Nonparametric tests are typically used for what types of data?

Nonparametric tests are specifically designed to handle data that does not necessarily conform to the assumptions of parametric tests, which require normally distributed and interval data. They are particularly well-suited for nominal (categorical) and ordinal (ranked) data types.

When data is nominal, it consists of categories without any intrinsic ordering, such as gender or color preferences. Ordinal data, on the other hand, consists of categories that have a meaningful order but do not necessarily represent equal intervals between ranks, such as survey responses that range from "poor" to "excellent." Nonparametric tests can effectively analyze these types of data by using ranks or counts rather than actual values, allowing for valid statistical inference when the data does not meet parametric assumptions.

In contrast, normally distributed interval data, interval or ratio data, and continuous data typically require the application of parametric tests, which utilize specific statistical properties that assume certain distributions and calculations involving means and variances. Thus, nonparametric methods provide a flexible alternative suitable for analyzing data that does not meet these criteria, confirming their appropriateness for nominal and ordinal data.