statistical inference can be avoided by applying information in this chapter. However, selection of the appropriate inferential test does not eliminate the possibility of Type I and Type II errors when making statistical inferences.