Almost half of Texas fourth graders scored a zero on the STAAR writing composition last year

From Houston Public Media:

Erika De La Rosa was excited when she heard about the sweeping changes to the STAAR exam.

“When they initially announced it, as an English teacher, it sounded like a fantastic idea because writing should be incorporated everywhere,” De La Rosa said.

She teaches seventh grade in Houston ISD. Before last year, writing was only tested on a separate exam in fourth and seventh grade, as well as in high school.

“What I saw for years was that because it was only tested in fourth and seventh grade, students would only get explicit writing instruction in fourth and seventh grade,” she said. “And so as a seventh-grade teacher, I would be scrambling to try to figure out how can I cover all of these grammar mechanics? How do I cover all of these writing techniques? What can I drop? What do we need to focus on?”

For the first time last year, every public school teacher in Texas from third through eighth grade had to grapple with those questions. The STAAR 2.0 redesign was complete, including a complete shift to an online testing platform, instead of the traditional paper exam. It also marked the end of the separate writing exams in fourth and seventh grade; instead, writing was included at every grade level.

“I’m not fully surprised because the test is so new and so different from how things have been in the past,” De La Rosa said. “Because years prior, the task was simply composition. Can they compose something that makes sense, that is all organized, that is grade-level writing? It didn’t require, like, a literary analysis.”

Students also performed poorly in grades where writing was assessed for the first time. Last year, 42% of third graders scored a zero on constructed response questions — along with about 25% of students in fifth, sixth and eighth grade.

Teachers, researchers and school leaders argue the new approach to writing assessment could kill creativity and curiosity in the classroom, especially with the way test scores tie into the TEA’s more stringent school accountability system. The TEA continues to project confidence about the changes, arguing the shift is intended to align the test with how teaching happens in the classroom.

‘Comparing apples to celery’

Chris Rozunik with the Texas Education Agency argued that year-to-year comparisons of constructed response scores are “kind of comparing apples to celery in some ways.”

As she pointed out, the shift from paper to online was not the only change. The writing composition questions are also different. They now include cross-curricular reading passages that reference multiple subject areas, and the grading rubrics for writing questions are more stringent.

“We are asking students to think differently about answering these questions,” Rozunik said. “When we set those standards last spring, we like to think about it as a very different world.”

It’s also important to note that an increase in zeros on writing compositions doesn’t necessarily mean an increase in the number of students who failed the STAAR. For example, 78% of fourth graders passed the RLA STAAR last year compared to 77% the year before.

Before the redesign, students had to get about half the questions on the test correct to “approach” grade level (to pass the test). Now, there are more types of questions, including ones that allow partial credit, and students only have to get credit for about a third of the questions to “approach” grade level.

“It’s not a percentage as we think of in school — you know, 70% is passing,” Rozunik said. “You need to step away from that scale entirely and think about the test as a whole, keeping in mind that we’re required to measure every student in the state of Texas, including the most high-flying, smartest kid on the planet. Within the scale which we generate, you need a lot of room above the approaches category, which is that very first level that students can be considered on track for success in the future.”

The TEA insists the online shift didn’t make the test more difficult, but researchers still have questions about that. With the exception of pre-approved accommodations for kids with disabilities, all students from third grade through high school now take the exam on a computer.

Jodi Pilgrim is associate dean with the College of Education at the University of Mary-Hardin Baylor. She’s also a former fourth-grade teacher.

“We’re wanting to assess how good they are at writing, but there’s a possibility that we’re also assessing how well they can type,” she said. “Those that are really good typers are going to do better on it.”

Concerns about creativity

Pilgrim also worries the new approach to writing could stifle creativity.

Before the STAAR 2.0 redesign, in the spring of 2021, fourth graders were given this quote: “Good teachers know how to bring out the best in students.” Their writing prompt: “Write about the best teacher you know.”

“I liked this a lot as a teacher,” Pilgrim said. “It was open-ended. Because if the prompt was a personal experience, any student can write about that. We tend to start our school years writing personal narratives because all students can tell about themselves.”

All students now read longer passages, and then write a response.

Last year, the fourth-grade passage was 500 words about salamanders, Barton Springs and the Edwards Aquifer. Fourth graders were then asked to write about the importance of the aquifer.

