Header text

EssayTagger is a web-based tool to help teachers grade essays faster.
But it is not an auto-grader.

This blog will cover EssayTagger's latest feature updates as well as musings on
education, policy, innovation, and preserving teachers' sanity.

Monday, February 25, 2013

Where we're headed: school-wide data unification

EssayTagger was initially developed as a tool to help individual teachers. Now we're taking aim at unifying all writing assessment data across an entire school or district. Here's why this is valuable and here's how we'll do it.

The problem: Disparate writing assessments
In a typical school there's a wide array of teachers who assess writing throughout the year. And with the increasing emphasis on "reading and writing across the curriculum," those numbers are growing. English and Social Studies teachers are busy as ever grading essays, but now there are Math teachers who are assigning reflection paragraphs. PE teachers are assigning sportsmanship essays.

Unfortunately writing assessments are almost always isolated within the confines of each individual classroom. The History teacher knows that his students are struggling with using evidence in their writing, but he has no idea that his students' English teacher is frustrated with the same problem. There simply aren't any lines of communication across departments to share this information and collaborate on a solution.

But the worst offenders are the district writing assessments. Many schools will do a school-wide writing assessment that is scored on a standardized rubric which is then coded into a database so that administrators can pore over the results. These district writing assessments exist outside of the normal curriculum (e.g. in the middle of the Huck Finn unit the sophomores will be asked to write about texting while driving). Worse, the students rarely ever see the results and almost never receive any feedback. They're writing into a black hole. And, oddly enough, teachers often don't even see the results. They might see some bullet points on an institute day slideshow or get the data second-hand from their department head.

Disparate writing assessments are pure silliness.

Writing--and assessing--across the curriculum
We've bought into the value of writing across the curriculum and now it's time to unify assessment data across the curriculum.

Every writing assessment tells us more about each student and our school's overall trends. Assessment data shouldn't be cloistered within the walls of each classroom, but rather should be contributing to a rich web of highly interconnected data. This is the "web-ification" of the school structure; teachers need to think of themselves as part of a network instead of individual sovereign islands.

When Bobby's English teacher grades his essay, his History teacher should be able to see the results down to each individual skill being assessed. The district writing assessments should add to this pool of information and provide more insight to all of Bobby's teachers. The Culinary Arts teacher should know what strengths and weaknesses to expect when she assigns a research paper on launching your own restaurant. The school's writing center or peer tutors could benefit from a detailed skills profile for  each student that comes for help.

Administrators need data, but...
Administrators understand that data is just step one. Despite teachers' gripes, data is not an end unto itself. Data is hungrily sought after so that diagnoses can be made, problems can be identified, solutions can be proposed, and solution effectiveness can be gauged.

District writing assessments attempt to do some of this in a very blunt fashion. Some weaknesses in student writing will certainly be revealed but the remediation cycle is long, slow, and not terribly direct. Administrators would have to wait until the following semester's district assessment to see if there were any noticeable gains. Any impact is at the aggregate level and is rarely tailored to individual students.

And meanwhile the message to classroom teachers is fairly diluted and indirect. The district office meets with department heads and sets goals, the department heads then meet with their teachers, the teachers then try to meet the stated goals--all while balancing their own day-to-day curriculum goals that have already been planned for their class. Months later the next district writing assessment comes and the teachers may or may not find out the results that year.

There are just too many layers here, too much lag time, too much distance.

But aside from district writing assessments, there really aren't many other sources of data for student writing.

Unifying writing assessment data with EssayTagger
The beauty of EssayTagger is that teachers can evaluate writing more efficiently and more consistently (but I always have to emphasize this: EssayTagger is not an auto-grader; teachers make 100% of the grading decisions), and at the same time, behind-the-scenes, EssayTagger is saving all of those evaluations and tabulating the results.

It would be impractical for that Culinary Arts teacher to meticulously catalog and report every student's performance on each individual aspect of writing that she assessed. And even if she did, where would it go? How would that information benefit the English teachers, the Social Studies teachers, Math, Science, PE?

EssayTagger will do all of this for her. All she has to do is grade; we'll do all of the data collection, organization, sharing, statistics, and charting for her.

Step 1: Common Core
For any network of data to be useful, it has to be comparing apples to apples, oranges to oranges. It needs a common framework that provides underlying structure. For those of us in the teaching biz, that's Common Core.

Forget all of the confusion and complaints surrounding Common Core. Just think of it as a way to categorize what you already do. Every essay needs a thesis. In Common Core parlance that's W.1a. History teachers need to focus a lot on referencing credible sources. That's WHST.8.

So when the Culinary Arts teacher looks to assess her students' theses, she's evaluating W.1a. And when their English teacher evaluates his students' theses, it's still W.1a. Now both teachers are on the same page; they're doing an apples-to-apples comparison.

This is a huge win.

Now we have the basic framework for integrating writing assessments across the curriculum.

