Testing data to determine proficiency.


I have been giving a work project and I believe that I need to use t-tests to get to my results, but don't know what results I need to get to determine how long it takes our new hire employees to reach proficiency.

I have 3 separate data sets (approx 70,000 lines of excel data) which are sorted by months tenure (0-36) over the last 12 months, so someone that started a year ago will now be in tenure group 12. These three data sets include sales which increases as time goes by, callbacks which have a random line around the average but should go down, and time to service that should go down (and it does).

So I need to determine when the newbies reach a level of proficiency of the tenured staff. I am using the Data Analysis Tools in Excel, and I believe I should be using the t-test: Two-sample Assuming Equal Variances (?). What results should I be looking for in the Excel results table below to ensure that the newer tenure reaches the performance level on any (or all) of the data sets that I have? I am taking the values of tenure group 1 + 2 + 3.... compared against whichever tenure group I stop at through 36 months.

t-Test: Two-Sample Assuming Equal Variances

Variable 1 Variable 2
Mean 2 2.333333333
Variance 1 2.333333333
Observations 3 3
Pooled Variance 1.666666667
Hypothesized Mean Difference 0
df 4
t Stat -0.316227766
P(T<=t) one-tail 0.383821958
t Critical one-tail 2.131846782
P(T<=t) two-tail 0.767643915
t Critical two-tail 2.776445105

Thanks for your help!