Are Children Learning

Explaining the ISTEP debate: 6 reasons why the test ballooned

PHOTO: Alan Petersime

The Indiana legislature is moving fast to cut at least three hours from the state ISTEP after two weeks of sharp words and behind-the-scenes negotiations over its length. Lawmakers are expected to rush a bill through both houses for the governor to sign next week to make the changes.

But with kids just days away from taking the exam, some are still asking: what caused the blow up?

The answer is a little complicated, but here are six reasons why ISTEP more than doubled in length from last year:

1. When standards change, tests must also change.

A big fight over Indiana’s academic standards last year ended when the state rapidly changed course and adopted quickly assembled new standards.

That disrupted a carefully coordinated plan in place since 2010 for the Indiana to adopt Common Core Standards along with 45 other states and use a shared exam that would test student knowledge with results that would be comparable across the country.

When Gov. Mike Pence and state Superintendent Glenda Ritz took office in 2012, Indiana had already adopted Common Core. Schools were putting it in place grade by grade, and a new Common Core-linked exam was scheduled to replace ISTEP this year.

But Pence was wary of the shared test — called the Partnership for the Assessment of Readiness for College and Careers or PARCC — and ordered the state to withdraw from the consortium creating the test in 2013. Six months later, both Pence and Ritz supported the idea of Indiana dropping out of Common Core and endorsed new locally made standards that were adopted last April.

Like Common Core,  Indiana’s new academic standards are more in-depth and ask students to do more analysis and critical thinking.

A test matching those expectations was needed in a hurry. Instead of taking years to adapt to the new standards and create the new exam, Indiana tried to do the whole process in a matter of months. That meant asking a lot of the 2015 ISTEP.

2. This year’s test had two extra goals — add questions to match the new standards and help create a test to replace ISTEP in 2016.

More difficult standards naturally meant Indiana needed a more difficult test. But there wasn’t time to completely overhaul ISTEP this year.

Instead, ISTEP was modified for this year to add several extra features. Many of the new standards were similar to the old standards, so many questions roughly matched the style and difficulty of past ISTEP exams. But new questions were added to also test students on new, tougher concepts included in the new standards, which were designed to make sure they graduate high school ready for college and careers.

The online version of ISTEP, for example, includes more advanced testing methods that ask kids to not only answer multiple-choice questions, but also answer questions in new ways, such as by dragging and dropping points on a graph or using drop-down menus.

Finally, this year’s ISTEP had one more job: Try out some questions that could be used on the 2016 exam.

But there was a problem. Indiana law requires release each year of all essay or short-answer test questions that are used in scoring. This would turn out to be a big factor in the length of the test.

3. A huge number of questions on this year’s test actually don’t count in a student’s score.

When test questions are released to the public they are effectively retired. They can never be used again on ISTEP.

So for this year’s exam, there were two big sets of essay and short answer questions: one group that counted toward each student’s score and must be released plus a large second set being tried out for use in 2016 that wouldn’t count.

Trying out questions is important. Test makers examine how students score on them to look for unexpected surprises. Questions they ask include: Was the question harder or easier for students than predicted? Was there reason to believe it was confusing to children? Was there any evidence the question was unfair to certain groups of students?

Trying out enough questions to be able to make a completely new test for 2016 was the main factor that caused what is normally a six-hour test to swell to more than 12 hours this year. All along, however, this was intended as a one-year problem. Future state exams are expected to be only slightly longer than the six-hour tests of the past.

The legislature appears poised to waive for one year the requirement that all essay and short-answer questions be released. This would allow some of this year’s questions to be reused so there could be far fewer extra questions that don’t count.

4. A longer test means more school days devoted to testing.

Indiana students don’t take all of ISTEP at once. They take sections of the exam in smaller doses over several days.

At its Feb. 4 meeting, the state board increased the number of days schools are allowed to use to give the test. The tests will be given over the course of almost a month, beginning Feb. 25 and ending in late March, followed by another set of testing days over three weeks at the end of April into May.

Schools can choose how to split up the parts of the test. Students might take just one section per day or do more depending on what teachers and principals decide. Danielle Shockey, the state’s deputy superintendent, said a testing day could take many shapes. In some schools, student take one 35-minute test section each day. In some schools, they spend an hour each day on testing. Other schools may do more.

