© 2005 INTERNATIONAL READING ASSOCIATION (pp. 22–27) doi:10.1598/JAAL.49.1.

Is reading fluency a key for successful

high school reading?
Timothy V. Rasinski, Nancy D. Padak, Christine A. McKeon, Lori G. Wilfong,
Julie A. Friedauer, Patricia Heim

Findings in this study suggest that risk students) by their regular classroom teacher
found that the lack of reading fluency appeared
reading fluency is a significant variable to be the area of greatest impairment in reading
in secondary students’ reading and (Rasinski & Padak, 1998). Pinnell et al.’s (1995)
study of the relationship between oral
overall academic development. Rasinski teaches at Kent reading fluency and fourth graders’
State University (402 White
Hall, Kent, OH 44242, USA). silent reading comprehension found
E-mail [email protected]. that nearly half of the 1,000+ sample
With the publication of the report of the Padak also teaches at Kent
State University. McKeon
of fourth-grade students had not yet
National Reading Panel (National
teaches at Walsh University achieved a minimal level of reading
Institute for Child Health and Human in North Canton, Ohio. fluency.
Development [NICHD], 2000), read- Wilfong is a doctoral
ing fluency has become more recog- student at Kent State One hypothesized explanation
University, and Friedauer
nized as a key element in successful teaches at Mayfield City for the connection between fluency
reading programs in the primary Schools in Highland Heights, and comprehension comes from
Ohio. Heim teaches at
grades. Indeed, Chall’s (1983) seminal Thornton Elementary School
LaBerge and Samuels’s (1974) theory
model of reading identified the attain- in Newton, North Carolina. of automaticity in reading. According
ment of reading fluency as one of the to this theory, readers who have not
earliest stages of reading achievement. yet achieved automaticity in word
Given that reading fluency deals with mastery of recognition (fluency) must apply a significant
the surface level of text—learning to recognize amount of their finite cognitive energies to con-
(decode) words in a passage automatically (effort- sciously decode the words they encounter while
lessly) as well as accurately and to express or inter- reading. Cognitive attention or energy that must
pret those words in a meaningful manner when be applied to the low-level decoding task of read-
reading orally—it is quite appropriate to think of ing is cognitive energy that is taken away from the
fluency as a goal in reading that should be mas- more important task of comprehending the text.
tered as early as possible in one’s reading develop- Hence, comprehension is negatively affected by a
ment. reader’s lack of fluency.
Recent research, however, has suggested that Our work in a university reading clinic indi-
the issue of reading fluency goes beyond the pri- cates that difficulties in reading fluency are mani-
mary grades. Our own work among struggling fested in the majority of students in grades 2
elementary-grade students (grades 1–5) referred through 8 who are referred for reading difficulties.
for Title I supplementary reading instruction Although the primary reason for referral may os-
(Title I is a U.S. federally funded program for at- tensibly be difficulties in reading comprehension

22 JOURNAL OF ADOLESCENT & ADULT LITERACY 49:1 SEPTEMBER 2005

 Is reading fluency a key for successful high school reading?

