We realize that teaching reading is serious business, but sometimes you need to retain your sense of humor in order to maintain a perspective. This article is intended to give you a little perspective on two of today’s major trends, RTI and testing, and their impact on using the Stevenson Program. Whether you enjoy using Stevenson now or you want to try it in the future, the following information may help you avoid becoming overwhelmed by all the initials.

 

RTI stands for Response to Intervention, and educators across the United States know it has become a big deal. In some places it is called “tiered intervention” or the “three-tier model,” but most frequently, it is referred to by its initials. RTI is not a curriculum, a method or a test. It is basically a model for adjusting instruction to help students who are struggling. (For the purposes of this article, we will focus only on reading instruction, although the same model can be used for other subjects.) RTI represents a major shift in how most schools determine which students get reading help, who will deliver the help, and when. In most cases, students’ responses to particular interventions will determine if they qualify for assistance through special education. Historically, most school systems used a “discrepancy” model (which we will discuss at another time) to determine such referrals. Since special education costs are a major issue in almost all school systems, and since a “special ed” designation has major implications for any student, you can see why RTI is considered a big deal. 

 

The basic thrust of RTI is simple enough. All students start out using the same reading method and same kind of materials. For the RTI purposes, it does not matter whether the initial curriculum is oriented towards phonics, guided reading or balanced literacy, and it doesn’t matter if the materials are a basal series, leveled readers or some combination of literature and supplements. The only thing that really matters is that students’ reading skills get assessed, and that the school system intervenes when students struggle.

 

The Tiers

 

Most schools adopting the RTI approach use a three tier model, although sometimes it is broken down further into four. In Tier 1, students who have difficulty learning to read receive the first intervention. Usually this is simply some additional instructional time using the basic general education reading curriculum in a small group in the regular classroom. The intervention is supposed to be closely monitored, often using a Curriculum Based Measurement (CBM). Students are not supposed to stay in this tier very long (perhaps eight weeks), but in practice the length varies considerably. Students showing significant progress return to “regular” instruction and students not showing adequate progress move on to Tier 2.

 

Tier 2 students receive more intensive instruction, usually delivered in small groups, and sometimes outside the regular classroom. The additional instruction often has to be carved out of regular scheduled activities. At this point students sometimes, but not always, begin to receive a different reading method. As in Tier 1, Tier 2 intervention is not supposed to last very long, but it is likely to last longer than in Tier 1, and the length again varies considerably from place to place. Parental involvement should begin early in the RTI process, but often it does not become formal until students reach Tier 2.

 

Students who do not respond well in Tier 2 will move on to Tier 3, where the situation changes more dramatically. Interventions become more intensive and more targeted. They are likely to include a more significant change in instructional methods and reading curriculum. Students receive more detailed evaluations. Parental involvement increases, and at Tier 3 the decision to designate students for special education is addressed. In some school systems, when students are actually given formal special education plans and services, they are considered to be in a different category. In effect, this creates a fourth tier.

 

The basic RTI plan seems like common sense. Still, many questions arise. How long should you implement any given intervention before you decide it is or is not working? Just how much do you change the instructional methods you use from tier to tier? What about materials? In the RTI process, one of the biggest questions by far is, “How do you assess whether a given intervention is actually working?” And, of course, as with many big questions, there is no simple answer.

 

A non-educator may wonder why educators cannot simply test a student to see what grade level material he or she is able to read, and use that measure to evaluate progress. If a second grade student scores at 2.1 grade equivalent and then six months later scores at a 2.7 grade equivalent, one might assume things are going well. But grade level scores in education are not like body temperature or blood oxygen levels in medicine. In Assessment for Reading Instruction (page 30), McKenna and Stahl state, “Without question, the worst norm typically reported on group achievement tests is the grade-equivalent score.” In individual achievement scores, grade level can be even more problematic. Therefore, educators have to consider a variety of measurements in addition to grade level.

 

As all the initials and acronyms in the opening paragraph imply, there are many kinds of tests. Although it is important to understand the purpose and value of different tests, almost all public school teachers are aware that one test looms larger than all others - whatever high-stakes test their state uses. In Massachusetts, the initials are MCAS (Massachusetts Comprehensive Assessment System), in Florida FCAT, in Texas TAKS, in New Jersey ASK, in Ohio OAT, and there are many more.

 

In some states the big test is simply called the “state CRT.” The initials CRT stand for Criterion Referenced Test, an important and prevalent category of tests in which specific criteria are being measured, and test results indicate how well the test taker has met the criteria. It is important to note that most of the high-stakes state CRT’s are not norm referenced. Norm referenced tests (NRT’s) are given to large numbers of students of a particular age range to establish a norm, and the scores reflect what statisticians call a normal distribution. Examples of norm referenced tests are the SAT9, the CTBS (which includes the “Terra Nova”), the Metropolitan, and the Woodcock-Johnson. Because many state CRT’s are not norm referenced, they can vary considerably from state to state, and they are not always the best indicators of group achievement. More importantly, neither the NRT’s or state CRT’s, although they receive a huge amount of attention, are practical as measures of individual achievement in the RTI model.

