ECELL Descriptive Reading Research Study
Cheryl Brown-Kovacic, Ph.D.
Introduction
As part of the Equal Chance, Equal Learning through Literacy (ECELL) Project, the Center for the Education and Study of Diverse Populations (CESDP) undertook a descriptive reading study in eight Corrective Action middle schools participating in the project. Schools were eligible for the ECELL project based on their high Native American populations. Only Native American students in the entering grade at each school were included in the study, which was designed to explore the foundational skills that might be contributing to low performance on New Mexico accountability assessments in reading comprehension. Navajo and Pueblo Indian students participated in the study.
In the fall of 2003, 314 sixth and 422 seventh grade students were tested with one subtest of the Woodcock Reading Mastery Tests – Revised (WRMT– R) and two subtests of the Test of Word Reading Efficiency (TOWRE). The subtest from the WRMT– R was Test 6 - Passage Comprehension. The subtests from the TOWRE were Subtest 1 – Sight Word Efficiency and Subtest II – Phonemic Decoding Efficiency. The TOWRE subtests are timed assessments, providing a measure of fluency.
These assessments offer some initial insight into the skills that students have developed in word attack, word identification, fluency, and reading comprehension strategies, which reflect the underlying reasons for low performance in reading. Since state reading standards at middle school focus on comprehension, there is an assumption that middle school students have already mastered foundational skills. If students have not done so, adjustments may need to be made in the middle school curriculum to meet the needs of such students.
The complexity of word reading ability reflects students’ ability to apply a variety of strategies to identify words while reading text. In general, five different strategies have been recognized through research.
1. Analyzing and blending together the individual phonemes, frequently referred to as phonemic decoding or “sounding out” the word
2. Noticing and bringing together familiar spelling patterns (orthographic processing) with more than one letter, often considered to be an advanced form of phonemic decoding
3. Recognizing printed words as whole units or reading by “sight”
4. Making analogies to words that are already known
5. Using clues from context to guess a word’s identity
These strategies contribute to fluent reading and are used by good readers to varying degrees, depending on the familiarity of the words encountered in text (Torgesen, Wagner, & Rashotte, 1999).
The intent of the study was to identify the potential underlying aspects of reading, which may contribute to individual student difficulties. Three areas were hypothesized as possible points of difficulty: phonemic decoding (the ability to analyze and blend together sounds), fluency (the skill to read easily, smoothly, and expressively), and comprehension (the ability to read for meaning and understanding). The following analysis will focus on these areas of reading.
Additional diagnostic testing could provide more extensive analysis relating to the sources of individual reading difficulty, while use of individual classroom assessments, such as anecdotal records, reading checklists, running records, and changes in text reading levels over time could contribute to greater depth of understanding regarding individual needs. Consideration should also be given to student interests, cultural background, relevancy, and attitudes in selecting engaging literacy experiences. Although some of these students are enrolled in Special Education programs, that information is not reflected in this study.
Scores
The resulting scores from the testing at eight Corrective Action middle schools with high Native American populations are summarized in the following table. Normal Curve Equivalent (NCE) scores will be used since this scale allows some comparison between different tests. Any student achieving a NCE score of at least 45 was considered to have adequate skills in the three areas tested. A NCE of 45 places the student at approximately grade level reading, with an error range of +.5 or -.5. A NCE of 45 translates to approximately the 40th percentile rank. Students achieving less than a NCE of 45 are considered to be at risk in reading and are further divided into two groups, those who are low–performing and those who are approaching adequate performance. A student who obtained a score of 35 NCEs or less is considered low–performing. This student will probably fall in the bottom quartile and is likely to be behind grade level by at least two years in performance. A student scoring at the 0 to 25 NCE level demonstrates an extensive lack of reading skills. His/her performance on state assessments is likely to be in the lowest quartile or proficiency category, as well as at chance or “guess” level on multiple-choice items in reading.
