Tammy-Lynne Moore
Michael F. Shaughnessy
Eastern New Mexico University
Portales, New Mexico 88130

Dr. Malbert SmithIII and Dr. A. Jackson Stenner founded Metametrics, a company focused on matching students with appropriate instructional materials and bringing meaning to measurement, in 1984.As a result of more than 15 years of research in reading comprehension, Metametrics launched the Lexile Framework for Reading.

Dr. Smith has served on the Computer Advisory Board for Inc. Magazine and the North Carolina Council for Economic Development. In 1997, North Carolina Governor Jim Hunt awarded Dr. Smith the prestigious Order of the Longleaf Pine, the state's highest civilian honor. He is chairman emeritus of the Duke Children's Classic and The Teddy Bear Ball for Duke Children's Hospital. Dr. Smith has served on the boards of The Hill Center in Durham, N.C., the National Alopecia Arecata Foundation and Durham's Child Guidance Clinic. He is also a member of and presents to various state and national educational research and measurement associations.

Dr. Smith received his Ph.D. in educational psychology from the University of North Carolina at Chapel Hill and his undergraduate degree in psychology from Duke University. He has published and presented numerous papers in the field of educational assessment and measurement, and has taught graduate seminars at Duke University and the University of North Carolina at Chapel Hill.

All biographical information is from the Metametrics, Inc. web site: www.metametricsinc.com/desktopdefault.aspx?view=mm&tabindex=6&tabid=66

In this interview, Dr. Smith responds to question about the Lexile Framework, reading and the teaching of reading.

1. How can teachers implement The Lexile Framework for Reading into their classrooms – can it be used with already existing reading programs, such as Accelerated Reader?

The Lexile Framework for Reading was developed by MetaMetrics, Inc. to provide educators with a tool to match readers to text. Reader ability and text readability are measured in a common unit called a "Lexile." The comprehension that a particular reader will enjoy when reading a particular book, article or Web site is a function of the difference between the reader measure and the text measure. For example, a median sixth grader reading at 900L will have 75-percent comprehension of Harry Potter and the Goblet of Fire (880L). Educators can access Lexile measures for more than 100,000 books at www.Lexile.com and millions of periodical articles using resources such as ProQuest, EBSCO and Thomson Gale. Thus, teachers can differentiate reading assignments to ensure that every child can access the text. For example, a middle school social studies teacher could prepare article packets on the life and times of Dr. Martin Luther King, Jr. for three or four Lexile zones, thus, increasing the likelihood that all students are both reading and learning about the content.

The Lexile Framework provides a psychometric backbone for an increasingly large number of instructional programs and materials. This commonality makes it easier for different programs to work together. Similarly, the cognitive overhead required by educators who must interpret multiple measures of reading ability is reduced by the availability of a common metric like the Lexile Framework.

Finally, educators can use the fact that text and readers are measured in a common metric to adjust scaffolding. A student who is forecasted to have 50-percent comprehension of a science textbook (e.g., 750L reader; 1000L text) needs a great deal of scaffolding. Without scaffolding, such as audio-supported reading, the reader may choose not to engage the text. NAEP survey results suggest that far too many students are not reading the textbooks in middle school and high school.

In summary, there are dozens of instructional programs that have adopted the Lexile Framework as a means of matching readers and texts. The fact that the Lexile Framework is an open standard available to all publishers and free to all educators is largely responsible for its popularity.

2. What is the average amount of improvement students experience in their reading comprehension abilities? How is this measured? How does student success with Lexiles compare with other reading measures?

There are three key variables in the Lexile Framework: reader ability, text readability and comprehension. Reader ability is measured by any reading test that has been linked to the Lexile Framework, such as Scholastic Reading Inventory, Pearson PASeries, The Iowa Tests from Riverside Publishing, CTB/McGraw Hill's TerraNova, NWEA's Measures of Academic Progress, and Harcourt's Stanford Achievement Test. Text readability is a measure of the vocabulary and syntax demands that text places on a reader. A computer program calculates the Lexile readability measure. Text readability is reported in the same Lexile metric as are reader measures. Finally, comprehension is the rate at which a particular reader reading a particular text makes new meaning. Comprehension rate is modeled as a difference between reader ability and text readability. Thus, comprehension rate describes the interaction of a reader and text. What grows from kindergarten through early adulthood is reading ability. Comprehension rate as a concept only makes sense in reference to a particular text or text readability, as illustrated in our Harry Potter example. Educators manipulate comprehension by choosing text that results in a targeted comprehension rate (consistent with the scaffolding the educator has in place) for that particular reading assignment.