“‘Explain why the aquifer is important in the article,’ so I would just take that exact same sentence and say, ‘The Edwards Aquifer is important because …’ Well, that didn’t take much thought,” Pilgrim argued. “One of the things we really focus on in schools right now is critical thinking. The world we live in requires the critical thinker. I don’t really have to think critically to write these responses.”

Rozunick said the TEA intended to make the writing questions more universally accessible through the redesign.

“There’s enough information in that prompt, in any prompt, in any text we have moving forward to be able to really get enough information in order to answer the question,” Ruzunick said. “You do not have to bring in other life experiences or emotions in order to answer these questions. And quite honestly, they’re probably more fair and equitable to students than, you know, ‘what’s your favorite teacher’ or ‘what you did last summer?'”

Pilgrim expects teachers will develop and teach a writing formula for students, especially with the rollout of “machine scoring” for writing questions this year. Most student writing is now graded by a computer, not a person, as first reported by the Dallas Morning News. According to the TEA, machine scoring is not significantly different from human scoring, but it is cheaper and faster.

De La Rosa believes teachers will still have to spend time teaching to the test, but she doesn’t think the pressure will kill creative writing entirely — at least not in her classroom.

“I’ve been really fortunate to be on the campus that allowed me that flexibility or kind of trusted me to make those calls,” De La Rosa said. “There are plenty of campuses that I know and plenty of teachers in situations where they’re not trusted and they’re told exactly what they are allowed to do and what they’re not allowed to do.”

“So I think that is a very valid fear that STAAR being this way will limit a lot of creativity,” she continued. “Because as much as we would love to say, like, ‘Oh, don’t teach to the test,’ it really is circumstantial to what campus and what district you’re in because not all of us are at campuses that allow us the freedom, and not all of us are in districts that trust us to make those decisions for our kids.”

This debate is not new

The pressure to teach to the test doesn’t come from the STAAR exam itself — it comes from the way the state holds districts accountable for test scores.

One extreme example is the state takeover of Houston ISD, which was partially motivated by low test scores at one school. The TEA-appointed management board of Houston ISD wants to significantly increase the percentage of students across the district who meet grade-level standards on the STAAR, and TEA-appointed superintendent Mike Miles has implemented a centrally-created curriculum in dozens of schools that previously struggled with the test.

The pressure to improve test scores is a consistent presence across the state.

“Yeah, accountability can do that — those pressures of teaching,” Pilgrim said. “The district feels the pressure of it, the principal feels it. It trickles down, and the teachers feel it. What’s really unfortunate is when the students feel the pressure.”

In many ways, the test is now low-stakes for students. They do not have to pass the STAAR to move on to the next grade level. But STAAR remains high-stakes for teachers, schools and districts.

The Texas Education Agency assigns school districts and campuses an A through F accountability rating, based largely on students’ test scores and post-graduation readiness. As the STAAR 2.0 redesign unfolded, the TEA also overhauled its A through F system.

At a press conference about the A through F revamp in May 2023, Texas Commissioner of Education Mike Morath explained the reason for the more stringent approach.

“We want to eliminate achievement gaps by race and by class, and we want to ensure that Texas is a leader in preparing students for postsecondary success across the nation,” Morath said. “This is kind of the rigorous expectation Texas needs to be a national leader in preparing students for postsecondary success.”

More than 100 school districts sued the TEA, and the new ratings are currently held up in court.

Tension over test-based accountability has been around for a while, as Duncan Klussman with the University of Houston’s College of Education explained.

“Over the last 20 to 25 years, this has been a constant process where the state establishes a testing system, gives it a name, and then establishes an accountability system,” Klussman said. “And what you see typically is in the first few years of a new system, you see kind of lower scores overall. Then districts start to understand it better, and teachers understand better, and you see scores kind of grow over time.”

Throughout that time, debates raged over whether tests measure social inequality more than school quality, to what degree policymakers should use test scores to dictate how teachers work in the classroom, and whether there are more meaningful ways to measure “student achievement” than one test on one day of the year.

H.D. Chambers is the executive director of the Texas School Alliance.

“I don’t want to dismiss any superintendent or any board member or any elected leader who is hyper-focused on student achievement,” Chambers said. “But is that student achievement being defined by just one thing?”

If you found the reporting above valuable, please consider making a donation to support it here. Your gift helps pay for everything you find ­­on texasstandard.org and houstonpublicmedia.org. Thanks for donating today.