“They have a long window of time,” Shockey said. “They can take one session a day if they so choose. It’s a local choice.”

5. Test makers had to consider that ISTEP is plays a critical role in school A-to-F grades and teacher evaluation ratings.

ISTEP is used to measure two things: how much students know of the content they were expected to learn this year, and how much they’ve improved from a previous year. Both factor into how Indiana measures the quality of schools with its A-to-F grading system, as well as how it evaluates teachers.

To determine a school’s A-to-F grade, the state considers both the percentage of students who pass ISTEP and how much students improved from last year. For teachers, the state expects to see their students’ test scores improve over the prior year.

When tests are roughly the same each year — measuring the same standards and using similar types of questions — it is easier to gauge how much students improved from the prior year. But when the standards change and the questions are crafted differently, test makers have to add extra questions to help determine each student’s improvement from the last test.

This spring’s test will include a few questions in English and math that are specifically designed to estimate roughly on what grade level each student best fits. For example, a fourth grade test might include a few third grade level questions and a few fifth grade level questions. Some students might do well on only the third grade questions but poorly on harder questions. Others might do well on all the questions, even the more challenging fifth grade questions.

Those extra questions help the test makers better estimate whether the student improved a little, a lot or not at all over the prior year. However, those extra questions also lengthen the test, but only by minutes, not hours, Michele Walker, testing director for the education department, said. The legislature agreed they were worth keeping — those questions will remain under the plan to shorten ISTEP.

6. Then, there’s the social studies question.

The federal No Child Left Behind Act, signed into law by President Bush in 2002, requires states to test students in English and math each year in grades 3 to 8, and once in high school, and also in science once during elementary, middle and high school.

Noticeably absent? Social studies.

Although Indiana’s social studies ISTEP test is only given to fifth- and seventh-graders each year, accounting for about an hour of testing for those grades, Pence’s test consultants recommended cutting that subject to reduce testing time further since it is only required by state law. That means the legislature could make an exception for this year.

State board members were divided on this idea. Some worried that it would send the message that social studies is not important. Others argued one hour for just two grades doesn’t add much test taking time.

But the legislature liked the idea of reducing test time further this way, so the Indiana Department of Education has told schools to expect the social studies exam to be optional this year. That means some students will take it, if the school decides they should, and others will be allowed to drop it for this year only.

Gradebooks

Three Chicago principals and the war against Fs

If you’re a principal intent on disruption, here’s one place to start: Ban Fs.

“Fs and Ds are worthless,” Principal Juan Carlos Ocon told a group of rapt educators Thursday. The principal of Benito Juarez Community Academy in the predominantly Latino neighborhood of Pilsen spoke as part of a panel on improving student performance at a conference hosted by the University of Chicago Consortium on School Research.

The event took place during a daylong look at the consortium’s latest round of pivotal research, which draws a clear line from ninth grade performance to high school graduation.

Conferees discussed the latest data showing freshman GPAs in core classes — such as reading, math, and science — dropping a third of a point from their eighth-grade GPAs. One key finding: Failure in non-core classes, like PE, far exceeds similar eighth- grade numbers. But researchers didn’t uncover why as many Chicago freshmen fail PE as science. (Read more here.)

PHOTO: Cassie Walker Burke / Chalkbeat
Sarah Duncan, left, of the Network for College Success, moderates a panel on grades at a conference Oct. 11, on findings of the To & Through Project. Also appearing on the panel at the University of Chicago are Juan Carlos Ocon, Chad Adams, and Wayne Bevis.

Joined on the panel by fellow principals Chad Adams of Roger C. Sullivan High School in Rogers Park on the North Side and Wayne Bevis of Robert Lindblom Math and Science Academy, a test-in school in West Englewood, Ocon said he took a hard position to “ban Fs from kids’ lives.”

“It actually increases rigor,” he said, explaining how the mindset of his school has shifted from punitive deadlines to encouraging learning at a student’s pace. Any high schooler who isn’t proficient in a subject by June must keep going to class until the light bulb glows, Ocon said. “Our classes do not end in June when classes end in traditional high schools — our classes extend through second week of August.”

Panelists Adams and Bevis are also “blowing up” the idea of Fs. At Adams’ school, located in an immigrant-rich neighborhood and inside which 40 some languages are spoken, Fs aren’t quite verboten — but, every five weeks, teachers have to come clean with how many Fs they give.