Growth rates for reader ability are remarkably similar in shape to those we observe for height and weight. Growth is faster in the elementary grades than in middle school, and high school growth rates are slower than in middle school. 200L of growth is what is typically observed in third graders and also what is observed in the four years spanning grades 9–12.

3. Many emergent literacy students receive little exposure to expository text and thus, as they become older, they struggle with comprehension of expository materials from textbooks to journal articles. What types of writing styles does the Lexile Framework include in its reading selections?

The Lexile Framework is restricted in its application to prose, including expository and narrative texts. Lexile readability measures are available for more than 100,000 titles (mostly narrative) and many millions of periodical articles (mostly expository). We have conducted several research studies that confirm the predictions the Lexile Framework makes about comprehension rates working for all prose. Thus, knowing a text readability and a reader ability is all that is needed to forecast comprehension rate; it is not necessary to also know whether the text is narrative or expository. That said, it is true that expository texts found in K–12 classrooms tend to have higher text readabilities than narrative texts. Similarly, narrative adult reading (Grisham, King, Rice, etc.) has lower text readabilities than expository adult reading (newspapers, periodicals, Web sites, etc.). As text type shifts from narrative to expository with the transition from "learning to read" to "reading to learn," the text readabilities increase. Concurrently, upper elementary and middle school teachers may not be as sensitive to the value of targeted reading as are elementary grade teachers. Thus, the text demand naturally increases as "one-textbook-fits-all" becomes more common place in the upper grades.

4. The Accelerated Reading program scores expository texts lower than most other text styles, making it a less-appealing choice for children who are trying to obtain their reading goals. Does The Lexile Framework for Reading take into consideration the type of text when determining text comprehension levels?

As noted earlier, it is not necessary to condition a forecasted comprehension rate (based on the difference between a reader measure and a text measure) on the type of prose to be read. An 880L reader reading a Readers Digest article on Islam (880L) is forecasted to have 75-percent comprehension, which is the same forecast that would be made for Harry Potter and the Goblet of Fire (880L). Whether the text selection is fiction or non-fiction is inconsequential to the forecasted comprehension rate. However, it is likely that much less attention to targeting (matching reader and text) is accorded expository text both because upper level teachers target less and because the text "diet" shifts to an emphasis on expository text with increasing grade level; lack of targeting and shifting text type conspire to increase the text demands on the reader. Note that this increasing text demand is not an artifact of readability formulas—whether Accelerated Reader's or some other—working better with narrative than expository texts.

The Accelerated Reader program uses a "grade equivalent-based" scale for text measurement and reader measurement. Unfortunately, a one-grade change in elementary school is not equal to a one-grade change in high school. It is like having a yardstick with different sized inches along the scale. This feature of the system could contribute to an artifactual misalignment of expository and narrative texts.

5. How do parents know what their child's Lexile measure is and how do parents know that the books their child is reading are at the appropriate level?

Lexiles are used in some fashion in every state, although not all students in the state are measured. States, such as California, Texas, North Carolina and Georgia, have adopted the Lexile Framework as part of their state-wide assessment (a complete list is available at www.Lexile.com). Currently, more than 20 million students receive a Lexile measure at school from a norm-referenced assessment, criterion-referenced assessment, formative assessment or reading program. Most states that have elected to report students' reading scores as Lexiles include the measures on the student report that is sent home to parents. In turn, parents can use a child's Lexile measure to find books, articles and other materials whose reading demand matches the reading ability of the child. One of the more popular resources is the Lexile Book Database at www.Lexile.com. This free database includes more than 100,000 fiction and non-fiction books searchable by such criteria as title, author, ISBN, keyword and Lexile range. Resources, such as the Lexile Book Database, allow parents to take a more active role in a child's literacy development by finding—and helping the child to find—materials that match his or her reading ability and interests.

6. How can parents make the most use of The Lexile Framework for Reading system at home?

One of the key benefits of Lexiles is the fact that the measures make student test scores actionable for parents.Parents do not know how to take action on a stanine, NCE score or a percentile rank. However, a Lexile measure provides greater insight into the child's actual reading ability. If a parent knows the child's Lexile measure, he or she can actively use that measure to connect the child to ability-appropriate reading materials. For example, the Lexile Book Database is particularly helpful in building reading lists for the summer months away from school when research has proven that students' reading skills diminish the most. Parents can also take the child's Lexile measure to the local library where, using library automation software that has incorporated Lexile measures, a librarian can locate materials that match the child's reading ability and interests.

7. How does the Lexile Framework measure reader ability and text difficulty?

As discussed earlier, a Lexile measure for a reader can be determined using one of the many reading assessments or programs that is linked to the Lexile Framework. A list of these reading measurement tools is available at www.Lexile.com.

