Test Yourself
True or False
Feedback
False
In most circumstances it is also useful to include the relative frequencies (%) for each category so the categories can be compared with each other.
Feedback
False
Either row or column percentages should be presented (these show the relative frequencies within each category of a variable). The choice of which depends on how the data were collected and which variable is the exposure and which is the outcome. Normally it makes sense to show the distribution of the outcome by levels of the exposure. However, when we sample individuals into a study based on whether they have the outcome or not (as a case-control study does), then the proportion with the outcome is fixed by the study design, in which case it is more sensible to present the distribution of the exposure by levels of the outcome.
Feedback
False
In exploratory and descriptive statistics we can make no claim about any true underlying association or difference in the population because we are only ever describing the data in the sample. We use statistical inference to try to make claims about the underlying population or any real behaviour.