Finding an overall level of Motivation (from a motivational questionnaire)

#1
Hi, i was hoping i could get some help concerning the statistics of factors in a Motivational Questionnaire.

Long story short, in the company i work for, we are about to test a bunch of people and pick some people over others. The test is based on motivation, and the end result is results in sten of 18 scales. (The company whos test this belongs to does not openly publicize the way they found these scales (Lets call the company - company X)). Our aim is to somehow create overall motivation of these people.

Sadly, i have lost most of my university knowledge in statistics, but company X has given us some advice about finding overall motivation, and i m having a gut feeling that statistically, it would be wrong.

Firstly, we were told to pick a handful of factors from the 18. say 7. Then from the 7 we pair or group the factors where we see a similarity. So we end up with Sten results of groups of 2, 2, and 3 factors. We, instead of looking at the results in sten, think of them as raw scores, we add the raw scores in each group. Then, we find the absolute value for these groups of factors. Thats not all.. Then we do it all over again... So we again add the three numbers (as there were three groups of factors) and find the absolute value.

We were told that this is a sure way of comparing people, however, my question to you would be, is this the right thing to do?