In the world of educational measurement nothing is what it seems
Achievement tests might in fact be aptitude tests. Aptitude tests essentially are tests of intellectual abilities. Tests on intellectual abilities in fact test mainly for differences between pupils in those same abilities, something else altogether! Most assessment specialists will tell you that reliability of tests is of the essence. No, it isn’t. Not even their validity. Nor their utility. In education the worth of achievement tests lies in the information one can get from them: for the pupil, the teacher. Wait a sec, aren’t they used for other purposes mainly? Placement? Selection? The award of diplomas? Accountability of schools or, god forbid, of teachers? International comparisons? To earn a buck? For another approach altogether, taking the individual pupil as the main actor, not the teacher, nor superiors, test developers, politicians or think tanks, see http://goo.gl/c4DmFL and be amazed.
To understand what is going on some history might be enlightening ( http://goo.gl/PQd1t ). Did you know that grading is just a standardised form of ranking? From social media I have picked up where and how teachers are struggling with assessment. If I have some advice to offer here, it will be based on psychology (its branches in cognition, personality, selection, experiment). And on a lifetime in educational research. How boring can that be?
Whatever. The interesting thing is: all of these viewpoints, inroads, aspects, they hang together and inform each other. It is easy to see, at least for hard core psychologists it is, how progressivist ideas of education, including those non-existent generic 21st century skills, work out as a forthright disaster in the field of educational measurement. Important issues here, especially on education(al opportunity).
Having ‘the whole picture’ might inform your daily assessment activities, at least by developing some sensivity for possibly unsuspected ethical issues involved.