Three decades of sustained silent reading: a meta-analytic review of the effects of SSR on attitude toward reading.
Literacy (Standards)
Reading (Standards)
Reading (Methods)
Silent reading (Usage)
Silent reading (Analysis)
Literacy (United States)
Yoon, Jun-Chae
Pub Date:
Name: Reading Improvement Publisher: Project Innovation (Alabama) Audience: Professional Format: Magazine/Journal Subject: Education Copyright: COPYRIGHT 2002 Project Innovation (Alabama) ISSN: 0034-0510
Date: Winter, 2002 Source Volume: 39 Source Issue: 4
Event Code: 350 Product standards, safety, & recalls
Geographic Scope: United States Geographic Code: 1USA United States

Accession Number:
Full Text:
The purposes of this meta-analytic study were to investigate the overall effect of Sustained Silent Reading (SSR) on attitude toward reading and to identify the moderator variables of SSR on it. A meta-analytical approach developed by Hedges and Olkin (1985) was used. Results indicated that the average of effect size on reading attitude is .12. For moderator variables, grade level was found to be significant for reading attitude. These findings suggest that providing a fixed period of time for students to read materials of their own choosing either for pleasure or for information facilitate their attitude toward reading. Furthermore, this study supports the recommendation of earlier intervention of SSR in improving students' attitude toward reading.


Most education systems worldwide have emphasized a high level of reading and a love of reading as important educational objectives (Elley, 1994). Also, the majority of teacher educators and teachers in the field of literacy education have rated highly the instructional goal of fostering students' reading attitude and reading performance (Morrow, 1991). Nevertheless, aliteracy, "lack of the reading habit in capable readers," (Harris & Hedges, 1995, p. 6) has been a serious concern for children in many countries. According to Anderson, Wilson, and Fielding (1988), American students spend only an average of 18.4 minutes per day reading printed materials out of school. Unfortunately, there has been no significant increase in the amount of time American students spent reading either for pleasure or for learning (Campbell, Voelkl, & Donahue, 1997).

Similarly, Korean students spend an average of 30 minutes of reading and an average 160 minutes of watching television per day at home. Since 1990, no significant change has been observed in the amount of time they reported reading for pleasure out of school (Yoon, Kim, Yi, & Yi, 2000). More recently, the OECD (Organization for Economic Cooperation and Development) report of international student assessment showed that most 15-year-olds worldwide do not read for enjoyment. Further, of those who read, most do for less than an hour per day out of school (Henry, 2001).

As a way of cultivating a love of reading, Sustained Silent Reading (SSR), which is an in-classroom reading activity wherein students are given a fixed period of time for the silent reading of self-selected material either for pleasure or for information, has gained popularity in many elementary and secondary classrooms in the United States, Canada, United Kingdom, and New Zealand (Dymock, 200; Halpern, 1981; Nagy, Campenni, & Shaw, 2000; Southgate, 1975, cited in Wheldall & Entwistle, 1988). Reviews of SSR (see Dymock, 2000; National Reading Panel, 2000; Sadoski, 1984; West, 1995), however, failed to establish its educational impact on attitude toward reading. Indeed, scientific evidence regarding the effectiveness of SSR on reading attitude is equivocal. For example, a few studies have indicated that SSR promotes positive attitudes toward reading (Aranha, 1985; Dully, 1989; Wilmot, 1975). In contrast, other research results question whether SSR has a positive influence on attitude toward reading (Collins, 1980; Dwyer & Reed, 1989; Langford & Allen, 1983; Manning & Manning, 1984; Summers & McClelland, 1982).

It was the purpose of this paper to examine the effects of SSR on reading attitude by employing Hedges and Olkin's (1985) meta-analytic procedure. Two research questions were posed: (a) Does SSR enhance students' attitude toward reading and (b) what contextual features of SSR are associated with students' reading attitude? The purpose of the first question was to evaluate the overall effect of SSR on reading attitude. The second question permitted the identification of SSR moderator variables on it.

Theoretical Framework

SSR and Reading Attitude

It has been argued that there are three important characteristics of SSR, which are supposed to mediate the relationship between reading attitude/reading comprehension and the activity. These include self-selection, role modeling, and nonaccountability.


