Setting and Sample Population
The sample population for this study is four third grade students in a general education class in an urban district elementary school. There are 19 students in the class. The four students read at a grade below their grade level. They are non-fluent readers and have consistently performed below par in assessments in reading. Student A is a 9 year old male Hispanic who reads at the first grade level. Student B is a 9 year old African American male who presently reads at the second grade level. Student C is a 9 year Hispanic female who reads at the second grade level. Student D is a biracial (Hispanic and African American) who reads at the second grade level. All of these students receive free lunch and have satisfactory attendance. They do not have an IEP and none of them are English Language learners even though the Hispanic students are bilingual. These four students were chosen because it was noticed that even though they were just reading at one level below their reading level, they exhibit a great lack in comprehension skills. Their reading could also be described as non-fluent. Their reading is slow, choppy, and uninspired. There are also a lot of miscues, insertions, omissions and repetitions of known sight words. After readings like this they fail to answer or answer satisfactorily comprehension questions based on the readings. Samuels (1974), believes this happens because too much of their cognitive resources are used up in trying to read the words that there is little brain capacity left to comprehend the text. The trend continues and manifests itself in their poor performance on language arts tests and reading benchmarks. There was concern because the reading employed by the district focuses on teaching comprehension strategies such as main idea, author's purpose, making inferences, and cause and effect relationships. In whole group and small group where there is a lot of shared reading, such as modeling of fluent reading by the teacher, choral reading by class, or reading while accompanying the teacher, these students showed no real deficit in answering comprehension questions from the reading both orally and in writing. The problem arises when they have to do independent work. The decision was therefore taken to assess their level of fluency to see if it was affecting their level of comprehension as purported by the review of the literature. These students were also chosen because they were nearer the point where they could benefit from instruction that could help them score proficient on the state test and help the school meet the Annual Yearly Progress (AYP). A sample population of this size is adequate since they will be given intervention during the regular literacy block. A larger sample would prove time consuming.

The Elementary school where these students attend is in a low socioeconomic urban school district which is characterized by violence and illegal activities. The school serves PK-6, with a student population of 570. They school has been struggling academically for many years. It has not met AYP in many years. Last year the school received the School Improvement Grant (SIG) to increase educational output. As a result new personnel were hired to fulfill the mandate. The school day was also extended by one and a half hours. Plus there are other late programs that cater to underachieving students. The school has a regular literacy coach and a reading coach supplied by the owners of the School's reading program, 100 Book Challenge (100 BC). This coach meets with teachers about once per month to analyze teachers' implementation of the program and offer feedback on instruction and also to model best practices in reading instruction.

At present, morale at the school is high. The school had shown improvements even though it did not achieve AYP. The results for last's year’s state test showed that the third graders had the highest gains in the whole school. Each year it is expected that 5% more student need to pass than the year before to show the needed amount of growth. This is especially important as the school jockeys to renew the SIG. For the renewal the panel will assess school wide improvement in student learning using results from the state test, district benchmarks such as Learnia, Scholastic Math Inventory (SMI), increased reading levels in 100 BC. Failure to show the necessary progress could result in the closure of the school, the removal of the principal, or the transfer of staff whose students did not show adequate growth. At this site Harcourt reading series basal and leveled reading materials (StoryTown) are utilized. There is also a distinct reading program, 100 BC. Social Studies and Science deviate from the regular curriculum and are called research labs. Children read trade books according to their interest and answer research questions. The finished product for each marking period is a student made book based on their topic of choice or interest. Social Studies and Science classes provide extra reading time. Reading is stressed at this school. The literacy block is two hours with 30 minutes of compulsory silent reading. All students are expected to read at least one hour per day. Most of this reading is done in school.