 

Many of the tests mentioned above are not given before third grade and many are not given each year. Therefore, they cannot be used to determine the effectiveness of particular interventions, which must be evaluated continually over both short and long periods of time. Some states give benchmark tests at several junctures during the school year. In some cases, these benchmark tests are designed to gauge how effectively students are progressing toward achieving the criteria measured on the high-stakes state test. However, some benchmark tests are not correlated in this manner. For example, in Texas, the TPRI benchmark test, however worthy it may be, does not necessarily indicate progress towards TAKS criteria, which were developed separately at a different point in time. Also, the TPRI is not administered frequently enough to be useful for evaluating an intervention in Tier 1. Therefore, for RTI evaluation, some educators may have to look beyond state benchmarks as well.

 

Options

 

One category of test that can be useful for a variety of purposes is the IRI or Informal Reading Inventory. There are many reading inventories and they are usually referred to by the name of their authors or coauthors, such as the Bader, the Burns, the Johns, the Silvaroli (one of our favorites) and the Stieglitz. These instruments are not usually as time-consuming to administer as the state CRT’s or even the benchmark tests, but they do require some practice and skill to administer. The IRI’s can identify strengths and weaknesses and compare their progress over time.                   

 

Another popular category of tests are the CBM’s or Curriculum Based Measurements. These tests are quick, and they are tied directly to particular reading skills that are part of the basic curriculum. They often involve reading passages taken from, or similar to those in, the curriculum. DIBELS (the Dynamic Indicators of Basic Literacy Skills), which is one of the most popular assessments in the country at the moment, is often referred to as a CBM. Basal reading series often provide their own progress monitoring tools that are sometimes considered to be CBM’s as well.

 

Between the CBM’s and the IRI’s, educators can assess student progress over relatively short periods of time. However, it is important to realize that the process of evaluating reading interventions is as much art as science.

 

Consider DIBELS. It can be administered quickly and frequently, and it is well organized. It offers simple instruments for data collection, and it is free online. For each of the early grades, there is a benchmark test to be administered three times a year and progress monitoring tests that can be given more frequently. The instruments provide information on different reading skills (initial sounds, letter naming, phoneme segmentation, oral reading, retelling and more). However, DIBELS appropriately refers to itself as measuring fluency. The measures tell little about accuracy, which is also extremely important.

You may remember that in last spring’s Food For Thought, we reported a study of two groups of students that were evaluated with DIBELS Nonsense Word Fluency Tests. One group had received phonics instruction in Stevenson and the other in a different method. The results showed that both groups made roughly the same amount of progress when looking at the simple raw score (number of phonemes correctly produced), BUT, when the data was further analyzed, the Stevenson group increased its accuracy greatly (from 76% to 96%) and the other group not at all (66% to 65%). (To review this article online, you can go to http://www.stevensonlearning.com/newsletter.asp and click on Spring 2007, Some Interesting Test Results from Ohio). If students read very accurately (above 90%), increasing speed helps comprehension. If they do not, increasing speed is virtually useless. Of course, there is a natural tendency to read slowly if you read poorly and read more quickly as you read more effectively, and DIBELS takes advantage of this tendency. Nonetheless, the educators using the test have to read between the numbers, or use multiple tests for multiple purposes. 

 

So how can you deal with RTI assessment if you are using the Stevenson Program? We have a few suggestions.

 

When possible, use our Mastery/Management Tests to assess or demonstrate progress. These are quick CBM tests. They do not yield a grade level equivalent, but they are good for monitoring progress, and progress is what reading intervention is all about. For some school systems, the Mastery/Management Tests books may not provide enough measurements for frequent RTI monitoring, so we have produced additional decoding tests that will soon be available free on our web site to people who have already purchased the Mastery/Management books. Look for the first of these tests for the Beginning Green and Basic Blue Levels on our web site under “Teaching Resources” after May 1, 2008.

We suggested using the Mastery/Management tests “when possible” because, if you are an American public school teacher in 2008, you will have to administer multiple test instruments, whether CRT’s or IRI’s or CBM’s. You may not feel that you have enough time to give more tests and still get any teaching done. We understand. Our Mastery/Management Tests help you demonstrate and assess progress, but they are not required.                   

 

Another simple, valuable step you can take for RTI purposes is to map out the progress you are likely to make with your students. We are not talking about complicated charts and graphs. We simply mean to look at where you are in the Stevenson Program, how long it took to get there and how long it might reasonably take to finish whatever level you are on. (If you feel uncertain about making this estimate, you can always call and ask for assistance, although a progress estimate may be more useful next fall.) Knowing where you are and where you will be in the course of the Stevenson Program makes it easier to anticipate how students will perform on other test instruments or in other reading materials.

 

As most of you know, the Stevenson Program teaches long vowel words before short vowel words, a sequence that is the reverse of that assumed in most tests. However, students still have to read many different kinds of vowel patterns - long ones, short ones, diphthongs and more - in order to succeed in the high-stakes state tests. Everyone does not have to take the same path at the same time in order to reach the same place. Mapping out progress will show you when students will reach key milestones in the program, and we are going to produce some free information this summer about such milestones.

 

Keep an eye on our web site for more information about assessments and how Stevenson relates to various reading measurements. After the additional free decoding tests for the Beginning Green and Basic Blue level are posted on the web site in May, we will publish Mastery Management Test books for the Lonely Vowels in June (see page 4) and then post additional Lonely Vowel Decoding Tests in July. In August we will post more information on grade levels and readability at different junctures in the program, and we will follow through on that subject in the fall newsletter. In the meantime, if you need some help with your RTI plans, do not hesitate to give us a call.