Vanja Pupovac

Summarizing and presenting numerical data

Scientific hypothesis and type of the study define variables that have to be measured. Measurements are determined by four distinct scales of measurement; nominal, ordinal, interval and ratio, producing two distinct types of data: categorical and numerical. Numerical data are usually summarized and presented by distribution, measures of central tendency and dispersion. For normally distributed data, arithmetic mean and standard deviation are used. For data not normally distributed, median with data range (minimum to maximum, interquartile range or percentile range) and mode are used. Commonly used graph types in descriptive statistics for numerical data presentation are error bar and box-and-whisker plots. Outliers are values that are numerically distant from the rest of the data and must be recognized.