Developing Fluent Readers

Because accuracy is a fundamental component of fluency, teachers who work with beginning readers must focus significant amounts of instructional time on basic word recognition and word analysis skills (Pikulski and Chard, 2005). To do this effectively, teachers should provide instruction that systematically presents daily opportunities for students to learn to read words accurately (Snow, Burns, and Griffith, 1998)-the important first step in becoming a skillful, proficient, and motivated reader. Pushing students to “read faster” too soon could cause some students to begin guessing or otherwise undermine their focus on reading carefully.

There is no guidance from empirical research about precisely when teachers should formally begin encouraging beginning readers to increase their speed, but teachers usually wait until about the middle of first grade. Fluency researchers Stahl and Kuhn (2002) recommend that students be given opportunities to re-read sentences and encouraged to make their reading “sound like talking” as soon as they are making good progress with basic decoding, demonstrating an understanding of the act of reading, and showing some degree of confidence-whether that happens in kindergarten or in first grade.

Teachers and parents should also frequently model fluent reading, demonstrating (and sometimes explicitly pointing out) how accurate reading can be done at a reasonable rate and with good phrasing, intonation, and expression. In the classroom, the teacher can read aloud from large-format books so the students can follow along.

What about students in grades two and higher who are making adequate progress with their reading? Three techniques can be used very frequently with a variety of texts to help maintain and develop students’ reading fluency: Choral reading, cloze reading, and partner reading. All of these procedures can be used with readers at any grade level, with small or large groups, and with fiction or content-heavy nonfiction materials. Two additional techniques can also be considered for use: Readers’ Theater and poetry readings. Let’s review each.

For choral reading, the teacher and students read aloud together, following the teacher’s pace-so students get the benefit of a model while they practice reading aloud. The teacher can stop at any time to ask questions, comment on the text, discuss a vocabulary term, or remind the class that she expects everyone to be reading. If choral reading is used with heterogeneously grouped students, it is possible that the lowest performing students may have difficulty keeping up with even a moderate pace. However, they can follow along, participating when they can, and still hear the text being read accurately and with good pacing and phrasing. Choral reading works best if the teacher directs all students-regardless of age or ability level-to use a marker or finger to follow along in the text as they read.

Cloze reading is similar to choral reading, except that the teacher does most of the oral reading while the students read along silently. Once or twice every few sentences, the teacher omits an important vocabulary or content word, not a simple sight word, and the students’ job is to read it aloud as a class. Notice that with cloze reading, as opposed to choral reading, students spend less time practicing oral reading. Therefore, cloze reading is best thought of as an alternative to Round Robin Reading.

Cloze reading allows teachers to cover text and keep students engaged while avoiding the pitfalls of subjecting the class to examples of poor reading and embarrassing struggling students. As with choral reading, it is likely that the lowest performing readers will be unable to keep up or to correctly read every omitted word, but they will not be singled out-and will be provided with examples of skillful reading.

Another method for improving fluency is to have students read aloud to a partner. This procedure works best when students are taught some techniques for giving feedback and managing their time, and when the partners have been selected by the teacher.

One technique for assigning partners is for teachers to first rank the students from the strongest reader in the class to the weakest (making judgments subjectively or from assessment data) and then consider whether there are students whose reading ability is so low that partner reading may be inappropriate. These students could meet with the teacher for more direct instruction or closely supported partner reading while the other students do independent partner reading.

The teacher then divides the remaining students in half, forming pairs such that the strongest reader is paired with a mid-level reader, and so on, ensuring that each pair has a slightly stronger reader, but that the difference in the students’ ability is not so large as to cause embarrassment or confusion.

At times, the stronger reader may be directed to read first, providing a model of fluent reading. Then the less fluent reader reads the same text aloud. The stronger student can help with word recognition and give feedback and encouragement to the less fluent partner. Another effective technique pairs students who read at the same level and asks them to re-read a story on which they have already received instruction from the teacher (Osborn and Lehr, 2004).

Readers’ Theater and poetry readings-both of which engage students in a reading performance-have become popular over the last few years. Much has been written about Readers’ Theater in particular, and about the apparent value of having students participate in dramatic readings (Rasinski, 2006). These kinds of activities provide students with an opportunity to read text that is enjoyable-and provides a clear incentive for students to read, and re-read, their assigned parts or poem.

