How would you estimate the amount of re-testing likely to be required?
-
Metrics from previous similar projects.
-
Discussions with the development team.
-
Time allocated for regression testing.
-
Both A & B.
The most reliable method to estimate re-testing effort is using metrics from previous similar projects - historical data provides objective evidence of typical re-test rates. While discussions with developers (B) can provide context, they're subjective and less reliable. Regression testing time (C) is a constraint, not an estimation method. Historical metrics offer the best statistical basis for estimation.
The most reliable way to estimate re-testing effort is to look at empirical data — metrics from previous similar projects — such as historical defect rates, fix-and-retest cycles, and time spent on regression in comparable work. This grounds the estimate in real evidence rather than guesswork or assumptions about how much time is theoretically available.