OK. Now, I am beginning to understand what you are saying. Thank you!
The student is to write only what was taught by the teacher and to not add any extraneous information, no matter how relevant it is. Thant would ensure most teachers not docking points off the answers.

Yes, but, that's only true to the extent that you're in it for the maximal points. I tend to encourage my students (and my daughter) to view the points given as a largely extraneous factor with little significance to the quality of any given educational experience.

But, yes, if you're trying to maximize your score, stick to the specific substance of what has been taught and clearly and concisely answer the question.

For some reason this reminds me of the year that #1's school switched to Pearson eTexts. The history teacher was initially unaware that there were quizzes embedded in the text that students could practice until they reached 100%. After a disastrous attempt at compiling her own exams (a long story involving test questions with no connection to the text, and technical difficulties with the tests supplied with the curriculum), the presence of these quizzes was pointed out to her, so she started using the same questions for actual paper quizzes. (Had to ask #1 how to screen capture and print them, though.) It's amazing how test scores rose after that!

As to automated grading: quite often, teachers don't set eyes on the work products or assessments at all. We routinely use computer-adaptive assessments for reading comprehension, essays, mathematics, vocational skills, even employability/soft skills.

There is nothing quite like the particular hell engendered by trying in vain to properly FORMAT an otherwise completely correct answer/solution to a lengthy problem in chemistry, physics, or calculus.

Just saying-- this experience is fairly fresh in my mind. It drove my daughter crazy that her chemistry homework portal would randomly select a number of significant figures for constants-- without telling the student what it had chosen for a particular problem. Sometimes it took 15 minutes or more to "guess" at the right value, and she lost points for every incorrect answer. Yes, really.



Let's also recall that such programs are created and instructed by human beings, too, and that they therefore contain errors. She found a few of those, as well.

My personal favorite was the resonance structure question which rejected her answer because--



it was entered "upside down" relative to the solution. There is simply no legitimate way to consider that answer incorrect.

So yes, while automation is a distinct improvement upon incompetent educators running classrooms, it only works well when it isn't some monolithic, inflexible beast that must be fed the proper input stream at all times. Computers are only as smart as those providing the instructions for them.

It's still not what I'd call good, relative to an actual expert human being, in other words.

Thank you for all of your replies. The teacher explained that earlier that day she told the class exactly what she wanted as a response and that it was as much a test in listening as in knowledge.

As long as she did state it clearly, tell them while she was stating it, everyone was there and they had the opportunity to write instructions down and seek clarification that is OK. I can never remember verbal instructions later and I don't think I am the only one.