Given Hapunda

154 The Wilcoxon Test This is another example of a non-parametric or distribution free test. It is used to test the null hypothesis that the median of a distribution is equal to some value. It can be used (a) in place of a one- sample t -test (b) in place of a paired t -test or (c) for ordered categorical data where a numerical scale is inappropriate but where it is possible to rank the observations. Performing the Wilcoxon test in SPSS 1. Go to Analyze 2. Nonparametric Tests 3. Legacy Dialogs 4. 2 related samples. 5. Enter the variables into the required boxes and click OK. 6. The SPSS output generates information for reporting whether there are significant differences in the groups or not. Tests of Correlation A correlation is a statistical method used to measure and describe the relationship between two variables, represented by r . This test makes no assumption of causality but that there is a relationship between two variables. In correlation, there are no independent (cause) variables. Both variables are treated as “dependent,” meaning that they have not been ordered causally. A relationship exists when changes in one variable tend to be accompanied by consistent and predictable changes in the other variable. Assumptions to perform a parametric correlation test include entail that: Focus Box 2: Non-parametric Tests These nonparametric tests will produce descriptive statistics that accompany the results. In the case of selecting measures of central tendency and dispersion for reporting, it is a normal practice to select the median and the range. The median and the range are more appropriate tests because they are distribution-free tests and do not assume normality of the data. In addition, they are more appropriate for data based on ranks. These tests rank the data, and calculations are carried out on the ranks.