Data Collection Procedures

The purpose of this study is to see the extent to which fluency intervention will impact comprehension. The intervention proposed is repeated reading. This could be teacher mediated, peer mediated or self mediated. The National Reading Panel (NRP), (2000), identifies fluency as one of the five main components of excellent reading instruction. Fluent reading is described as fast, accurate, and expressive reading. . One of the best ways to improve fluency is through repeated reading… (Fountas & Pinnell, 2006). Research in this area will further highlight fluency as one of the essentials reading skills, and prove that it can increase comprehension.
At the beginning of the school year, all students in the sample group were leveled for appropriate reading instruction. Students were leveled at their independent level. At this level reading should be easy, fast and fun. Students took part in the leveling by choosing the leveled book that they thought they could read as per the three criteria listed above: easy, fast, and fun. After they have decided on their independent level they were crossed referenced by the teacher using a norm referenced test. Readers at each level had certain prerequisite skills. The students are assessed using a cold read text at the independent level that they decided. Failure to meet the requirements at that level will see them drop to lower levels. If they are too advanced for one level they will advance to a higher level. The students instructional reading levels are assessed using the Developmental Reading Assessment (DRA) benchmark test or putting them a level higher on the 100 BC level. Knowing the students reading level is important to provide individualized and appropriate instruction to take them to the next level and above. It also provids important information on what they should know and to give further insight as to what account for their non-fluent reading practices. In the DRA test students respond to questions posed by the teacher to ascertain their reading engagement. Students are asked to explain how they choose books, the types of books they like to read and also to give a brief summary of one of their favorite books, who reads with you at home? They will also be asked questions such as: Would you rather listen to story or read a story to someone and why? Do you like to read alone, with a partner or in a group and why? Whom do you read with at home? These questions give clues as to the amount of reading students’ do, their knowledge of books and authors, their ability to summarize and give important details, the variety of reading materials read, highlight their interest and show their awareness as a reader. These components will be important as the intervention is undertaken. For example if students are in interested in certain topics, passages will be selected with those in mind. The student’s reading log, which is updated on a daily basis, will also be used to check for reading engagement.
The students will respond to a teacher made questionnaire to assess their attitudes towards reading. The students answered questions with a rating scale of one to four with four being the highest. The scores were totaled. The higher the score, the more positive the attitude towards reading, the lower the score the more negative the attitude toward reading. Students’ attitude toward reading speaks to their motivation to read. This factor should be ruled out to provide the appropriate intervention.
Students will be observed on a daily basis in whole group, small groups, and one on one conference. Students reading behaviors will be observed and documented. Behaviors such as how comfortably they read, their level of fluency, the types of errors made, posture, and mannerisms. Oral and written responses to comprehension questions were noted, and relevant samples noted with teacher feedback.

Students will complete the pretest on an individual basis. The pretest will consist of a cold read of three independent level passage. The pretest will be administered during the 30 minutes silent reading period during literacy. The participants will called on the do the pretest one at a time. This is to prevent the other participants to hear the text and become familiar with it. Students will read for one minute.

While the students read a running record will be used to record the number of errors made using a chart similar to the ones used in the DRA. Students’ number of miscues will be documented, and the percent of accuracy and the number of words read per minute will be calculated. The teacher will also be observing for fluent reading, that is, speed, and expressiveness. Miscues include:
a. omission
b. insertion
c. skipped lines or words
d. mispronunciation plus
e.reversals
f.. failure to read a word after 3 seconds

The following will not be counted as errors
a. self correction of miscues b. hesitation

After the number of correct words per minute is recorded for each passage, the median score will provide an index of students' fluency development relative to the intervention. After reading each paragraph, students will answer five multiple choice questions on each of the passages based on the literal and inferential levels of the Bloom’s Taxonomy, to collect data for the comprehension component. Students’ response will be also be documented and a score of point given to each correct response. The teacher makes notes and shares findings the next day with students and informs students the areas they will need to improve. The teacher models the intervention and ensures students understand. For example one of the students had a problem with reading phrases correctly. Teacher teaches mini lesson and models using a phrased-cued text to recognize the natural pauses in reading. Rasinki, 1990, 1994 believes that students’ ability to identify phrases as units will enhance their comprehension.

Over the four week period students will meet on a daily basis to complete fluency building activities. They will reread passage increments of one minute. The repeated reading in the first week will be teacher mediated, since teacher is using herself as a model for the students. Each time they will be told specific reason why they are rereading- to improve expression, to increase speed, and for comprehension. Student’s progress will be observed and noted.
Over the next few weeks a similar procedure will be followed. This time the strategy will be peer mediated or self mediated on a more regular basis. Students will chart their word per minute growth using a simple bar graph. On alternative days students have to answer one or two comprehension questions on the topic. Students and the peer mediator will chart their own progress. This was to get them involved in the study and take responsibility for their own fluency development. Near the end of the fourth week, students were given a posttest. The post test took the same format as the pretest. Students were given three cold read independent level passages and then answer five multiple choice questions based on each paragraph.

Discussion of Action (Intervention)
Although there is some debate about the

This study seeks to analyze the impact of repeated reading on fluency development and comprehension.
Discussion questions: What were the relationships, if any, between fluency interventions and comprehension? What was the nature of students' performance on the pre and post test?
How did students perform in peer mediated repeated reading as opposed to when it is teacher mediated. What factors influenced the success of the fluency intervention? Dependent variable - fluency and comprehension
Independent variable: repeated reading
Discussion of Action (Intervention)