(especially among intermediate and middle-grade ing or Oral Reading Fluency (ORF) Assessment
students), we also find that a lack of fluency ac- (Deno, 1985; Deno, Mirkin, & Chiang, 1982;
companies the difficulties in comprehension. Our Marston, 1989; Rasinski, 2004). Working individ-
clinical intervention program provides work in ually with one of us, students read a ninth-grade-
fluency and comprehension, and, for the most level passage, taken from the Secondary and
part, students make significant gains in both areas. College Reading Inventory (Johns, 1990), for one
minute. Although we recognized that the passage
may have been at a frustration level for some stu-
Fluency beyond the elementary dents, using grade-level materials is the conven-
grades tion for CBM/ORF assessments (Rasinski).
Students were asked to read orally in their normal
Although fluency is generally thought of as an el- voices and were told that they would be asked to
ementary grade issue, we wondered if fluency retell what they had read at the end of the read-
could be still be an issue in the reading difficulties ing. During the reading we marked any uncor-
experienced by large numbers of students beyond rected errors students made during the
the elementary grades. In particular, middle and one-minute period. We also asked them to do a
high school students from urban areas appear to quick retell of what they had read. The primary
experience more difficulty in reading than stu- purpose for the retelling was to ensure that stu-
dents from nonurban areas (e.g., National Center dents read in a normal manner—to read for un-
for Educational Statistics, n.d.). Could one source derstanding rather than speed.
of their difficulties in reading stem from a lack of
reading fluency? The high school at which we worked was di-
vided into individual “houses” in order to provide
To answer this question, we assessed the de- students with a smaller, more intimate learning
coding accuracy and fluency levels of a large environment. Students were randomly assigned
group of ninth-grade students at the end of the at the beginning of the school year to one of the
school year. In this study, fluency was defined as houses. We positioned ourselves at each house so
students’ reading rate. Although reading rate does that reading samples of students from all seg-
not capture the full meaning of fluency, it is con- ments of the school could be taken. Teachers who
sidered a useful and valid measure of fluency
were willing to allow their students to leave class
(Rasinski, 2004). The ninth graders in this school,
for periods of less than five minutes provided us
which is part of a moderate-sized urban district
with students to assess. We were assured by the
in the U.S. Midwest, have generally performed
teachers and administrators in the school that the
poorly on the state high school graduation
students we tested were a representative sample of
tests—a series of tests across important content
all of the students in the school.
areas in which students read and respond to text
passages that reflect the various content domains. From the one-minute reading we were able
On one day during the last week of the school to determine each student’s word-recognition
year (June, 2003) we visited the high school at level, as measured by percentage of words read
which half of the school district’s freshmen were correctly, and reading fluency, as determined by
enrolled. We selected the last week of the school number of words read correctly in the time peri-
year to ensure that the reading samples we ob- od. We were also able to obtain students’ per-
tained reflected the most advanced levels of read- formance scores on the state high school
ing exhibited by students during the year. During graduation test—a silent reading comprehension
the day, we tested 303 students using a one- test across all major subject areas that they had
minute reading probe, also known as taken earlier. The high school graduation test
Curriculum-Based Measurement (CBM) in read- consists of a series of passages read silently, and

JOURNAL OF ADOLESCENT & ADULT LITERACY 49:1 SEPTEMBER 2005 23

 Is reading fluency a key for successful high school reading?

each passage is followed by a set of comprehen- spring fluency norms for grade 8 students (Johns
sion questions. The test is given in the ninth & Berglund, 2002). We chose to use these norms,
grade, and students are required to pass it in or- recognizing that they are conservative estimates
der to qualify for a high school diploma. of ninth graders’ reading and generally understate
their reading performance. According to these
Based on our reading assessment, we found
norms, the 50th percentile spring norm for
that the end-of-the-year ninth graders in this ur-
eighth-grade students is 171 words correct per
ban school read with an average word-recognition
minute (wcpm); the 25th percentile norm for
accuracy rate of 97.4% (standard deviation =
eighth graders is 145 wcpm. Extrapolating these
2.8%) correct and a reading fluency rate of 136.4
data, we could logically expect ninth-grade stu-
(standard deviation = 33.2) words correct per dents to read at an even higher fluency level than
minute. Now just what do these scores mean? For eighth graders. Once we put this in context it was
word recognition, it appears that the students apparent that, on average, these ninth graders’
were able to decode words quite accurately. fluency levels were below the 25th percentile for
Normally, a 95% word-recognition accuracy level eighth graders. These ninth graders read at a flu-
is considered to be an instructional level. Thus, ency level that was about 80% of what might be
the students in our sample displayed, on average, considered the norm (50th percentile) for eighth-
strong proficiency in word recognition. grade students. It was clear then that these stu-
Fluency, however, seems to be a different dents, as a whole, had not achieved a level of
matter. Because students’ reading rate increases as fluency that would be considered normal or aver-
they mature across and within grade levels, it is age for their grade level.
necessary to compare students’ oral reading flu- To further detail the performance of this
ency performance against established norms. It group of ninth-grade students, we report their
was unfortunate that we were not able to find es- fluency (rate) scores in Table 1. Using the eighth-
tablished norms for students in grade 9 or above. grade norms, we should expect 25% percent of
To the best of our knowledge, such norms do not eighth-grade students to fall at and above the
currently exist and reflect the conventional wis- 75th percentile, 25% of students to fall at or be-
dom that reading fluency is not an issue at the low the 25th percentile, and 50% of students to
secondary level. We were, however, able to find fall between the 26th and 74th percentile. In our

