To my knowledge it was common late 70's and early 80's (maybe outside this period - I don't know) for companies to adminster programming apptitude tests to applicants for trainee programming posts. Some companies (BIS was one) marketed such tests and maintained a central register of results. The tests were general IQ type test plus some questions on following instructions to write values into boxes and then operate on them (e.g. add the contents of the first box to the contents of the second box and if the result is greater than the contents of the third box subtract 9 from the contents of the fourth box etc) - very much like executing a program rather than writing one in fact! Were the tests useful? Not sure. I was recuited by a company (now sadly defunct) called Centre-File to as a trainnee programmer on the basis of such a test. Of the approx 7 people recruited at the same time 2 dropped out mainly for personal reasons during the 3 month training course. The rest of us survived and went on to be sucessful programmers (COBOL and assembler). I suspect that people who score well on such tests probably have a high chance of finding learning to program relatively easy BUT it's quite probable that people who score poorly on such tests may also make perfectly good programmers. At that time and in that situation companies were probably happy to miss out on some potentially good programmers if they could have reasonable confidence that most of those they did recruit would also be good programmers. When recruiting to HE courses we haven't really got the same luxury. It's no good identifying the top 10% and being certain they'll learn to program easily if that means rejecting 90% of whom perhaps half (but which half?) could also learn with a reasonabe investment of teaching resource. Regards, Gill <<<<<<<<<<<< Gill Windall [EMAIL PROTECTED] >>>>>>>>>>>> School of Computing and Mathematical Sciences University of Greenwich Wellington Street LONDON SE18 6PF Tel: 0181 331 8545 Fax: 0181 331 8665
