Current information about this program can be found under the Education Research program.

Carnegie Mellon University
Principal Investigator: James Callan
Reader-specific lexical practice for improved reading comprehension

Texts that closely reflect an individual reader's vocabulary and comprehension level offer the potential to provide stronger support for the development of the reader's reading comprehension and vocabulary growth. But classroom teachers seldom have the time to individualize the selection of reading materials for each of the students in their classes. The purpose of this project is to develop a computerized web-based search engine to support an instructional strategy designed to increase students' vocabulary knowledge and improve reading comprehension by enabling students to select passages from across the Internet on topics of interest which also meet specific standards of reading difficulty and content of unfamiliar vocabulary. As students use the search engine, it builds a model of the individual student's level of acquisition and fluency for each word, producing an individualized framework for selecting reading materials that will strengthen that student's reading comprehension.

The researchers are carrying out three studies in the process of developing the search engine, two of which involve university students and one of which involves students in the third through sixth grades. The first study involves adults reading texts in three different topic areas (business, arts and leisure, and sports) and three levels of different percentages of unfamiliar words, to measure the effects of these differences on reading times, comprehension, and readers' ability to infer word meanings, assessments of difficulty, and degree of interest. The second study involves students reading texts chosen according to the search engine's criteria for gauging their reading difficulty and taking into account the individual readers' degree of familiarity with the vocabulary, to evaluate the feasibility of individualized text selection using materials drawn from the web. In the third study, the researchers are having students in grades three through six use the search engine to select and practice reading with materials selected from the web that reflect increasing levels of difficulty in the vocabulary words included in the texts, and measuring the development of their reading comprehension.

Florida State University
Principal Investigator: Richard Wagner
Origins of individual and developmental differences in reading comprehension

Current assessment measures of students' reading comprehension are good for the purpose of evaluating how well students read compared to their classmates and grade level, but they are not especially well-suited to explain why students are performing at a given level, and what kind of instruction might serve to improve it. The purpose of the project is to develop knowledge to guide the development of measures of reading comprehension designed to identify the source of problems in reading comprehension, and guide instructional practices.

The researchers are carrying out five studies of the sources of individual and developmental differences in reading comprehension with students in the second and fourth grades. The first two studies are designed to test different models of the possible sources of individual and developmental differences in students' reading comprehension. The researchers are collecting student performance data on reading tests over a three-year period of time and using statistical analyses to investigate the relationships among such characteristics as students' vocabulary, working memory of what they're just read or heard, awareness of how words are formed, and ability to 'decode' (i.e., to recognize and read words by translating the letters into speech sounds to determine the word's pronunciation and meaning). In the third and fourth studies, the researchers are carrying out two experiments in which the instruction of different groups of students focuses on the improvement of one or more of these student characteristics of decoding and vocabulary, in order to study how training in one characteristic leads to changes in the other characteristics and in overall reading comprehension. In the fifth study, the researchers are examining different kinds of measures of reading comprehension by comparing traditional published reading comprehension tests with tests that more closely resemble the material that students normally read in class, and with a new test that focuses on students' comprehension of specific passages of reading material, to explore the tests' potential for identifying the sources and solutions to problems in student progress in reading comprehension.

Teachers College, Columbia University
Principal Investigator: Joanna Williams
Teaching elementary students to comprehend expository text

The ability to read materials that are primarily intended to convey information is key to success in school, in the workplace, and in the community. Research suggests that students' lack of exposure to and instruction in such expository texts slows their progress in reading as they progress through the fourth grade and beyond, and that a lack of understanding of the structures of such texts frequently contributes to their difficulties. The purpose of this project is to develop and evaluate instruction designed to improve students' knowledge and use of expository text structures before they reach fourth grade, so that they are better prepared to comprehend the expository text in the content textbooks they will encounter as they progress through school.

The researcher is carrying out three studies designed to examine the effects of integrating explicit instruction about text structures into regular teaching, in classrooms of second grade students of mixed abilities in urban schools with large proportions of minority students. The first study focuses on compare/contrast text structures, using content drawn from the social studies. Classrooms of students are randomly assigned to receive instruction according to one of three different approaches: 1) instruction in which explicit instruction about the compare/contrast text structure is embedded into the social studies content; 2) instruction using the same reading materials as in the first approach, except without the emphasis on the compare/contrast text structure; or, 3) instruction unaltered from the teachers' normal educational practice. The second study uses the same design, except that it focuses on instruction in cause/effect text structures rather than compare/contrast structures. The third study also uses the same design, but incorporates instruction in both compare/contrast and cause/effect text structures. The researcher is comparing student performance from the various groups of students using a variety of measures designed to assess their learning of both the content and their understanding of the text structures, and then using the data to further revise and improve the most effective instructional approaches.

The Pennsylvania State University
Principal Investigator: Bonnie Meyer
Intelligent tutoring using the structure strategy to improve reading comprehension of middle school students

Learning to read is fundamental to learning and life-time success. Some students have difficulties understanding what they read, not because they can't read at all, but because of how they read. The purpose of this project is to develop and test the efficacy of various features of a web-based intelligent tutor designed to teach middle school students who are struggling readers how to use the structure of the text to help them understand what they are reading. Students are taught to recognize common organizational structures used in expository texts (such as comparison, problem/solution, cause/effect, sequence, description, or listing) and how to use those structures to help them identify the main ideas in expository texts.

