What does the nonparametric method mean?
The non-parametric method refers to a type of statistic that does not require that the population analyzed respect certain assumptions or certain parameters. Well-known statistical methods such as ANOVA, Pearson correlation, t-test and others only provide valid information about the data analyzed if the underlying population meets certain assumptions. One of the most common assumptions is that demographics have a “normal distribution”.
However, parametric statistics can also be applied to populations with other known types of distribution. Nonparametric statistics do not require that demographic data meet the assumptions required for parametric statistics. Consequently, non-parametric statistics fall into a category of statistics sometimes called without distribution. Often, nonparametric methods will be used when the population data have an unknown distribution, or when the sample size is small.
Explanation of the non-parametric method
Parametric and non-parametric methods are often used on different types of data. Parametric statistics generally require interval or report data. An example of this type of data is the age, income, height and weight in which the values are continuous and the intervals between the values have meaning.
On the other hand, non-parametric statistics are generally used on nominal or ordinal data. Nominal variables are variables for which values have no quantitative value. Common nominal variables in social science research, for example, include sex, the possible values of which are separate categories, “male” and “female”. The other common dummy variables in social science research are race, marital status, education level and employment status (salaried versus unemployed).
Ordinal variables are those in which the value suggests a certain order. An example of an ordinal variable would be if a survey respondent asked, “On a scale of 1 to 5, 1 being Extremely Dissatisfied and 5 Extremely Satisfied, how would you rate your experience with the cable operator?
Although non-parametric statistics have the advantage of having to answer a few hypotheses, they are less powerful than parametric statistics. This means that they may not show a relationship between two variables when in fact one exists.
Common nonparametric tests include Chi Square, Wilcoxon rank sum test, Kruskal-Wallis test, and Spearman rank order correlation.