Descriptive statistics
Frequency distributions
The absolute
frequency shows how often certain responses occur. The relative frequency
(percentage distribution) means the percentage of individual responses
in relation to the whole data set. The frequency distribution can be
presented in a table or in a diagram (e.g. in a histogram).
Statistical parameters
Different statistical parameters can be calculated in order to obtain information about special characteristics of the attribute distribution:
Measures of central tendency: Which value best represents the distribution?
- Arithmetic mean: the sum of all the scores
divided by the number of scores (only for interval or ratio-scaled
data!). It is also used to calculate a scale value (an average score calculated
over all participants and all questions belonging to a scale).
- Analysis example for a comparison of project activities with the weighting of program goals
- The number of projects that have a certain program goal is compared with the weightings of the program goals (arithmetic mean): Projektaktivitäten und Zielgewichte [Project activities and goal weights] (pdf, 48 kB, in German).
|
- Mode: the most fre quently occurring score.
- Median: divides the distribution into two
halves with 50% of the scores above it and 50% of the scores below it.
Scatter: How different are the scores within a distribution?
- Variance: The sum of the squared deviations of scores about the mean, divided by the number of scores. Only for interval or ratio-scaled data!
- Standard deviation: The square root
of variance. Only for interval and ratio-scaled data!
- Range of variation: Difference between the
smallest and the largest score.
- Percentile: Difference between the
value that falls above the x % of the scores and that which falls below
the x % of the scores.