A levels have been with us since the early 1950s
Suppose it had been decided at the beginning to forego the complicated system of exams for a much simpler test
All children would have their height measured on their 18th birthday. Those over 5'8" would 'pass'. This is a guess, but perhaps the 1950 pass rate would have been 5%
If we still used the same test in 2007 there would not be much room for carping about whether standards had slipped or not. Though there would be interesting scope for, & acres of newsprint on, the relative advantage of being measured early or late in the day, the elasticity of tape measures, how precisely to convert from metric .... And accusations of unfair use of growth hormones, hair thickening agents, or ways of increasing the thickness of skin under your childs heels
The children who passed would be those over 5'8" tall, just as in their parents or grandparents generation. Standards would indubitably have risen, because more children would be passing
But we would not be selecting only the tallest 5% of children. If we wanted to do that, we would have to raise the bar, perhaps to 6 feet
Would height be a more or a less arbitrary test than ability to pass exams?
Well this is only a thought experiment, so we dont need to discuss that
But its interesting to note that the biases in the 2 systems would be remarkably similar. The pass group would include relatively more of the following categories:
boys
higher income/social class
educated parents
southern England
All factors which are in turn correlated with private education