However, while these techniques are motivating, teachers should not assume that either one could possibly provide as much practice for the whole class as choral or partner reading, much less anything close to the amount of instruction and practice necessary for struggling students to improve their fluency.

The research literature provides some clear directions on what to do with struggling readers: Interventions must combine the modeling, repeated reading, and feedback that research has demonstrated effective (Shaywitz, 2003). Several commercial programs have been developed, including Read Naturally (Ihnot, 1991), the Six Minute Solution (Adams and Brown, 2003), Quick Reads (Heibert, 2002), and the Great Leaps Reading program (Campbell, 1996). Each of these programs includes at least some of the instructional components that have been shown to improve students’ reading fluency and has its own approach to student engagement.

Unfortunately, research that directly compares the effectiveness of these various programs has yet to be done. In my own review of the available research, I’ve concluded that the strategy developed by Read Naturally makes the best use of the research base on fluency and has the strongest evidence of effectiveness as a fluency intervention. And, in using the Read Naturally strategy with students in many different grade levels, I’ve found that it engages them in the repeated reading they so desperately need.

However, I encourage readers to keep in mind that over the next several years, research may provide evidence of effectiveness on these other programs that equals or surpasses that of the Read Naturally strategy.

The Read Naturally (RN) strategy was developed by Candyce Ihnot, a Title I reading teacher from Minneapolis. Candyce developed and tested it in 1989 as part of her master’s thesis in special education. After finding that her approach was effective with struggling students in her school, Candyce and her husband, Tom Ihnot, developed a set of instructional materials that are commercially available from their company, Read Naturally, Inc.

To implement the RN strategy, students’ fluency levels (WCPM) are assessed to place students at an appropriate instructional level. The teacher then helps each student set a reasonable, achievable fluency goal (approximately 80-90 WCPM for primary students or older students reading at a primary level; from 90-120 WCPM for upper elementary students).

Instruction begins with an unpracticed, “cold reading” of a student-selected passage from the targeted level. Passages may range in length from approximately 100 words at the mid-first-grade level to 350 words at the sixth-grade level. As they read, students use a timer and keep track of the words they skip or stumble over (by lightly underlining the problem word). They then calculate their WCPM and graph this first, unpracticed WCPM score on a bar graph.

In step two, students practice reading this same passage three to four times along with a model to learn how to accurately pronounce all the words in the text. This step is not timed, and the students read the entire passage. The modeled reading can come from a recording or a person trained to read the passage at a rate that is comfortable for the student.

The key here is that a student does not just listen to the model, but actually reads aloud (softly) with the narrator’s voice, giving full attention to the text. Encouraging students to point to the text being read and informing them that they will be responsible for answering a set of comprehension questions after completing all the steps in the strategy helps students stay focused.

Once students feel comfortable with the text, they begin step three in which they read the text independently, again aloud, but softly. Students set a timer for one minute and read the text several times until they are comfortably reaching their predetermined goal level-and are ready to be checked by the teacher. Having some kind of silent signal for the teacher such as a flag or colored card at the students’ desk can help manage this step. Students keep practicing the passage until the teacher can meet with them because this maximizes their engaged practice time-a key to improved skills in low-performing students (Brophy, 1988).

In the final step, the student reads for the teacher, who then calculates the WCPM score. The student “passes” if four criteria are met:

1. the WCPM score meets or exceeds the predetermined goal;
2. three or fewer errors are made;
3. the student reads the passage with correct phrasing and attention to punctuation; and,
4. the student can correctly answer a few comprehension questions.

When students do not pass, they continue practicing this same text. When they do pass, they graph their new score onto the same bar with their initial, unpracticed score, using a different colored pencil or marker. This graph gives tangible evidence to the students that they are improving-and keeps motivation high by showing them that their own effort makes the difference. For an external check on progress, the teacher should also periodically assess students’ performance on an unpracticed passage by following the progress monitoring procedure described in the article “Screening, Diagnosing, and Progress Monitoring: The Details.”

Students repeat these steps until they complete 10-12 passages of equivalent difficulty. At that point the student and teacher collaboratively examine the data on the student’s graph to decide what step to take next. If the student is making steady progress in the current level, but is not yet approaching his goal level on the first, unpracticed reading, he should stay in that same level for another 10-12 passages.

