Analyzing employee productivity in insurance claims

[chuie]

Hi There,

I  need some guidance in measuring a employee productivity  in insurance claims. I  have a following data elements of each employee  by per period.

EmployeeID  Payperiod  #productivehours  #netactivity #decisiontoappeal #closecase  #draft/sendapproval  #readmessage

1                        1                   80                          60                     5                           2               43                                 10

1                        2                   -                                -                  -                              -                 -                                      -

-                         -                   -                               -                    -                              -                -                                       -   

2                        1                  -                               -                    -                              -                -                                        -

The questions I have :

1. How is employee productivity measured?

2. Test the hypothesis that productivehours is correlated with #netactivity. If so how strong?

3. I am doubting on the  the 2 hypothesis is because # net activity is so much related to 1. type of activities( one in bold in data element) and 2. the activities might have weights like one can read 100 emails a day but can only send 2 draft/send approval.

4. So my question on this what is the best way to give weights to each activities ?

5. in future we want to do some sort of simulation research like we how to optimize our resources, what if we make a protocol that each employee has to  complete at least 60 activities how will that affect the com,completion rate  ?

 if you could help me answer those 5 question that would be great.

Thank you in advance

C

[tomrvincent]

I'd say Q#1 is something your boss or customer would need to answer. Might be a combination of items.

[chuie]

thank you... can you please also answer other questions or guide me to right path

[tomrvincent]

Are you familiar with proc corr?

[chuie]

yes I do

[tomrvincent]

So wouldn't that be useful for Q2?

[chuie]

 I did that the correlation  between # hours worked vs # of activities  and the correlation is 0.12. So that comes my question # 3 and 4 which I do not know the logic and any other option 

[tomrvincent]

so you'll have to get into multivariate regression. This should help you:

https://stats.idre.ucla.edu/sas/dae/multivariate-regression-analysis/