Readers may pay closer attention, persist in their reading for longer periods of time, learn more, and enjoy their involvement to a greater degree when they read materials that interest them. According to Deci and Ryan's (1985) theory of self-determination and intrinsic motivation, children's natural curiosity energizes their desire to learn. The self-determination theory suggests that children are motivated when they have choice and ownership in what they read. Therefore, providing children with an opportunity to select reading materials promotes their literacy development because they have their own ownership of what they read (Shannon, 1995).

Several studies have demonstrated that individual preference of topics has a facilitative effect on cognitive and affective functioning. For example, Rehder (1980) sought to determine "how much liking affects reading achievement." Secondary school students involved in a popular fiction course who allowed to choose paperback books significantly outperformed control group students who participated in a composition class. Similarly, Cecil (1984) reported that Native American students who read books in which they were interested demonstrated higher reading attitude and reading comprehension of those books compared to books in which they were less interested. More recently, Schiefele (1991) investigated how college students' interest in materials influenced their comprehension when their prior knowledge of the materials and general intelligence were controlled. Schiefele found that students who were interested in the materials processed those materials more deeply than did students less interested in the materials. Furthermore, Fink (1996) showed that even dyslexics, who were hampered by persistent deficiencies in basic reading skills, could construct meaning in a single high interest domain. In contrast, when self-choice is removed, individual intrinsic motivation to reading may be diminished (Gottfried, 1990). These results suggest that preference might be a crucial motivational element leading to an increase of reading attitude and reading comprehension. Thus, by providing opportunities for self-selection in SSR, a teacher can foster children's involvement in reading materials and promote their literacy development.

Role Modeling.

Human behavior is learned in part by observation and imitation (Bandura, 1986). For young children, in particular, a given behavior modeled by parents, teachers, peers, or celebrities may facilitate their learning of it. From this lens, showing a reading behavior to them may be one of the most important sources of developing their attitude toward reading. Several researchers in the field of literacy education have suggested that role modeling is a crucial factor for reading attitude acquisition and development. For example, Gambrell (1981) stated that "students need to see that we value reading and that reading is important in our lives. Share with your students. What better way to show them that reading is important?" (p. 898). Wheldall and Entwhistle (1988) conducted a study to determine whether teacher role modeling is an important factor in children's reading behavior development. Fourth-grade students of mixed ability participated in SSR activity over a six-week period. They reported an overall mean increase in on-task-reading behavior of 32% from initial baseline to second intervention when a teacher modeled reading during SSR sessions. These results clearly demonstrate that a teacher role modeling as nonverbal feedback plays a crucial role in fostering children's reading attitude.


Children may not be required to keep records, prepare book reports or daily reading journals, or write summaries during SSR sessions. In high accountability, it may lead even active readers to invent ways of showing the autonomy of their reading for the teacher, while for reluctant readers it may be so threatening that they never experience the pleasure of reading. In this regard, one teacher who participated in the Nagy et al.'s survey (2000) strongly commented that "reading should be a spark to ignite a fire--heavy accountability tends to throw water on the spark. If it is graded, it defeats the purpose of reading class ... to become life-long readers" (p. 6). Schiavone's (2000) study also showed that accountability for reading did not play a crucial role on the reading comprehension and attitude of seventh-grade children. Thus, instead of imposing heavy-accountability, rather teachers should always exert all possible efforts to share his or her reading experience with children and to entice comments from them about reading out of the activity.

Variability in Study Features

Studies included in this paper differ in methodological and procedural features as well as sample characteristics such as the type of treatment conducted, participant characteristics, sample size, and type of control group used. Such contextual features known as moderator variables (Rosenthal, 1984) may be associated with variations in the magnitude of the relationships between an independent variable and a dependent one. That is, some contextual characteristics across the reviewed studies may influence significant variability in the magnitude of the relationships, that is, in the effect sizes of those studies. From the literature review, several contextual features (e.g., publication type, duration of treatment, treatment type of control group, participant's grade level and ability level, methodological quality, and teacher's role) were found as potential moderator variables of SSR studies. Of them, two contextual factors, duration of treatment, and participant's grade level, were analyzed because they are more significant moderators that exert on influence on the effect sizes of the studies as compared with the other ones.