Common Core-aligned progress tracking
We already provide Common Core progression tracking for an individual teacher's classes. Now imagine this sort of data expanding beyond an individual teacher and reaching across the curriculum to include all writing assessments from every class in every subject:

This sort of analysis can also extend to the aggregate level. Are the 6th graders as a whole improving on W.1a? How are this year's sophomores doing compared to last year's sophomores? How about slices by department, by course, by teacher?

And aligning assessment to Common Core is actually shockingly easy, thanks to our Common Core Rubric Creation Tool. Just check off the skills you want to assess and then use that rubric to grade the students' writing; that's all EssayTagger needs in order to add your assessment data to the pool of Common Core-aligned data!

Step 2: Google apps accounts
If your school is not using Google Apps for Education, stop reading this and go get your school signed up. Now. Seriously. It's free for goodness' sake!! Go!

Through Google apps, each student gets a standardized school email account, Google Docs and Drive access, and a whole slew of other tools.

EssayTagger has just released support for student Google apps and Drive account integration. Students can now sign in with their Google apps account as a way to identify themselves. They can direct EssayTagger to pull a document directly out of their Google Drive accounts and submit it to their selected assignment. Pretty cool.

But it's the identification piece that really matters here. Now we'll know that the Jimmy in this English class is the same Jimmy in that Driver's Ed class. So when he submits his Distracted Driving Journal entry, the assessment data from that assignment will be added to Jimmy's growing writing assessment data profile.

Year-to-year persistence
Even better, Jimmy's Google apps account will follow him for his entire career at that school. This means that when he submits assignments next year, his future teachers will continue to add to his existing writing assessment profile.

This year-to-year data profile is another huge win.

Now next year's teachers won't have to spend time finding out Jimmy's strengths and weaknesses; they'll have years of data to look back on and will be able to compare his performance to that of his peers. It's absurd that teachers start from scratch each year with each incoming crop of students and, as a result, a good chunk of the first quarter is wasted on diagnostic let's-see-where-you're-at writing assessments.

More alignment to goals
This whole new data ecosystem allows schools to get more aligned around specific goals. Let's return to those district writing assessments. The district assessments should also be scored within EssayTagger so that they can immediately contribute to the pool of data that can benefit all teachers.

Administrators can analyze the district assessment data and the ongoing classroom assessment data to identify aggregate weaknesses. And because everything is aligned to Common Core, district goals can be stated much more explicitly: "We have to get better at W.1a!"

Classroom teachers can then immediately focus on the problem area and do a series of major and minor assignments that include W.1a. Administrators can then see new results in mere days or weeks as new classroom assessment data comes in for W.1a. By the end of the year administrators can see if this year's 8th graders are ahead on W.1a relative to last year's 8th graders.

The remediation cycle is so much faster, much more direct, vastly more effective, and much easier for classroom teachers to implement.

At the end of the year successes and failures can be reviewed and the school can iteratively improve its curriculum. If W.1a really is a persistent, difficult problem, let's add more focus on it in just about every class right at the start of first quarter.

We can even start to look at Common Core coverage. "Hey, we seem to be neglecting W.3 - Narrative Writing. How about we add a short creative writing unit? Or maybe the History classes can add a fictional first-person account of a major event?"

More coordination, more consistency
Different teachers will likely have wildly different ideas of what a good thesis looks like. But now that this assessment data is aligned, we can open up collaborative conversations to level expectations across the school. "Hey, I teach regular level sophomores too, but why are my W.1a scores so much higher than yours? Am I being too generous?" They can then have EssayTagger retrieve the exact thesis text excerpts and compare.

The most organized PLTs may have common expectations for their Sophomore English course, but how many schools have common expectations for what a sophomore-level thesis looks like in all of their sophomore courses, regardless of department?

If we're really focused on writing across the curriculum, it should no longer be acceptable for teachers in one department to have different writing standards (or none at all!) than their counterparts across the building. It's exactly this isolation, these varying standards that Gerald Graff, 2008 President of the MLA (and my former professor at UIC!), bemoans as constituting a detrimental "mixed message curriculum."

Imagine a school-wide effort to improve W.1a where teachers use consistent terminology with consistent expectations in every classroom across every subject. A thesis is a thesis is a thesis instead of a claim, a main idea, or a central point. Imagine PE teachers and Autos teachers actually collaborating with English and Social Studies teachers to determine what constitutes a good thesis.

This evolution won't all magically happen just because we unify all writing assessment data, but that unification seems a necessary prerequisite for those conversations to happen.

Assessment data unification is the future
We're a long way into the push for more and more data in schools. Now it's time to take the next step forward and unify classroom data across the school and align all of it with district goals via Common Core.

Schools have to evolve from being a collection of isolated classrooms to more of a highly-interconnected data web.

EssayTagger is building that future right now. We're looking for trailblazing, enlightened schools to join us and guide this effort. What are the must-have features of a tool that offers school-wide unified data? We are completely driven by teacher and administrator feedback; if you're onboard with us, you'll get a disproportionately large say in how these features take shape and where we set our priorities.

And we treat our friends well. Early partners will see substantial discounts to our school-wide or district-wide pricing. Start by filling out our quote request form.

Let's make it happen!