Results
The first task in analysis was to identify the performance levels of these Native American students in reading comprehension. Then the relationship between comprehension performance, word reading and fluency could be identified for those students demonstrating low performance in comprehension. Students could potentially need instruction in word attack, fluency, or comprehension strategies.
NCE Scores in Reading Comprehension by Grade
(WRTM-R Passage Comprehension Subtest)
Number (Percent) of Students
Grade / N / Low Performing / Total Low Performing / Approaching Adequate / Adequate0-25 NCE / 26-35 NCE / 0-35
NCE / 36-45
NCE / 46 NCE or Above
6 / 314 / 91 (29%) / 112
(36%) / 203
(65%) / 84
(27%) / 27
(9%)
7 / 421 / 153
(36%) / 148
(35%) / 311
(74%) / 80
(19%) / 40
(10%)
Two-thirds of the sixth-grade students and three-fourths of the seventh grade students fall into the low performing category on a reading comprehension assessment that controls for prior knowledge and vocabulary. Only about ten percent of students at either grade are performing at the level expected.
NCE Scores in Sight Word Efficiency
(TOWRE Subtest 1)
Number (Percent) of Students
Grade / N / Low Performing / Total Low Performing / Approaching Adequate / Adequate0-25 NCE / 26-35 NCE / 0-35
NCE / 36-45
NCE / 46 NCE or Above
6 / 312 / 58
(19%) / 54
(17%) / 112
(36%) / 113
(36%) / 87
(28%)
7 / 422 / 41
(10%) / 102
(24%) / 143
(34%) / 163
(39%) / 116
(27%)
About one-third of these students, at both grade levels, demonstrate low performance in sight word efficiency, while approximately one-fourth are performing at expected levels.
NCE Scores in Phonemic Decoding
(TOWRE Subtest 2)
Number/Percent of Students
Grade / N / Low Performing / Total Low Performing / Approaching Adequate / Adequate0-25 NCE / 26-35 NCE / 0-35
NCE / 36-45
NCE / 46 NCE or Above
6 / 311 / 117
(38%) / 63
(20%) / 180
(58%) / 49
(16%) / 82
(26%)
7 / 421 / 131
(31%) / 78
(18%) / 209
(49%) / 83
(20%) / 129
(31%)
On the Phonemic Decoding Subtest 50% to 60% of the students demonstrate low performance, approaching twice the percentage that were in the low performing category on Sight Word Efficiency. However, similar percents of students performed at the adequate level on these two subtests.
Interpretation of Scores
On these assessments high percentages of students demonstrate difficulty in reading comprehension, with more seventh grade students than sixth grade students falling into the low performing category. Of the students demonstrating performance less than adequate on comprehension, 77% in grade 6 and 74% in grade 7 also have less than adequate performance in phonemic decoding. Strong research findings suggest that good phonemic decoding skills are necessary for the growth of a rich sight word vocabulary and that a rich sight word vocabulary is important for fluent reading and good reading comprehension. Thus, the TOWRE subtests focus on these areas. The Sight Word Efficiency test provides a measure of vocabulary for fluent reading, while the Phonemic Decoding Efficiency subtest focuses on phonemic decoding skills necessary, but not sufficient, for the development of a strong sight word vocabulary (Torgesen, J., Wagner, R., & Rashotte, C., 1999). Of the 732 students tested, 389 (53%) have very poor skills in decoding words, scoring less than 35 NCEs. If a student is not able to decode words in a timely manner, the student is likely to also have difficulty with fluency, since both the accuracy and the flow of reading will be impacted by the lack of ability to decode words with which the student is unfamiliar. About 1/3 (34.5%) of the students show serious difficulty in reading sight words in a fluent manner. These students require direct instruction in decoding and word attack skills.
