A time and motion expert is interested in comparing two new methods of completing a certain task. The participants are divided into two groups of equal size and, for each group, their times (in minutes) to completion of the task will be monitored. It is anticipated that the completion times for each group will have a range of 10 minutes. If an estimate of the difference in mean time to completion of the task is recognized to be within ±2 minutes with 95% reliability, how large of a sample is needed (for each group)? What is the sampling assumption being made?