The Woodward News

February 16, 2014

Concerns pop up over third grade reading tests

By Rachael Van Horn
Associated Press

Woodward, Okla. — A new law that could impact a number of Woodward School District’s  third graders has school administrators working overtime to ensure the students take a test that is fair.

The Reading Sufficiency Act requires third graders in Oklahoma to score satisfactorily on the reading portion of the Oklahoma Core Curriculum Test (OCCT) to be promoted to fourth grade.

Students who do not pass the test will be given the opportunity for intensive remediation over the summer months provided by schools then retested using one of the four state approved alternative tests. If they cannot pass the alternative test or do not qualify for one of the six “good cause” exemptions they will be retained, according to the literature issued by the Oklahoma Department of Education.

But through the study of samples of those alternative tests as well as the third grade OCCT tests from 2011 and 2012, an administrator locally noticed the complexity of some reading phrases contained within the tests.

The test questions seemed to require a higher grade level of reading ability, according to Woodward Assistant Superintendent Tom Fisher.

That got Fisher interested. So he studied the first alternative test he received in the mail called the Terra Nova.

“When I was looking at it – Deana (Fisher, the dean at Northwestern-Woodward) and I have a second grade granddaughter who is a very good reader – I knew, looking at this test, that Kate (his granddaughter) would have a hard time reading it,” he said.

Intrigued and concerned about what this must mean for Woodward third graders, Fisher, a mathematician and statistician, began to dig a little deeper.

Fisher Googled “readability tests” and found several types of tests - also used by the state - that determine the grade level readability under which a test is written.

“I did it the old fashioned way by counting syllables and I entered that information into the formula the readability test gave me and it came out 5.8, almost sixth grade reading level that’s required to read some of those test questions,” he said.

Alarmed by this and thinking he must have made a mistake he requested help from Susan Viles, Woodward School District assessment coordinator.

The two then ran all four of the alternative tests through a computerized battery of approved readability tests.

This was done by applying randomly chosen phrases from the tests into the readability monitor on seven different methodologies used nationally to basically “test the test.” Each uses a computerized model that counts syllables, word length and sentence length to determine what grade level the test question is written, Fisher said.

Again, it came out to about an average of a sixth grade reading level for all four approved alternative tests.

That got Fisher thinking about what grade level of reading was required for third graders to pass the OCCT test. That is the actual “high stakes” test slated to be given in April which will determine who will be promoted to fourth grade or who will be at risk for being retained.

“We applied the same testing to the 2012 and 2011 OCCT test and it was averaging at the fourth grade reading level,” he said.

Fisher took his concerns to Woodward Superintendent, Tim Merchant who has begun to pose the questions to state leadership. Fundamentally, the district believes if third graders are going to be tested with a “high stakes” outcome – that is being retained - based on their ability to read at the third grade level, the test should be written at the third grade level.

According to Terri Brecheen, executive director of literacy and early childhood for the Oklahoma State Department of Education, there are aspects of how each test, the alternative test and the OCCT high stakes test, are being utilized that need to be understood before a determination they are somehow unfair is made.

Brecheen said Oklahoma students are being tested using two different types of testing methods. One is called “criterion-referenced testing” and the other is called “norm-referenced testing.”

Criterion-referenced tests are used to measure student mastery of instructional objectives or curriculum (absolute performance), rather than to compare one student with another or to rank students. They are often used as end-of-unit tests in textbooks or as a "benchmark" to identify areas of strength or weakness in a given curriculum, readiness to move on to a different level of instruction, etc. . .  she said.

According to Brecheem, the OCCT criterion-referenced test should include grade level work, which students in the third grade should know.

The alternative test, which would be offered to a student who scored unsatisfactorily on the reading portion of the OCCT, is a “norm-referenced test”.

Norm-referenced tests compare an individual child's performance to that of his or her classmates or some other, larger group. Such a test will tell you how your child compares to similar children on a given set of skills and knowledge, but it does not provide information about what the child does and does not know. Scores on norm-referenced tests indicate the student's ranking relative to that group,” she said.



So for instance, if your child scores in the 25th percentile, he or she is reading equal to or better than 25 percent of the students the child is being compared against, she said.

This is a test that does indeed have questions that are at fifth, sixth, seventh and even eighth grade levels as well as second and third grade levels because there are a certain percentage of children who can answer those eighth grade test questions, Brecheen said.