Duration of Treatment.

Each of the studies included in this analysis varies on its duration of treatment. Intervention in several studies (Dully, 1989; Schon, Hopkins, & Vojir, 1985) lasted more than a semester, approximately 6 months, while that of others (Box, 1984; Collins, 1980) did less than 6 months. However, evidence regarding the effectiveness of duration of treatment as a moderator variable on reading attitude is equivocal. Even one school-year study (Dully, 1989) did not report a positive effect on reading attitude, whereas just a 13-week study (Box, 1984) showed a positive effect on it. In light of the cost and the effort of producing a given effect, duration of treatment gives us indicator to evaluate practical value of the program. For instance, if short duration of treatment can produce a significant impact on reading attitude as much as longer duration of intervention, less expensive and effortful one may justify broad dissemination ad adoption. In this regard, I chose duration of treatment as a moderator variable to investigate if such contextual factor influences significant variability in the magnitude of the effect sizes of those studies. The level of this moderator was classified as more than 6 months and less than 6 months because a semester-treatment-duration, approximately 6 months or less, is not substantially short in school context.

Grade Level.

A variety of grade levels of students ranging from a second grader to a college level have participated in SSR. However, empirical support regarding the effectiveness of grade level on reading attitude is not entirely convincing. Wilmot (Wilmot, 1975) reported that there is a statistically significant effect for the fourth- and sixth-grade students, whereas no significant one is found for the second graders. Holt and O'Teul (1989) showed that there is a positive effect for the seventh graders, while Einhorn (1979) reported that there is no significant impact on reading attitude for the fifth-grade students. In view of the effectiveness of intervention, it is important to know at which grade levels such contextual factor influences the magnitude of a relationship because it may provide us with an optimal time to maximize the effectiveness of SSR intervention. In this regard, grade level of students was selected as a moderator variable to investigate if such a contextual factor moderates significant variability in the magnitude of the effect sizes of those studies. The grade level was classified as below 3rd grade and above 4th grade because after the third grade more and more children are likely to participate in leisure options such as sports and social communities, which can influence the magnitude of the effectiveness of SSR intervention.


Data Collection

To evaluate the effectiveness of SSR on attitude toward reading, I employed a meta-analytical approach developed by Hedges and Olkin (Hedges & Olkin, 1985). To identify relevant studies, ERIC (Educational Resources Information Center) and UMI (University Microfilms) databases were searched. In particular, including unpublished doctoral dissertations as well as unpublished articles in this analysis helps me avoid more or less publication bias related to the "file drawer problem" (Rosenthal, 1979), which is tendency for authors not to submit and journals editors not to accept for publication studies that fail to produce statistically significant results. However, the search was limited to the research and literature from 1970 to the present because research on SSR has been widely conducted since Hunt (1970) introduced Uninterrupted Sustained Silent Reading (USSR) as one of the important elements of an Individualized Reading Program (IRP). Through such search, 350 studies were collected as a preliminary data set. At the initial selection, abstracts of 350 studies were reviewed and 307 studies were removed because they did not provide statistical information necessary for calculating effect size. For the review of the second list, full texts of the remaining 43 studies were reviewed based on the following criteria: (a) an SSR group was compared to a control group, (b) studies contained enough statistical information to estimate effect size, (c) studies were published or unpublished after 1970, and (d) outcome measures included reading attitude. After the second reviewing, 7 studies remained in this analysis.

Effect Size Calculation

The meta-analysis program developed by Schwarzer (1996) was used to calculate effect sizes. For studies including multiple treatments case, the effect size of each comparison was calculated individually. For studies in which multiple outcome measures were used, the effect size of each dependent measure was averaged. Thus, although 7 studies for this analysis were selected, 11 effect sizes were obtained and calculated.

In addition to the calculation of effect sizes, statistical analyses of the effect sizes were performed. For the first purpose of this study, the null hypothesis that the population value of weighted average of corrected effect sizes equals zero was tested. To identify the moderator variables of SSR, a Q statistic (Hedges & Olkin, 1985) that assesses if the effects produced by a group of studies vary primarily because of sampling error or represent systematic differences among the studies, was used.


