Recommendations for online essay grading systems?

Who else remembers using the Flesch-Kincaid Grade Level metric in Word to evaluate school essays? https://en.wikipedia.org/wiki/Flesch%E2%80%93Kincaid_readabi...

Imagine my surprise when I learned that this metric is not one that was created for authors to maximize: reading ease for the widest audience is not an objective in some deparments, but a requirement.

What metrics do and should online essay grading systems present? As continuous feedback to authors, or as final judgement?

I'm reminded of a time in highschool when an essay that I wrote was flagged by an automated essay verification engine as plagiarism. I certainly hadn't plagiarized, and it was up to me to refute that each identified keyword-similar internet resource on the internet was not an uncited source of my paper. I disengaged. I later wrote an essay about how keyword search tools could be helpful to students doing original research. True story.

Decades later, I would guess that human review is still advisable.

This need of mine to have others validate my unpaid work has nothing to do with that traumatic experience.

I still harbor this belief in myself: that what I have to say is worth money to others, and that - someday - I'll pay a journal to consider my ScholarlyArticle for publishing in their prestigious publication with maybe even threaded peer review (and #StructuredPremises linking to Datasets and CreativeWorks that my #LinkedMetaAnalyses are predicated upon). Someday, I'll develop an online persona as a scholar, as a teacher, maybe someday as a TA or an associate professor and connect my CV to any or all of the social networks for academics. I'll work to minimize the costs of interviewing and searching public records. My research will be valued and funded.

Or maybe, like 20% time, I'll find time and money on the side for such worthwhile investigations; and what I produce will be of value to others: more than just an exercise in hearing myself speak.

In my years of internet communications, I've encountered quite a few patrons; lurkers; participants; and ne'er-do-wells who'll order 5 free waters, plaster their posters to the walls, harass paying customers, and just walk out like nothing's going to happen. Moderation costs time and money; and it's a dirty job that sometimes pays okay. There are various systems for grading these comments, these essays, these NewsArticles, these ScholarlyArticles. Human review is still advisable.

> How can we identify biases in these objective systems?

Modern "journalism" recognizes that it's not a one-way monologue but a dialogue: people want to comment. Ignorantly, helpfully, relevantly, insightfully, experiencedly. What separates the "article part" from the "comments part" of the dialogue? Typesetting, CSS, citations, quality of argumentation?

You could call it something like "Because I Want You To Grade My Essay Again" (BIWYGMEA) and just pay people who submit to it.