“Teachers didn’t like it as first, but then they started to hold each other accountable,” Adams said. I have the same kids (as you do) in your class, but, look, I gave 4 Fs versus your 54. What are you doing?”

Bevis has done away with As through Fs entirely and moved to a numeric grading system that runs 1 to 4. He’s also implemented a buildingwide revision policy, which can be controversial at some schools. After receiving a grade, students have at least two weeks to resubmit revised work and show they have improved their skills. “Some teachers go longer than two weeks, up to a semester,” he said.

Though located in very different areas of the city, each school has seen significant gains in student performance, with consistent, year-over-year rises in graduation rates and “freshman on track” percentages — that is, the percentage of freshmen who are on track to graduate as measured at the end of ninth grade, a metric developed by the University of Chicago and a key measure of success in Chicago.

The principals used the panel session to share other practices they see improving performance in their schools.

At Lindblom, for example, a revolving weekly “colloquium class” offers students extra help in a particular subject. Students must submit requests by Monday night, and with input from teachers a computer spits out their assigned special class, which can change week-to-week. “There’s a consistent understanding among teachers and students that we need to target which skills they struggle with,” Bevis said.

At Juarez, teachers spent the past year studying and recommending a set of core developmental competencies, a list that includes perseverance and relationship skills. Daily lessons are built in during an advisory period, and the staff is on board since they helped create them, Ocon said.

Adams echoed the idea of building a high-performance culture starting with his teacher corps. He’s likewise building a set of core values to express what a Sullivan High School graduate represents. When it comes to creating a learning culture, staff buy-in is essential, he said. When it comes to change, “if the teachers aren’t ready, the kids won’t be ready.”

 

held back

Holding middle-schoolers back causes dropout rates to spike, new research finds

PHOTO: Seth McConnell/The Denver Post
A student opens his locker between classes at Overland Trail Middle School on August 17, 2017, in Brighton, Colorado. (Photo by Seth McConnell/The Denver Post)

To hold back or not to hold back? For many policymakers in the early 2000s, the answer was clear: it was time to stop allowing struggling students to keep moving through school.

“It’s absolutely insidious to suggest that a functionally illiterate kid going from third grade, it’s OK to go to fourth. Really?” explained Jeb Bush, the former governor of Florida, where he curtailed the practice known as social promotion.

Former New York City Mayor Michael Bloomberg felt the same way. He introduced a policy of holding back low-performing students and fired appointees to the city’s school board who pushed back in 2004.

The idea was that the stricter standards would help students and schools alike. More time in school would give students the chance to catch up, allowing them to avoid the academic failure that could result from being continually promoted with big gaps in their skills. Thousands of additional students in Florida, New York, and across the country were held back in line with that theory.

Now, enough time has passed to see what happened to some of those students years later — and two recent studies reach a decidedly dire conclusion.

Being held back a grade in middle school, researchers found, substantially increased the chance that students dropped out of high school. In Louisiana, being retained in either fourth or eighth grade increased dropout rates by nearly 5 points. In New York City, the spike was startling: dropout rates were 10 points higher than similar students who weren’t held back.

A policy meant to make sure students stay on track, then, appears to have caused more students to leave school altogether.

“The takeaway from this would be that, at a minimum, we should be retaining fewer middle school students,” said Paco Martorell, a professor at the University of California – Davis who studied the New York City policy.

“If we’re talking about a middle school policy, I would strongly suggest against that at this point,” said Marcus Winters, a professor at Boston University who studied the effects in Florida.

Whether retention ultimately helps or harms students remains a crucial question. Though some places have relaxed their policies, others are adopting stricter rules. Michigan’s new retention law, for one, threatens to ensnare the vast majority of Detroit’s third graders.

The research also offer some better news, including out of Florida. Holding back students when they are younger doesn’t have such clear negative effects. And summer school, which often goes along with retention, can help students, potentially outweighing the downsides of retention policies.

Here’s what else the new research tells us.

Retention seems to increase drop-out rates.

The latest studies focus on Louisiana, New York City, and Florida. Each compares similar students, some who just barely earned a passing score on a test and others who just missed the cut-off, allowing researchers to zero in on the effects of being held back.

In New York City, the grade retention policy initially seemed promising. A 2013 analysis showed that retained students scored higher on state tests when they eventually reached the next grade.