The researchers are carrying out three studies of various features of the web-based tutor, with fifth and seventh grade students from four diverse school districts serving suburban or rural communities that include minority and low-income families. The researchers are developing and pilot testing the web-based tutor in the project's first year. Following that, in the first study students are randomly assigned to use the tutor programmed to provide either simple or more advanced responses to their input, and either with or without the option of being able to choose what passages they practice their reading on. In the second study the researchers are interviewing students to discover how well they understand the structure strategy. In the third study, students are being randomly assigned either to a group whose use of the tutor is programmed to adjust to the individual student's learning progress, or to a group with a tutor that delivers the standard intelligent tutoring of the structure strategy.

University of Connecticut
Principal Investigator: Michael Coyne
Project VITAL: Vocabulary Intervention Targeting At-risk Learners

Research indicates that while vocabulary knowledge plays an important role in the reading process and reading comprehension, little is known about how to teach vocabulary effectively. Evidence also shows that a gap in vocabulary knowledge emerges between at-risk students and their peers before they enter kindergarten, and grows larger in the early grades. The goal of this project is to develop and evaluate strategies for teaching vocabulary to young children at risk of experiencing reading difficulties. The project researchers are developing, implementing, and evaluating strategies designed to improve the vocabulary knowledge of at-risk urban first graders.

In the first phase of this three phase project, the researchers are performing experiments to develop, field-test, and refine vocabulary instructional strategies with small groups of students. In the second phase, they are evaluating the efficacy of those vocabulary instructional strategies by comparing student learning in first grade classrooms randomly assigned to use either explicit vocabulary instruction inserted into storybook reading activities or traditional storybook reading activities. In the third phase, the researchers are repeating the second phase study in a wider range of diverse classroom settings. The research team is also training classroom teachers from urban schools serving a high percentage of students from diverse ethnic backgrounds to implement the vocabulary intervention strategies. The project is designed to test the feasibility of having classroom teachers implement and sustain the use of effective vocabulary instruction under real conditions.

University of Illinois
Principal Investigator: Richard Anderson
Improving comprehension and writing through reasoned argumentation

While learning how to use various argument strategies is an integral element of education across the curriculum, existing research has not yet identified the most effective ways of doing so. The purpose of this project is to evaluate Collaborative Reasoning, an education intervention using intellectually engaging classroom discussion to improve children's ability to comprehend, evaluate, and produce arguments. In Collaborative Reasoning, students take positions on issues in texts they read, provide reasons to support their positions, cite text information and background knowledge as evidence, challenge other children when they disagree, and respond to the challenges of others. The researchers are measuring the effects of student participation in this activity on learning.

The researchers are comparing the effects of student participation in classrooms where Collaborative Reasoning is introduced with classrooms continuing to carry out their normal educational practices. Pairs of fourth and fifth grade classrooms of students with similar backgrounds are being randomly assigned to have one classroom initiate Collaborative Reasoning while the other classroom continues as before. The study sample includes 32 fourth- and fifth-grade teachers and approximately 1,400 children attending rural and urban schools serving families with a variety of income levels and a diversity of racial and ethnic backgrounds. The researchers are measuring and analyzing student performance using a variety of outcome measures, including the number of satisfactory arguments, counter arguments, rebuttals, uses of text evidence, and the uses of argument stratagems in students' written essays; their responses to questions on a standardized reading test; a test of critical thinking ability; and, interviews in which students rate their own interest in reasoning and argument, and express a preference for either collaborative reasoning or conventional classroom discussion.

University of Illinois
Principal Investigator: Georgia Garcia
Instruction of reading comprehension: Cognitive strategies or cognitive engagement

Levels of student proficiency in reading comprehension in the United States remain low, and there is little evidence to suggest any increase is occurring in the use of effective instruction in reading comprehension. Research has established the efficacy of several approaches to teaching reading comprehension, but little is known about why these approaches work, and teachers find it difficult to actually implement effective approaches in their teaching. The purpose of this study is to develop and evaluate teacher-implemented strategies for reading comprehension instruction based on two established general approaches, and then to combine the positive features of the strategies of both approaches into a comprehensive model of instruction.

The researchers are carrying out this three phase project with second and fourth grade teachers and their classrooms in four school districts serving different populations of students, one of which includes a substantial portion of students whose first language is Spanish. In the first phase, the researchers are working with second and fourth grade teachers to develop and implement teaching strategies using two different approaches. In the first approach, called cognitive strategies, the emphasis is on teaching students to carry out specific operations with the material they are reading, such as questioning, clarifying, summarizing, or reacting to what they read. In the second approach, called cognitive engagement, the teacher uses questions to encourage students to think about various features of what they are reading, in order to strengthen their intellectual grasp of the material. In the second phase the researchers are formally evaluating the effects of teachers' use of these two approaches, by randomly assigning teachers in 2nd and 4th grade classrooms in participating schools into three groups of classrooms using cognitive strategies, cognitive engagement, or continuing to teach as they had before. In the third year, the researchers are using the results of the prior study to synthesize the two approaches into a more comprehensive model and then testing the effects of the synthesized approach in comparison with normal teaching practices.