Signs (epidemiology)

The subject of statistical research is signs (properties). These can be quantitative (they can be expressed by a number) or qualitative (the character does not have a numerical character). We divide quantitative traits into continuous ones, which can take on arbitrary values ​​from some interval (e.g. height, weight or solution concentration), and discrete ones , which can only take on selected values ​​(e.g. number of siblings, year of birth or number of CFU in bacterial suspension). Qualitative signs can be coded with a number for further processing, but this number may not have any meaning (e.g. 1 = red, 2 = blue, 3 = green).


 * Qualitative signs: Their values ​​can be given by verbal expression.
 * We can only list the values ​​of nominal signs (e.g. a man's marital status: single, married, divorced, widowed).
 * Ordinal characters allow you to arrange their values ​​according to size (e.g. education: primary, secondary, university).
 * Quantitative signs
 * continuous – take on arbitrary values, e.g. height, weight;
 * discrete – only acquire separate numerical values, e.g. the number of children.

The interval scale of quantitative signs allows you to arrange the observations in relation to the degree of the investigated property and allows you to determine their exact distance. On an interval scale with a variable start, the zero position is a matter of choice (eg arbitrary time scale). The zero position on an interval scale with a fixed start is precisely given and expresses the absence of the measured property (e.g. weight measurement in kg).

Související články

 * Statistický soubor