There are 107 (34%) students at sixth grade and 127 (30%) at seventh grade who are low performing in both phonemic decoding and sight word efficiency. They are at greatest risk. This pattern is an indication that explicit and intensive instruction in word-level skills is needed. These students will require instruction that focuses on stimulating the growth of phonemic awareness, teaching grapheme-phoneme correspondences and blending skills, and providing extra opportunities for guided reading experiences so that a teacher can directly assist with phonemic decoding skills while the student is in the process of reading text and building fluency in word recognition. Carefully supported writing experiences that are monitored can also help these students make progress. They will also need extensive opportunities to practice their newly acquired skills (Torgesen, et.al.,1999).
The TOWRE manual suggests that the scores on the two subtests should be similar to one another for a student progressing at grade level equivalence. Cases that do not follow this pattern should be identified by individual student and considered an indication of a specific reading disability. In examining cases within this population that do not have similar scores on the TOWRE subtests, it appears that some students may be relying on their sight word vocabulary as the basis for their reading ability. As students grow older, a reliance on sight words will become more and more difficult. It is easier for students in early grades at the elementary level to maintain normal progress in reading by simply memorizing words than they can later on when the range of words in grade-level text is greatly expanded. If a student in middle school is able to maintain normal growth in word reading ability while utilizing weak phonemic decoding skills, then the student has probably learned a strategy for expanding sight vocabulary that doesn’t depend heavily on phonemic decoding ability (Torgesen, et al., 1999). The 160 (51%) of the sixth grade and 207 (49%) seventh grade students, who scored at least 5 NCEs higher on the Sight Word subtest than on the Phonemic Decoding Efficiency subtest, should have a good chance of improving their vocabulary given decoding strategies. Both grades have approximately 1½ times as many students demonstrating severe difficulties with phonemic decoding as with sight word efficiency.
The Passage Comprehension test offers a measure of basic reading comprehension while controlling for the effects of prior knowledge and vocabulary. It measures the student’s ability to study a short passage and identify a key word missing from the passage. It is a modified cloze procedure in which the students must use a variety of comprehension and vocabulary skills. It is unlikely that students can offer an acceptable response without a complete reading and understanding of the entire item. Most items are drawn from a variety of real-world reading materials. A variety of short passages and longer passages, typically encountered in textbooks and newspapers, as well as household and business documents, are used. The results of this WRTM-R subtest can be supplemented by asking the student to read several pages from a textbook relevant to the instructional setting and respond to inferential questions or to paraphrase the content of the materials (Woodcock, R., 1998). It is likely that the 514 students who are low performing in passage comprehension struggle with fluency or some aspect of fluency. In fact, the data clearly shows that 112 (36%) at grade 6 and 143 (34%) at grade 7 of these students struggle with fluency, as these students were low performing on either the Sight Word Efficiency Subtest or both of the TOWRE subtests.
The 164 students in the Approaching Adequate performance in passage comprehension are likely to require assistance in both building vocabulary and strategies for comprehension, which will assist with selecting the main idea, summarizing, finding key points, making inferences, etc. The 43 grade 6 and 57 grade 7 students who struggle in comprehension, but have adequate scores in word attack and sight words, are likely to need vocabulary building, practice, and comprehension strategies. These students have the building blocks for reading in place, but need extensive practice.
There were 64 (21%) grade 6 and 86 (20%) grade 7 students whose Phonemic Decoding Efficiency score was higher than their sight word reading ability by at least 5 NCEs. This often points to problems in vocabulary and/or simple lack of practice. Increasing the amount of reading both in and out of school may be helpful. “Coaching” the students on faster and more efficient strategies for recognizing words as whole units may also be of assistance (Torgesen, et al., 1999). On the other hand, there were 160 (51%) of the sixth grade and 207 (49%) seventh grade students whose sight word reading was higher than their word attack skills by 5 NCEs, indicating problems with decoding and fluency, hence the problems with comprehension. Of the 143 students who were adequate on both sight words and word attack, only 29% were also adequate in comprehension. These students are likely to require direct instruction in strategies for comprehension, extensive practice at their independent reading level, and vocabulary building.