The latest study, released earlier this year by RAND, looks at the long-run effects for those students held back between 2004 and 2012 and paints a starkly different picture. Students who were held back in middle school were much more likely to drop out of high school than the students who also went to summer school but who moved to the next grade on schedule.

There were no clear effects for students held back in elementary school, according to that recent RAND study. (An older Chicago paper found something similar: retaining eighth-graders increased future dropout rates, but retaining sixth-graders had no clear effects.)

In Louisiana, the recent research found that retention increased high school dropout rates for fourth or eighth graders who were held back between 1999 and 2005.

The rules around retention vary widely. In most cases, students are held back after they fail to pass a test, sometimes after summer help. In Florida, policymakers focused their policy on third grade, but other places, like New York City, introduced strict holdover policies in a number of grades.

There’s also lots of variation in just how often students are held back. Nationally, about 2 percent of students are retained each year, a number that has held steady or modestly declined since the mid-1990s.

In New York City, only 1 percent of students were retained across a number of grades. But in Louisiana, about 7 percent of fourth-graders and 8 percent of eighth-graders were held back. When the policy was first introduced in Florida, around 13 percent of third-graders were kept back, a number that eventually fell to around 5 percent.

Helping students catch up over the summer is beneficial.

Another recent study offers better news: In Florida, retention of third-graders in the early 2000s had no effect on their high school graduation rates, and it actually improved students’ grades in high school. The study also found that retained students saw an immediate test-score bump, though that faded over time.

What explains the more positive results? It’s hard to know, because the Florida study looks at not just retention but a package of policies that went along with it, including summer school and assigning students in the repeated grade extra reading help.

The Louisiana paper may shed some light on this question. It was able to separate the consequences of being held back — which appear to be negative — from the consequences of going to summer school. Sending eighth graders to summer school decreased their chances of dropping out of school down the line and their likelihood of being convicted of a crime before their 18th birthday.

In other words, the different results suggest that being held back hurts students, but the summer support that goes along with it helps them.

Retention is costly, though perhaps less so than some think.

There’s another downside to holding students back: it’s expensive to pay to keep students in school for more time. It costs both the school system and the student, who potentially misses out on an extra year of earning as an adult.

“Being retained may not confer benefits that justify spending an additional year in the same grade,” the New York City researchers concluded. “This is especially true given our finding that retention entails significant financial costs.”

The New York City study finds that each retained student costs the system roughly an extra $2,600 — a large amount, though far less than annual per-student spending.

White students are more likely to avoid being held back.

The consequences of retention, good or bad, are disproportionately felt by some groups of kids.

For instance, in Louisiana 85 percent of retained students were black, even though black students represented less than half of students in the state’s public schools at the time. In New York City, black students were more than twice as likely to be retained as white students with similar test scores.

Nationally, black and Hispanic students are substantially more likely to be held back. Some of that can be tied to test scores, but other research shows that white, affluent families are particularly likely to circumvent policies around holding students back.

In Florida, children whose mothers did not hold a high school degree were 7 percentage points more likely to be retained compared to their peers with equal academic performance whose mothers were college educated, another study found. The students who moved ahead anyway often took advantage of exemptions, like portfolios created by teachers to demonstrate that students should move on to the next grade.

There’s still a lot we don’t know about the effects of retention.

Where does this new long-term research leave us?

Although retention itself may be harmful to students, the combination of retention and summer school in Florida and Louisiana was neutral or positive. One potential takeaway is that districts should maintain extra help for struggling students while scrapping retention.

But those policies are intimately connected in many places, so it’s not clear that you can pull out one part of the policy like a Jenga piece and have the rest of the apparatus remain intact. Indeed, new research by Winters, the Florida researcher, suggests that the threat of retention can cause students do better in school.

It might also spur changes across a school or community. That’s what is happening in Detroit, where the retention law has focused attention on young students’ reading. “We have to get involved now and do anything we can to get the proficiency level up for the second-graders,” as one Detroit principal told Chalkbeat in August.

Martorell, the Davis professor, says we still need more evidence to know whether there are hidden benefits to holding students back. But he warned that existing research indicates that some students are paying a price.

“Policymakers should think long and hard about whether these other effects that are not captured by these studies … are significant enough to incur monetary costs and potential negative effects on students,” he said.