**Cornell University Ergonomics Web**

** CUErgo Statistics
Helper: Descriptive Statistics**

**What type of data do you want to analyze?**

NOMINAL or ORDINAL DATA

I have nominal data (e.g. counts, frequencies) or ordinal data (e.g. rank data, rating scales with unequal intervals, such as a scale like "very poor", "poor" "good" "very good").

The most common descriptive statistics that are calculate to summarize nominal or ordinal data are:

- Simple counts (e.g. number of men and women in a sample)
- Percentages (e.g. percentage of men and women in a sample, % saying "good" or "bad")
- Proportions (e.g. proportion of men and women in a sample)

INTERVAL or RATIO DATA

I have interval data (e.g. equal interval rating scale, such as a Likert scale, measures, such as inches, degrees etc.) or ratio data (e.g. miles per hour, meters per second).

The most common descriptive statistics that are calculate to summarize interval or ratio data are:

- Central tendency measure (mean, median or mode)
- Variability measures (e.g. standard deviation of the sample, standard error of the mean, variance)
- Normality measures (e.g. measures of skewness and/or kurtosis)