Hi Kashish,
These types of inference problems are done using a tree unfortunately this site has no drawing tools. I will just talk you through the calculations.
If i understand the problem correctly student #1 processes 60% and student #2 does 40% of the applications. We are also told that #1's error rate is 40% meaning for every 100 applications he processes 40 have errors and 60 are correct.
Student #2's error rate is 50% meaning 50 wrong and 50 right. I don't know about you but I would fire both of these students :)
We now have to calculate the probabilities of the 4 possible outcomes the first is p(#1/E)*p(E)=.6*.4=.24
The second is p(#1/C)*P(C)=.6*.6=.36
The third is p(#2/E)*p(E)=.4*.5=.20
The fourth is p(#2/C)*p(C)=.4*.5=.20
The meaning of these results are; out of 100 applications 24 were processed in error by student #1, 36 were processed by student #1 correctly and likewise for #2 20 in error and 20 correct.
If we now randomly choose an application and it has an error the probability that it was processed by student #1is p(E/#1)=p(#1/E)/p(E) (Bayes theorem)=.24/(.24+.2)=.24/.44=.545. So the probability is slightly over 50%
and p(E/#2)=.2/.44=.455.
So even though student #1 has a lower error rate #1 processes more applications and if you pick one at random and it has an error chances are it was processed by student #1, Cool eh?
Hope this helps
Jim