Those alternative tests - the norm referenced tests - are graded on a Bell Curve using scores of all third graders nationally as the standard that sets the Bell Curve, she said.

A student who takes the alternative test after receiving remediation must score in the 45th percentile on the alternative test to be promoted, Brecheen said.

The norms on every one of those tests are in the third grade range, she said.  

“If you look at many of the other states that have also enacted this law, they are setting theirs at the 50th percentile and we chose the 45th because we wanted to be fair and give them every chance of succeeding,” Brecheen said.

According to Florida Department of Education spokeswoman Cheryl Etters, in April students there take a state assessment and must score satisfactorily on the reading portion just like Oklahoma students.

In Florida, the state third grade reading test is a criterion-referenced test that does include some questions of a more difficult nature to capture who in each school is reading above the third grade level, Etters said.

“We have to provide those more difficult questions otherwise how would we know about those students who are reading above grade level,” she said. “But it is still the third grade reading test.”

Students there also have a chance to take an alternative test, which is also criterion-referenced.

The argument from Woodward School District administration is pretty straightforward and not at all difficult to understand, Fisher said.

“The bottom line is, if they are going to test third graders on the ability to read at the third grade level, they should use third grade material,” Fisher said. “Those norm referenced tests are for assessment. You do not use assessment tests to determine if a child can read at the third grade level.”

His argument drills down to the fundaments of teaching and working with students.

“What they are not considering is the emotional state of a third grade student who gets into the test and maybe the first question in written at the eighth grade level,” Fisher said. “He gets to that first question and says to himself, ‘I can’t even read this word,’ and then his anxiety level goes up and maybe the student who could read perfectly well at the third grade level, still performs poorly on the test.

That opinion couldn’t be more simply stated than by Mandy Cheap, the parent of a Woodward third-grader.

“I do not agree with high stakes testing, especially at such a young age. Basing a child’s pass/fail status on one test score is ludicrous,” she said. “Telling an 8 year old child that they will have to flunk a grade if they don’t pass a test is absurd.  When do kids get to be kids?  I know how important reading skills are to a child’s education. I know without a firm foundation in reading all school subjects will be difficult but shouldn’t we be more focused on teaching these children to read and not so focused on testing.”

 

That is the emphatic and vocally proclaimed opinion of Professor Shane Jimerson of the University of California – Santa Barbara. He calls high stakes testing with the threat of retention “complete garbage.”

For more than 20 years, Jimerson has studied the effects of retention as well as proven methods to teach reading that are, he said, successful and produce positive results.

He notes that within these reading sufficiency laws cropping up in many states, there are truly productive and evidence based teaching methods that are also included such as reduced teacher-student ratios and high-performing teachers. But he says the linking of the test to being retained is so damaging that it cancels out. almost, the effects of the positive teaching models proven to work on students who have difficulty reading.

According to Jimerson’s research and numerous data, retaining children doesn’t help the child read better. Instead studies show it produces stress at the same level as that experienced with the death of a parent or going blind, Jimerson said.

 Jimerson added that retaining students because they failed a test shows in many studies to be the single most significant predictor of a student dropping out of high school as well as more aggressive acting out and other deleterious effects.

“This is legislation and policy making that is inconsistent with science,” Jimerson said. “It should be about the reality not the rhetoric.”

For Fisher’s part, he understands the legislation is in force. But he wants legislators to do the right thing and create a program that uses sound science, proven teaching methods and assessments that are fair, don’t set students up to fail and measure what they are supposed to.

His request is for the legislature to rewrite the law in a way that allows teachers and administrators to write a test that measures actual third grade reading ability.

“We need to get a committee formed of third grade teachers at all levels, get superintendent’s recommendations for those teachers from all different levels of schools and meet in Oklahoma City,” Fisher said. “You get together and you write reading passages and you check the reading level using approved methods and it has to be on the third grade. It can be done, because I did it for eight grade math and I know it can be done.”

*****

Reading Sufficiency Act  requires benchmark assessments in kindergarten through third grade. Many schools assess pre-kindergarten students in literacy. The law mandates schools to use these benchmarks to identify children who need intensive intervention in reading beginning in these early grades and to notify the parents of these students in writing.

 

Moreover, the school must develop for that student an individualized program of reading instruction that includes:

• the child’s specific reading difficulty

• the intensive teaching practices to be

implemented

• how often progress will be monitored

 

Moreover, the plan must ensure enough time is given to the student to achieve grade-level reading.