Software to Simplify Statistical Practice ...

Wilcoxon Signed Rank Test

 

Performs Wilcoxon Signed Rank test for Single Group Median

The Wilcoxon signed-rank test is used to test whether the median for a variable has a particular value. Unlike the one- sample t-test, it does not assume that the observations come from a Gaussian distribution.

 

 Data

In a semiconductor manufacturing process flow, we have a step whereby we grow an oxide film on the silicon wafer using a furnace. In this step, a cassette of wafers is placed in a quartz "boat" and the boats are placed in the furnace. The furnace can hold four boats. A gas flow is created in the furnace and it is brought up to temperature and held there for a specified period of time (which corresponds to the desired oxide thickness). This study was conducted to determine if the process was stable and to characterize sources of variation so that a process control strategy could be developed.

Source: http://www.itl.nist.gov/div898/handbook/ppc/section5/ppc511.htm

 

 

Analysis 

  1. Open the DataBook nonparam.vstx
  2. Select the sheet oxide
  3. Choose the tab Advanced Statistics, the group Nonparametric Tests and the command Wilcoxon Signed Rank Test
  4. In Variables, enter THICKNESS
  5. Click Options page. In Hypothesized median, enter 560.
  6. Click OK

 

 

Output

Wilcoxon Signed Rank Test for Single Group Median

Test of median = 560 versus median not = 560
alternative hypothesis: true median is not equal to 560

                  THICKNESS
N                 168
Estimated median  562.5000
Mean Rank         10.3452
Rank Sum          1738.0000
Z Statistic       1.3765
Distribution      Normal Approximation
p-value           0.1687
alpha-level       0.05
Z Critical        1.9600
Conclusion        Reject the Alternative Hypothesis                  

Interpreting the results

There is insufficient evidence to reject the null hypothesis (p > 0.05). The population median is not statistically different from 560. The estimated median is 562.5. This median may be different from the median of the data, which is 560 in this example.