Ta b l e 1
Ninth-grade student fluency performance compared against
existing norms for grade 8

Eighth-grade Number of ninth-grade students Percentage of ninth-grade

student norms in current study in current study
75th percentile and above 13 4.3
≥ 193 wcpm
26th–74th percentile 114 37.6
146–192 wcpm
25th percentile and below 186 61.3
≤ 145 wcpm

24 JOURNAL OF ADOLESCENT & ADULT LITERACY 49:1 SEPTEMBER 2005

 Is reading fluency a key for successful high school reading?

study of ninth graders, fully 61% of students included more higher achieving students so as to
scored at or below the 25th percentile. More than be representative of the full population of high
two times the number of ninth-grade students school students in the state or nation, it is likely
fell within the bottom range than what should that the correlation between fluency and compre-
normally be expected of eighth graders. hension would have been even stronger.
We also determined the number of students Nevertheless, the relationship between fluen-
from our sample who read at less than 100 words cy and comprehension that we report in this arti-
correct per minute. We selected 100 wcpm as a cle suggests that reading fluency is indeed a factor
conservative indicator of significant concerns in that needs to be considered even among high
reading fluency. A rate of 100 wcpm is generally school students, and especially among struggling
reflective of an end-of-year reading rate for readers. The high school students in our study, on
grades 2–3 (Hasbrouck & Tindal, 1992). A total of average, read at a fluency level that is below what
36 ninth graders (12%) read below this bench- would normally be expected of eighth graders;
mark rate. Taking 167 wcpm as the average read- moreover, their reading fluency levels were related
ing rate against which teachers measure reading to their comprehension performance.
assignments, any reading assignment given to this
group of students (nearly one out of every eight
students) requires at least 150% more time to What does this mean?
complete than what the teacher might otherwise
expect. It is clear that these various analyses indi- Although a correlation between fluency and com-
cate that a significantly large number of students prehension does not prove causation—that fluency
in this low-performing school are not close to ad- or lack of fluency leads to improved or deficient
equate levels of fluency and may benefit from comprehension—the findings do suggest that this
specific instruction in reading fluency. is a possibility. The theory of automaticity again
offers a compelling explanation for this finding.
Because fluency appears to be an area of Although the high school students in this study
concern among this group of students, we at- read with a high degree of accuracy, they had to
tempted to determine the relationship between invest so much of their limited cognitive energy
reading fluency and reading comprehension for in accomplishing this task that they drained cog-
these ninth graders. We did this by running a cor- nitive capacity away from where it could and
relation between the fluency (rate) scores and the should have been used more profitably—to com-
students’ scores on the state high school gradua- prehend the text.
tion test—a test of reading comprehension. We
found a statistically significant (p < .001) and Although variation in fluency does not ac-
moderately strong relationship between these two count for a majority of variation in comprehension
variables (r = .530). This means that about 28% on the state high school graduation test, the 28%
of the variation in student achievement on the that is accounted for by fluency represents a signifi-
high school graduation test could be accounted cant portion of comprehension performance. The
for by variation in students’ reading fluency. results of our study lead us to conclude that im-
provements in fluency could account for significant
We believe that the correlation statistics re-
and substantial gains in students’ reading compre-
ported here actually underestimate the relation-
hension.
ship of fluency to comprehension among high
school students because those in our study repre- It is clear that this hypothesis needs to be
sent a restricted sample—they performed, on aver- tested. High school students deficient in reading
age, at a level that is below grade-level expectations fluency could be given an instructional interven-
for fluency and comprehension. Had the study also tion that focuses on reading fluency. If fluency

JOURNAL OF ADOLESCENT & ADULT LITERACY 49:1 SEPTEMBER 2005 25

 Is reading fluency a key for successful high school reading?

