Adjusted Wald assumes a 95% Confidence Level (we'll talk about this next). If you're using something other than 95% Confidence Level, a different method is needed.
To keep it simple, we won't address an ALL failure or ALL success scenario here.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Now we come to our second adjustment. This one will impact our Confidence Interval. We need to again refer to our observed Completion Rate:
| Observed Data | Symbol | Formula | Value |
|---|---|---|---|
| Completion Rate | p | x/n | .90 |
Compared to the previous adjustment, this one is pretty straight forward. If the observed Completion Rate is between 0 and 1 (i.e. not ALL failures or ALL successes), we add two successes and two failures to our observed data. This is called the Adjusted Wald Method. As per the table below, we just add 2 to the numerator and 4 to the denominator.
| Calculated Data | Symbol | Formula | Value (given x=9; n=10) |
|---|---|---|---|
| Adjusted Success Probability via Adjusted Wald Method | pa | (p+2)/(n+4) | 11/14 = .79 |