I've often thought to myself I'd like to see a sort of test for how subjective English is. Pick a random sample
n English essays
to get marked (where
n is determined beforehand, the larger the better statistically, though less feasible), and get several English teachers to mark each one independently (getting them all from your school to mark essays from your school will mean this is a test for subjectivity of marking at your school). Then, for each essay
, calculate: the mean mark
, the standard deviation
, and then the coefficient of variation
. Then, look at the mean
CV , the standard deviation of the
CV's
, and the
CV of the
CV's
over all
n essays.
Then compare these statistics with those of similar tests conducted on other HSC subjects, e.g. get HSC Physics markers to mark ~20 marks worth of extended responses, or get HSC Modern History markers to mark essays etc.
Then we could get some notion of how subject English actually may be compared to other subjects. (I'm guessing the
CV's for maths should be very low, for example. In general, the higher the
CV's, the greater the subjectivity.)