Overall Effect

Eleven comparisons from 7 studies were used to get the average effect size of SSR on reading attitude. The mean of effect size (ES) was .12, and its standard error was .04. The statistical test did not support the hypothesis that the population ES is zero ([x.sup.2](1) = 12.11, p < .01). The result indicates that there is empirical support for SSR affecting students' reading attitude. Table 1 shows the descriptive statistics on the effect size of reading attitude.

Moderator Variable Effect

The purpose of the second question is to identify the moderator variables of SSR on reading attitude. The table 2 displays descriptive statistics for the effect size by duration of treatment and grade level.

The [H.sub.b] statistic, the test of heterogeneity, indicated that effect sizes are not different by duration of treatment ([x.sup.2] (1) = .05, p > .05). This result shows that there is no effect size difference between more-than 6-month duration of treatment and less-than-6-month one. That is, SSR equally fosters students' reading attitude regardless of duration of treatment. In contrast, the Hb statistic for grade levels showed that effect sizes differed on students' grade levels ([x.sup.2] (1) = 6.67, p < .01), indicating that SSR is a more effective reading activity to increase attitude toward reading for the lower grade students than for the higher grade students.

Discussion and Implications

One of the most important findings of this study was an affirmative evidence for significant reading attitude gains from a fixed period of time for students to read materials of their own choosing either for pleasure or for information. As outlined above, the average effect size of SSR on attitude toward reading was .12, which corresponds to the percentile of .55, indicating that the reading attitude score of the average individual in the SSR group exceeds the scores of 55% of individuals in the control group. This result provides evidence to support the effectiveness of the SSR reading activity at enhancing students' reading attitude. It is consistent with findings from several models of reading attitude acquisition and development (e.g., Mathewson, 1994; McKenna, 1994) that show the characteristics of SSR presented above function as external and internal motivators to play a crucial role in improving a reader's reading attitude. According to Cohen's (1988) criterion of the magnitude of effect sizes, however, an effect size of .12 is small. Nevertheless, this small effect may be regarded as a considerable and practical progress in classrooms because the development of attitude toward reading takes time and evolves in a slow way (Sadoski, 1980).

The results from the moderator variables yield an affirmative answer to the second research question. The fact that students in the experiments of less-than-6-month duration of treatment (ES = .12) did not gain significantly more reading attitude scores than their peers in the experiments of more-than-6-month duration of treatment (ES = .14) provides evidence to support that the effectiveness of the reading activity at improving students' reading attitude is not associated with duration of treatment. That is, the activity equally enhances students' reading attitude regardless of duration of treatment. An alternative explanation of it may be that a semester-treatment-duration might be substantially long enough to provide opportunities for students to have a joyful reading experience in school context. However, students may not maintain this positive effect of the activity over a relatively short period time, for example, one month because the possibility that they have a pleasing reading experience may decrease as duration of treatment reduces. To enhance children's positive reading attitude, thus, a relatively long duration of treatment should be considered.

In contrast, the heterogeneity test for grade levels showed that effect sizes differ on students' grade levels. For the lower grade students, the average effect size was .32, which corresponds to the percentile of .63, whereas for the higher-grade students, the mean effect size was .06, which corresponds the percentile of .52. The results indicate that the reading attitude score of the average individual in the SSR group exceeds the scores of 63% and 52% of individuals in the control groups, respectively. According to Cohen's (1988) criterion of effect sizes, the effect size for the lower students is medium, while one for the higher students is small. Taken together, primary low-grade students benefit from the activity compared with upper high-grade students possibly because higher grade students' reading attitude tends to decline gradually as they participate more and more in leisure options such as sports and social communities. This result supports the recommendation of earlier intervention in improving children's attitude toward reading.

Admittedly, the current study has limitations. First, all eligible studies were not exhausted because this analysis relied on ERIC and UMI databases. Second, by the very nature of a meta-analysis, qualitative research relevant to SSR was not included. Finally, there was a lack of theoretical background of multi-way interactions among moderator variables. Thus, the findings from this study should be carefully interpreted.


Anderson, C.R., Wilson, P.T., & Fielding, L.G. (1988). Growth in reading and how children spend their time outside of school. Reading Research Quarterly, 23, 285-303.

