Considering that the CT rates are fairly high during training, it could be surmised that the selection process has something to do with this. I don't think it is news to anyone that SHL won't be doing the testing for NavCanada by the end of 2008 or so (
http://www.avcanada.ca/forums2/viewtopi ... 68&t=38999 ), which is an indication that the end results aren't what NavCanada are looking for at this point.
(let me put up here that I am a two time interview washout with SHL, I have been through the mill twice and I know where I passed and where I failed)
The basic methods of SHL (beyond the initial testings) has been using is situational questions. "Give an example of when you had a problem with a manager and how did you resolve it" sort of questions. The concept is that past behaviors (actual actions) will forecast and predict your future actions in an entirely different situation.
Long details on a variation of it here:
http://www.era.org.in/Library/BestPract ... ctInt1.pdf
Better explaination here:
http://www.quintcareers.com/behavioral_ ... ewing.html
Behavioral Interviewing is said to be significantly more successful than other methods. Depending on the study (and the company hyping them) numbers from 25% to 60% more effective than unstructured interviews are mentioned. However, depending on the questions asked and the weighting of the value of the answers, you can end up with a system that is more likely to be successful for certain types of candidates and less successful for others. Effectively, candidates with less life experience, less job experience, and perhaps even less exposure to challenging work enviroments would do better than someone who has worked in a higher pressure enviroment.
BI basically looks at negative situations and sees how you dealt with them. Someone who has little real work experience might not have too many truly negative experiences, and it is likely the situation discussed was resolved reasonably well. However, many people work in high stress environments where the resolution to an issue might not be entirely to everyones satisfaction. That would plummet you down the scoring in a BI interview / testing situation.
As a result, there is potential that the ideal SHL candidate is someone who has never faced much true stress, but has reacted well to small stresses placed on them. Training to be a controller can be a heavy stress environment, and I think that the failure in the system may be as a result of the types of candidates that the current system prefers by it's very nature.
Aptitude testing is a pretty easy deal, it is a real go or no-go deal. It is a very functional way to filter out a large percentage of the people. In two sets of testing, I have seen people unable to fill out their name on the form, unable to follow simple directions (don't open the book until told to), and I am sure the intial tests wipe out about 80-90% of the people who come for testing, either as not able to do the work or not scoring high enough. There is no memory test, although the audio part of the test does require a certain amount of extreme short term memory to make it really work out.
At the end of the day, the results are what they are. For 1000 applicants to testings, about 100 are getting past stage 1, and probably less than 50 are getting past stage 2, and I am sure more are lost after that. Even after all of that filtering and poking and prodding, the TC rates for some of the training classes is extremely high. According to some numbers I have seen, the failure rates in some cases are as high as 80%.
Nothing against the people who are getting into the system, but with high failure rates either the training isn't very good or the wrong people are getting in to start with, and based on NavCanada dropping SHL and moving to some other setup, I am thinking the latter rather than the former.
