I'm sorry, but does alignment not occur after all the raw marks have been processed, and they determine the spread/avg marks across the state? I mean, raw marks differ each year, but are they not determined AFTER all the raw marks have been processed, analysed and weighted? Ir would not make sense otherwise, as each year, responses and total raw marks across the state may differ. If a subject has more raw marks towards the 90's, it is probably deemed easier. However, if it has more raw marks towards 70's and 80's, it is probably moderately difficult, thus would align better.