If the student’s first unpracticed readings are occasionally meeting or approaching the goal, the teacher and student may decide to move the student up to the next level of difficulty with the same goal, or stay in the current level of difficulty and raise the “pass” goal a bit higher. Of course, if at any time the student is having difficulty reading at the goal level after the practice readings, the decision can be made to move the student down to an easier level or make a downward adjustment in the WCPM goal.

In addition to requiring the students to answer a set of comprehension questions at the end of each passage, some teachers have added other comprehension activities to this process, such as having the students write a five-minute re-tell response after each passage.

Using the RN strategy for 20-30 minutes per day, for three or more days per week, can have a significant impact on improving students’ reading fluency. In two studies reported on by Hasbrouck, Ihnot, and Rogers (1999), second- and third-grade Title I students, as well as sixth-grade special education students, showed significant improvement in their fluency. The second- and third-graders received, on average, 32 weeks of RN instruction. From fall to spring, the second-graders’ average WCPM increased from 17.9 to 71.6, meaning that they moved from well below the 25th percentile to well above it (see norms table in “Screening, Diagnosing, and Progress Monitoring: The Details”); they showed an average gain of 1.68 WCPM per week, much greater than the 1.2 WCPM per week gain that second-graders typically make.

Third-grade students had similar results. From fall to spring, their average WCPM increased from 42 to 93, meaning that they moved from just below the 25th percentile to well above it; they gained 1.60 WCPM per week, as compared to the typical growth of 1.1 WCPM per week. The study of sixth-grade special education students also found significant improvements. These students were reading at levels ranging from grade 1.5 to 4.0. They received RN instruction in a special education class for 20 to 32 weeks and improved their fluency by an average of 1.4 WCPM per week, which is double the 0.7 words per week that sixth-graders typically gain.

I would like to add two caveats regarding reading fluency. First, as this skill has recently garnered greater attention, and awareness of the link between fluency and comprehension has grown, there appears to be a tendency for some to believe that raising a student’s fluency score is the main goal of reading instruction. As important as fluency is, and as valuable as the information obtained from fluency-based assessments can be for instructional decision-making, I want to caution teachers and administrators to keep fluency and fluency-based assessment scores in perspective.

The ability to read text accurately, at a reasonable rate, and with appropriate expression and phrasing is certainly a key factor in being able to understand what has been read and to enjoy the process of reading. Nonetheless, fluency is only one of the key components of reading. I urge teachers to use the 50th percentile as a reasonable level of proficiency for students, and keep in mind that it is appropriate and expected for students to adjust their rate when reading texts of varying difficulty and for varied purposes.

Pushing every student to reach the 90th or even the 75th percentile in fluency is not feasible or necessary and, for students at or above the expected level in fluency, the instructional time could be better spent by enhancing other critical aspects of reading, such as increasing their vocabulary and becoming better at monitoring their comprehension.

The second caveat is that we still have much to learn about fluency. Ongoing debates in the research community include questions regarding the value of reading lists of words versus sentences and paragraphs; repeated reading of the same passage versus reading several different passages that have lots of the same vocabulary; the nature of the text in which students would benefit most for fluency practice (i.e., narrative or expository, randomly selected or highly controlled passages); the exact role of silent reading in a comprehensive reading instructional program; the role of prosody in the impact of fluency on text comprehension, etc.

For example, we know that the ability to instantaneously recognize high-frequency sight words is an essential element of fluent reading. Researchers continue to explore whether or not having students practice reading word lists or passages is the more efficient way to develop this automaticity. Until research provides a definitive answer, having students orally read passages seems more beneficial because of the added opportunity to work on prosody and comprehension.

Likewise, we know that repeated reading of a single passage is highly effective, but it is not clear whether or not a set of passages on a single topic that has been carefully written with a large number of repeated words could be equally or even more effective. If reading a set of passages turns out to be as effective as re-reading a single passage, the set could conceivably be used to enhance students’ fluency, vocabulary, and domain knowledge simultaneously.

We will leave researchers to continue their valuable efforts to address these important but yet-to-be-answered questions. However, this article should help practitioners feel confident that there is sufficient guidance from research to support the use of fluency-based assessments in their professional data-collection procedures, and to select instructional practices for both those students who are on-track and those who are struggling to develop the essential skill of reading fluency.

Jan Hasbrouck is president of JH Consulting, as well as an affiliate of the Behavioral Research and Teaching Group at the University of Oregon. Her most recent book, which she co-authored with Carolyn Denton, is The Reading Coach: A How-To Manual for Success.