does indeed contribute to comprehension among readings also lead to gains in fluency, comprehen-
high school students, gains in fluency and com- sion, and overall reading on other passages not
prehension should be detected. previously encountered. In other words, student
practice on certain passages generalizes to im-
Beyond future research, however, these find-
proved performance across all reading.
ings indicate that some attention to reading fluency
in high school is called for. At the very least, it is Repeated or practiced reading is best ac-
clear that overly slow and disfluent reading is a complished through performance activities.
detriment to reading proficiency. Readers who When students are asked to perform for others,
read at an excessively slow pace, even without af- they have a natural inclination and desire to prac-
fecting comprehension, are at a disadvantage when tice the passage to the point where they can read
compared with their classmates who read at a it accurately, with appropriate rate, and especially
more normal rate. In our study, we found that 186 with meaningful expression and phrasing. Texts
of 303 students (61.3%) read at a rate that was at such as poetry, scripts, oratory, and song lyrics are
or below the 25th percentile rate for eighth-grade meant to be performed and could be incorporated
students. This means that these students require into any secondary content area classroom with a
significantly more time to accomplish any reading bit of creative planning by the teacher.
assignment than do students who read at a normal
A second proven method for developing
reading rate. Such levels of reading performance
fluency has been termed assisted reading (Kuhn
can easily lead to frustration, avoidance of reading,
& Stahl, 2000; NICHD, 2000; Rasinski &
and, ultimately, school failure.
Hoffman, 2003). That is, students read a passage
Students learn what teachers teach. And be- while simultaneously listening to a fluent oral
cause reading fluency has generally been thought rendering of the same text by a person or persons
of as within the domain of the elementary grades, or on a previously recorded version of the read-
it is unlikely that fluency is taught directly or sys- ing. Choral reading is certainly one way to pro-
tematically in the middle and secondary grades. vide assisted reading. Another is for the less fluent
Students who lack sufficient fluency entering into reader to read along with a more fluent partner.
the middle grades are not likely to find much in- That partner could be the teacher, a tutor, a class-
structional support for their difficulties. If fluency mate, or a parent. Teachers might also tape record
is a concern among middle and high school stu- selected passages for less fluent readers and ask
dents, it needs to be taught. them to read the passage while listening to the
tape repeatedly until they feel that they can read
Fortunately, it is not difficult to integrate
the passage independently.
reading fluency into regular classroom instruction
in literacy and other content areas. Reading fluency Employing methods such as these to im-
develops with contextual reading practice. Wide prove students’ reading fluency does require addi-
reading of independent-level material (Allington, tional work for the teacher. However, if our goal
2000) and guided reading of instructional-level is to improve student performance across content
material are clearly good ways to develop reading areas, then improvements in general reading abil-
fluency. Repeated readings, another form of read- ity must be a goal. As the study reported here ten-
ing practice, is one of the most powerful ways to tatively suggests, lack of reading fluency, an
increase reading fluency (Dowhower, 1994; Kuhn instructional goal of the reading curriculum that
& Stahl, 2000; NICHD, 2000; Rasinski & Hoffman, has not traditionally been given importance in
2003; Samuels, 1979). Through repeated readings secondary schools, may be one important cause
of a particular text, students increase their fluency for reading comprehension difficulties among
and comprehension of the passage practiced. What secondary school students. Some attention to flu-
is more important, however, is that the repeated ency for those students who are not fluent readers

26 JOURNAL OF ADOLESCENT & ADULT LITERACY 49:1 SEPTEMBER 2005

 Is reading fluency a key for successful high school reading?