Aranha, M. (1985). Sustained silent reading goes East. The Reading Teacher, 39, 214-217.

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall.

* Box, L. (1984). A study of achievement and attitudes of high-achieving, average-achieving. and low-achieving third grade students involved in daily sustain silent reading. Unpublished doctoral dissertation, Mississippi State University, Mississippi.

Campbell, J.R., Voelkl, K., & Donahue, P.L. (1997). Report in Brief: NAEP 1996 trends in academic progress. Washington, DC: National Center for Education Statistics.

Cecil, N.L. (1984). Impact of interest on the literal the literal comprehension of beginning readers--a West Indian study. The Reading Teacher. 37, 750-753.

Cohen, J. (1988). Statistical power analysis for the behavior science (2nd ed.). Hillsdale, NJ: Erlbaum.

Collins, C. (1980). Sustained silent reading periods: Effect on teacher's behavior and students' achievement. The Elementary School Journal. 81,109-114.

Deci, E.L., & Ryan, R.M. (1985). The general causality orientations scale: Self-determination in personality. Journal of Research in Personality, 9, 109-134.

Dully, M. (1989). The relation between sustained silent reading to reading achievement and attitude of the at-risk student. (ERIC Document Reproduction Service No. ED 312 631)

Dwyer, E., & Reed, V. (1989). Effects of sustained silent reading on attitudes toward reading. Reading Horizons, 29, 283-293.

Dymock, S.J. (2000). The effect of sustained silent reading on reading comprehension: A review of the research. Paper presented at the National Reading Conference, 50th annual meeting, Scottsdale, AZ.

Einhorn, J.A. (1979). Sustained silent reading practice: Its effects upon student reading habits and attitudes toward reading. (ERIC document reproduction service No. ED 174 069)

Elley, W.B. (1994). Voluntary reading activities. In W.B. Elley (Ed.), The IEA study of reading literacy: Achievement and instruction in thirty-two school systems (pp. 65-148). Tarrytown, NY: Author.

Fink, R.P. (1996). Successful dyslexics: A constructivist study of passionate interest reading. Journal of Adolescent & Adult Literacy, 39, 268-280.

Gambrell, L.B. (1981). The clip sheet. The Reading Teacher, 34, 836-838.

Gottfried, A.E. (1990). Academic intrinsic motivation in elementary and junior high school students. Journal of Educational Psychology. 77, 631-645.

Halpern, H. (1981). An attitude survey of uninterrupted sustained silent reading. Reading Horizons, 21, 272-279.

* Hanson, R.A. (1972). A study of the effects of sustained silent reading on the reading attitudes and habits of second, third and fourth grade students in a low income area school. Unpublished doctoral dissertation, The University of North Dakota, Grand Forks.

Harris, T.L., & Hodges, R.E. (1995). The literacy dictionary. Newark, DE: International Reading Association.

Hedges, L.V., & Olkin, I. (1985). Statistical methods for meta-analysis. Orland, FL: Academic Press.

Henry, T. (2001, December 4). Getting a read on US teens. US Today, p. 7D.

Holt, S.B. & O'Tuel, F.S. (1989). The effect of sustained silent reading and writing on achievement and attitudes of seventh and eighth grade students reading two years below grade level. Reading Improvement, 26, 290-297.

Hunt, L.C. (1970). Updating the individual approach to reading: IPI of IRP? (ERIC Document Reproduction Service No. ED 044 244)

* Langford, J.C., & Allen, E.G. (1983). The effects of U.S.S.R. on students' attitudes and achievement. Reading Horizons, 23, 194-200.

Manning, G., & Manning, M. (1984). What models of recreational reading make a difference? Reading World, 23, 375-380.

Mathewson, G.C. (1994). Model of Attitude Influence upon Reading and Learning to Read. In R.B. Ruddell, M.R. Ruddell, & H. Singer (Eds.), Theoretical models and processes of reading (4th ed., pp. 1131-1161). Newark, DE: International Reading Association.

McKenna, M.C. (1994). Toward a model of reading attitude acquisition. In E.H. Cramer & M. Castle (Eds.), Fostering the love of reading: The affective domain in reading education (pp. 18-40). Newark, DE: International Reading Association.