A Lexile measure for text is based on two well-established predictors of how difficult a text is to comprehend: semantic difficulty (word frequency) and syntactic complexity (sentence length). If there are words that readers do not encounter very often, the likelihood that those readers will know the words' meanings is lower than with common words encountered every day. The longer the sentence, the greater the effort required to read the sentence all at once, break its structure down mentally, and understand the concepts being communicated. In order to determine the Lexile measure of a book or article, the text is split into 125-word slices, which is the typical length of a long paragraph. Each slice is then compared to the nearly 600-million word Lexile corpus—taken from a variety of sources and genres—and the words in each sentence are counted. These calculations are put into the Lexile equation. Each slice's resulting Lexile measure is then applied to the Rasch psychometric model to determine the Lexile measure for the entire text.

8. How is Lexile unique from other measures?

Readability formulas have been around for at least 50 years. What makes the Lexile Framework unique, and what has led to its widespread adoption, is the fact that the framework measures the reader and the reading material on the same scale. When MetaMetrics developed the Lexile Framework, our goal was to develop an 'actionable' measure. Educators and parents can use a student's Lexile measure to 'target' reading, by connecting him or her to materials that match his or her reading ability. Today, Lexiles are recognized as the most widely adopted reading measure.

9. What other supplemental materials are available with the Lexile Framework?

In addition to the Lexile Book Database, educators and parents can use the Lexile map (available at www.Lexile.com) to help match readers to ability-appropriate reading materials. The Lexile map provides a graphic representation of books at their Lexile level. Many Departments of Education that have adopted the Lexile Framework in support of their state-wide literacy initiatives have also elected to customize the Lexile map with titles from their curriculums. We have also worked with these agencies to make Lexile information and resources available on their Web sites. Additionally, we offer a free version of the Lexile Analyzer on our Web site, which allows educators and parents to analyze a piece of text and generate a Lexile measure.

For educators, we offer Lexile professional development. Every year, school administrators, teachers, librarians and other educational professionals attend our professional development workshops to learn how they can use Lexiles to inform instruction and improve achievement for all students.

10. Explain briefly how partners can license Lexile Framework materials.

MetaMetrics offers a broad range of services and technologies that enable our partners to utilize the Lexile Framework. For testing partners, we can design and develop custom assessments and scoring applications, as well as link their existing assessments to the Lexile Framework. To date, all major standardized tests have been linked to the Lexile scale.

For publishing and technology partners, we license various Lexile measurement technologies, and provide training and technical support to ensure the successful integration of these applications with their products and processes. We also offer various measurement service options to produce "certified" Lexile measures for published materials, or to support the development of new content using Lexiles. These certified trade titles are available through the online Lexile Book Database.

11. Has there been a lot of research on your materials?

The Lexile Framework for Reading is the result of more than 15 years of research by MetaMetrics psychometricians and faculty from the University of North Carolina at Chapel Hill and the University of Chicago. The development work was initially funded through a series of grants by the National Institute of Child Health and Human Development, part of the National Institutes of Health. These grants supported the research on reading and psychometric theory that culminated in the Lexile Framework. Our nationally known team of psychometricians continues our pioneering work in Lexiles today.

The Lexile Framework is also used as an outcome variable by other researchers to measure reading ability, including David Francis (University of Houston) and James Kim (University of California, Irvine), and companies such as Scholastic for Read 180. Additionally, Lexiles are used as an outcome measure for some state-level Reading First programs (Oregon).

12. What about Sustained Silent Reading? Do you feel this helps or hinders reading improvement?

The work of Tim Shanahan (University of Illinois at Chicago) and James Kim has shown that reading fiction that is well below your reading ability (900L reader; 600L text) is not going to improve your reading ability. However, reading targeted fiction (900L reader; 900L text) in which the reader has high interest does positively effect reader growth trajectories. The lesson is that the "treatment recipe" is more important than what you call it. Daily reading of high-interest, well-targeted text is a "treatment recipe" well supported by research. Ignoring targeting leaves out a vital ingredient in a recipe designed to improve reading ability. If the treatment outcomes are "love of reading" or "feeding the reading habit," then recreational reading, well below target, may be useful. John Grisham's novels are typically around 800L, which is perfect for adult readers that want to "get lost" in the story. In contrast, Charles Frazier's Cold Mountain (1210L) is a very different reading experience for most adult readers. MetaMetrics would love to see a randomized field trial of sustained silent targeted reading. The objective would be one million targeted words read per year for every 5–12 grade student.