When should we use Kolmogorov–Smirnov test in SPSS research?

When should we use Kolmogorov–Smirnov test in SPSS research?

1 comment to When should we use Kolmogorov–Smirnov test in SPSS research?

• admin

  July 31, 2011 at 12:39 pm

  This is a non parametric test used by SPSS researchers, which is employed when the assumptions of any distribution are not put into consideration and also when the measure of the data is ordinal.
  In most cases it’s used to test central tendency and shape of distribution of two samples if they are equal.

  For instance, when testing for two sample group means one can use the independent samples t test with the assumptions that the data is normally distributed. If the distribution is not considered and /or the measure of the data is ordinal, then we can test the means of the two sample using Two Independent samples test.

  From SPSS Analyze menu, Non parametric tests, Two Independent samples test. Select test type Kolmogorov Smirnov Z from the dialog box.
  This test is also used by SPSS researchers when they are investigating the distribution of a data set. SPSS provide us with the one sample Kolmogorov Smirnov test which test if a data set follows Uniform, Normal, Exponential or Poisson distribution. Caution should be taken into consideration for a set of data that can be classified, which may result to biasness. For example, gender, religion etc. should be split before carrying out the test of the distribution.

  Procedure from SPSS, from the Data menu, Split file…, select compare groups from the split file dialogue box.
  For the other distribution SPSS researchers can use P-P plots from Graphs menu