SPSS Parametric or NonParametric Test
 In this section, we are going to learn about parametric and nonparametric tests. If we use SPSS most of the time, we will face this problem whether to use a parametric test or nonparametric test.
 The first person to talk about the parametric or nonparametric test was Jacob Wolfowitz in 1942. He tried to draw a distinction between those tests, which make assumptions about the nature of a variable in their population. If we already know about the population and we develop a test basis on those assumptions and apply a test, in that case, our result is more generalizable.
 Suppose we are studying an age variable. Suppose we want to find out some conclusions about the age. We are already aware of how the age is distributed in the population or entire population or the Indian population or American population. In that case, whatever test we are going to use will give us a more generalizable result.
 While other cases, when we are not aware of the features of variables that we are studying, especially in the population, then we will not create a situation where the result would be generalizable. So that was the beauty of the parametric test. That's why our researcher, supervisor, or general editor often nudges us to use parametric tests more often as compared to nonparametric tests.
 The results of parametric tests are more generalizable as compare to nonparametric tests. In the Parametric test, we are sure about the distribution or nature of variables in the population. So if we understand this, we can draw a certain distinction between parametric and nonparametric tests.
Difference between Parametric and NonParametric Test
The following differences are not an exhaustive list of distinction between parametric and non parametric tests, but these are the most common distinction that one should keep in mind while choosing a suitable test.
S.NO. 
Parametric Test 
NonParametric Test 
1 
Normality of Distribution 
Nonnormal Distribution 
2 
Homogeneity of Variance 
Nonhomogeneity of variance 
3 
Independence of Observations 
Dependence of observations 
4 
Randomness 
Nonrandom 
5 
Interval scale measurement 
NonInternal Scale Measurement 
1. Normality of distribution shows that they are normally distributed in the population.
Nonnormal distribution specifies that we are not aware of the distribution of the population.
2. Homogeneity of variance specifies that different groups which we are using must have the same variance.
A1^{2} = A2^{2} = ……= An^{2}
Nonhomogeneity of variance specifies that the parametric condition might be violated in a nonparametric test.
A1^{2} ≠ A2^{2} ≠ ……= An^{2}
3. Independence of Observations specifies that observation of one candidate or subject in no way affect the observation of other candidate or subject.
Dependence of observations specifies that observation of one candidate or subject affects the observation of other candidates or subjects.
4. Randomness specifies that the sample must be randomly drawn from the population.
Nonrandom specifies that we are not randomly drawn to our sample, and all the subjects which are part of our study will not be randomly selected.
5. Interval scale measurement specifies that our data will be measured in an interval scale, and the quantity of measurement between two intervals of a scale remains constant throughout the scale.
NonInterval scale measurement specifies that the parametric condition might be violated in a nonparametric test.
