Statistics for #Research - L2 - The Concept of Descriptive Statistics

Descriptive statistics boil down to two jobs: summarizing what counts as “typical” in a dataset and quantifying how widely the values spread. Central tendency answers the typicality question—what a person’s height is likely to look like in a group—while dispersion (variability) answers whether those values cluster tightly or scatter across the range. Together, they turn a list of measurements into interpretable research information.

Central tendency is introduced through three measures. The mean is the arithmetic average: add all height measurements and divide by the number of observations. It’s presented as the go-to measure for interval and ratio scale variables. When values are ordered but not evenly spaced—like categories with a rank structure—the median becomes the appropriate “middle” value once the data are sorted from lowest to highest; it’s tied to ordinal scale variables. For nominal scale variables, where categories have no inherent order, the mode is used—the value that appears most frequently. An example with five people (male coded as 1 and female coded as 2) shows how the most repeated code becomes the mode.

Dispersion then shifts from “typical” to “spread.” The range is defined as the difference between the minimum and maximum values, giving a quick sense of how far apart the extremes are. But the transcript emphasizes that dispersion is broader than just extremes: variability describes how individual responses are distributed across the entire range. For a more reliable picture of spread, standard deviation is highlighted as the quickest way to gauge how much observations generally differ from the mean. Low standard deviation means most values sit close to the mean, while high standard deviation signals greater scattering.

Finally, the accuracy of the mean is linked to standard error. Standard error is described as standard deviation divided by the square root of the total number of responses. This connects sample size to precision: as the number of observations grows, the denominator increases, standard error falls, and the estimated mean becomes more accurate relative to the true population mean. The session frames these descriptive statistics—mean/median/mode for central tendency and range/standard deviation/standard error for dispersion—as the core summary information typically reported in research theses and papers.