offers promise of significant improvements in Deno, S.L. (1985). Curriculum-based measurement: The
reading comprehension and overall academic emerging alternative. Exceptional Children, 52, 219–232.
Deno, S.L., Mirkin, P., & Chiang, B. (1982). Identifying valid
performance across the content areas.
measures of reading. Exceptional Children, 49, 36–45.
Dowhower, S.L. (1994). Repeated reading revisited: Research
into practice. Reading and Writing Quarterly, 10, 343–358.
A caveat Hasbrouck, J.E., & Tindal, G. (1992). Curriculum-based oral
reading fluency norms for students in grades 2 through 5.
Although we have identified reading rate as a Teaching Exceptional Children, 24, 41–44.
method for assessing reading fluency, we want to Johns, J.L. (1990). Secondary and college reading inventory
make it clear that reading rate does not represent (2nd ed.). Dubuque, IA: Kendall/Hunt.
the full meaning of fluency. We define reading flu- Johns, J.L., & Berglund, R.L. (2002). Fluency: Answers, ques-
tions, evidence-based strategies. Dubuque, IA:
ency as reading with appropriate accuracy and Kendall/Hunt.
rate but also with good and meaningful phrasing Kuhn, M.R., & Stahl, S.A. (2000). Fluency: A review of devel-
and expression. This oral interpretative aspect of opmental and remedial practices (CIERA Rep. No. 2-008).
fluency is the pinnacle. Teaching students to read Ann Arbor, MI: Center for the Improvement of Early
quickly will not necessarily lead to more fluent Reading Achievement.
LaBerge, D., & Samuels, S.A. (1974). Toward a theory of au-
readers or better comprehenders.
tomatic information processing in reading. Cognitive
Our point is to suggest that teachers beware Psychology, 6, 293–323.
Marston, D. (1989). A curriculum-based measurement ap-
of fluency programs or interventions that seek
proach to assessing academic performance: What it is and
solely to boost student reading rate. Rate-building why do it. In M.R. Shinn (Ed.), Curriculum-based meas-
exercises and admonitions to read faster will result urement: Assessing special children (pp. 18–78). New York:
only in students who read quickly but still do not Guilford.
comprehend what they read. Our own work in National Center for Educational Statistics. (n.d.). Cross-
district comparisons of percentage of students at or above
fluency suggests that the instructional activities
Proficient in reading, grade 8 public schools: By urban dis-
outlined earlier—repeated and assisted reading for trict, 2002. Retrieved July 7, 2004, from http://nces.ed.
meaning—will lead to faster reading and, what is gov/nationsreportcard/reading/results2002/district
more important, readers who read with and for compareprof-g8.asp
meaning. National Institute of Child Health and Human Development.
(2000). Report of the National Reading Panel. Teaching
Although clearly not definitive, this study children to read: An evidence-based assessment of the scien-
suggests that fluency needs to be a concern for tific research literature on reading and its implications for
teachers at all grade levels, not just teachers of be- reading instruction (NIH Publication No. 00-4769).
Washington, DC: U.S. Government Printing Office.
ginning readers. It makes good sense that even
Pinnell, G.S., Pikulski, J.J., Wixson, K.K., Campbell, J.R.,
older students who read with a lack of sufficient Gough, P.B., & Beatty, A.S. (1995). Listening to children
fluency will have difficulty comprehending what read aloud. Washington, DC: U.S. Department of
they read. We hope this article will lead to further Education, Office of Educational Research and
research into the role of fluency in the middle Improvement.
Rasinski, T.V. (2004). Assessing reading fluency. Honolulu,
and secondary grades and will inspire middle and
HI: Pacific Resources for Education and Learning.
secondary teachers, regardless of their content Rasinski, T.V., & Hoffman, T.V. (2003). Theory and research
specialty, to attempt to make reading fluency an into practice: Oral reading in the school literacy curricu-
integral part of their reading instruction. lum. Reading Research Quarterly, 38, 510–522.
Rasinski, T.V., & Padak, N.D. (1998). How elementary stu-
dents referred for compensatory reading instruction per-
REFERENCES form on school-based measures of word recognition,
Allington, R.L. (2000). What really matters for struggling read- fluency, and comprehension. Reading Psychology: An
ers. New York: Allyn & Bacon. International Quarterly, 19, 185–216.
Chall, J.S. (1983). Stages of reading development. New York: Samuels, S.J. (1979). The method of repeated readings. The
McGraw-Hill. Reading Teacher, 32, 403–408.

JOURNAL OF ADOLESCENT & ADULT LITERACY 49:1 SEPTEMBER 2005 27