Confidence Level for Change
Each change detected by the change-point analysis has a confidence level associated with it. The confidence level ranges from 0% to 100%. The higher the confidence level, the greater the certainty that a change took place. By default, 90% confidence is required to state that the change is significant. You can change the required confidence level using the Advanced Options dialog box.
The confidence level is based on what Statisticians call the significant level, alpha-level or p-value. The p-value is the probability that the observed effect could have been due to the natural variation in the data assuming no change took place. A p-value of 0.05 would indicate that the chance of the observed effect is low, 1 in 20, due to variation alone. This is good evidence that a change took place. A p-value of 0.5 would indicate that there is a 50-50 chance of observed effect even if no change took place. This offers little or no evidence of a change. The smaller the p-value, the greater the evidence of a change.
The confidence level is 100*(1 - p-value). Therefore:
Confidence Level p-value