Morrow, L.M. (1991). Promoting voluntary reading. In J. Flood, J.M. Jensen, D. Lapp & J.R. Squire (Eds.), Handbook of research on teaching the English language arts (pp. 681-690). New York: Macmillan.

Nagy, N.M., Campenni, C.E., & Shaw, J.N. (2000). A survey of sustained silent reading practices in seventh-grade classrooms. Reading Online [On-line]. 1-12. Available:

National Reading Panel. (2000). Teaching children to read: An evidence-based assessment of the scientific research literature on reading and its implication for reading instruction. Available [Online]

* Reedy, J.D. (1994). Effects of a sustained silent reading program with literature response journals on third graders' attitude reading achievement and writing. Unpublished doctoral dissertation, Baylor University, Waco.

Rehder, L.G. (1980). Reading skill in a paperback classroom. Reading Horizons, 21, 16-21.

Rosenthal, R. (1979). The file draw problem and tolerance for null results. Psychological Bulletin, 86, 638-641.

Rosenthal, R. (1984). Meta-analytic procedures for social research. Beverly Hills, CA: SAGE.

Sadoski, M.C. (1980b). Ten years of uninterrupted sustained silent reading. Reading Improvement, 17, 153-156.

Sadoski, M.C. (1984). SSR, accountability and effective reading instruction. Reading Horizons, 24, 119-123.

Schiavone, J.P. (2000). Sustained silent reading and reading workshop: The impact of accountability on seventh graders' reading comprehension and reading attitude. Unpublished doctoral dissertation, George Washington University, Washington, D.C.

Schiefele, U. (1991). Interest, learning, and motivation. Educational Psychologist, 26, 299-323.

* Schon, I., Hopkins, K., & Vojir, C. (1985). The effects of special reading time in Spanish on the reading abilities and attitudes of Hispanic junior high school students. Journal of Psycholinguistic Research, 14, 57-65.

Schwarzer, R. (1996). Unpublished manuscript. Available [Online]

Shannon, P. (1995). Texts, lies, & videotape: Stories about life, literacy, & learning. Portsmouth, NH: Heinemann.

* Summers, E.G., & McClelland, J.V. (1982). A field-based evaluation of sustained silent reading (SSR) in intermediate grades. The Alberta Journal of Educational Research, 28, 100-112.

West, K.M.M. (1995). Silent reading: What the research says. Unpublished masters' thesis, Texas Woman's University.

Wheldall, K., & Entwistle, J. (1988). Back in the USSR: The effect of teacher modeling of silent reading on pupils' reading behavior in the primary school classroom. Educational Psychology: An International Journal of Experimental Educational Psychology, 8, 51-66.

* Wilmot, M.P. (1975). An Investigation of the Effect upon the Reading Performance and Attitude toward Reading of Elementary Grade Students, of Including in the Reading Program a Period of Sustained, Silent Reading. Unpublished doctoral dissertation, University of Colorado, Boulder.

Yoon, C.-K., Kim, J.-Y., Yi, S.-K., & Yi, H.-S. (2000). A national survey of reading. Seoul, South Korea: Korean Research Center for Publications.
Table 1
Descriptive Statistics on the Effect Size of Reading Attitude

Dependent Variable   Number of Effect Size   Effect Size   Standard
                       (N = Sample Size)                    Error

Attitude                   11 (3183)             .12         .04

Dependent Variable   95% Confidence

Attitude               .05, .19

Table 2
Descriptive Statistics on the Effect Size of the Moderator on
Reading Attitude

Moderator                Level            Number of ES       ES
                                        (N = Sample Size)

Treatment Length   Less Than 6 Months       5 (1327)        .12
                   More Than 6 Months       6 (1856)        .14
Grade *             Below 3rd Grade         7 (2164)        .32
                    Above 4th Grade         4 (1019)        .06

Moderator          Standard   95% Confidence
                    Error        Interval

Treatment Length     .04       .04,  .20
                     .07       .00,  .27
Grade *              .09       .14,  .50
                     .04      -.01,  .14

* p < .05, for the test of heterogeneity

Jun-Chae Yoon
Department of Reading Education
University of Georgia
Athens, GA 30602
Gale Copyright:
Copyright 2002 Gale, Cengage Learning. All rights reserved.