Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Feb 24.
Published in final edited form as: Sci Stud Read. 2015 Feb 24;19(3):224–242. doi: 10.1080/10888438.2015.1007375

Text (Oral) Reading Fluency as a Construct in Reading Development: An Investigation of its Mediating Role for Children from Grades 1 to 4

Young-Suk Grace Kim 1, Richard K Wagner 1
PMCID: PMC4384883  NIHMSID: NIHMS657455  PMID: 25848201

Abstract

In the present study we investigated a developmentally changing role of text reading fluency in mediating the relations of word reading fluency and listening comprehension to reading comprehension. We addressed this question by using longitudinal data from Grades 1 to 4, and employing structural equation models. Results showed that the role of text reading fluency changes over time as children’s reading proficiency develops. In the beginning phase of reading development (Grade 1), text reading fluency was not independently related to reading comprehension over and above word reading fluency and listening comprehension. In Grades 2 to 4, however, text reading fluency completely mediated the relation between word reading fluency and reading comprehension whereas it partially mediated the relation between listening comprehension and reading comprehension. These results suggest that text reading fluency is a dissociable construct that plays a developmentally changing role in reading acquisition.

Keywords: Developmental relations, Listening Comprehension, Oral reading fluency, Reading comprehension, Text reading fluency, Word reading fluency

Reading comprehension is a critical, necessary skill for success in schools and civic life. Theories and empirical studies have shown that reading comprehension is a complex skill drawing from multiple component skills (Cain, Oakhill, & Bryant, 2004; Kendeou & van den Broek, 2007; RAND study group, 2002; Kim, in press; van den Broek, Fletcher, & Risden, 1993). Of these multiple skills, language comprehension and word reading skills are necessary foundations for reading comprehension (i.e., simple view of reading, see Hoover & Gough, 1990; Catts, Adlof, Hogan, & Weismer, 2005; Johnston & Kirby, 2006; Joshi, Tao, Aaron, & Quiroz, 2012; Mancilla-Martinez, Kieffer, Biancarosa, Christodoulou, & Snow, 2011). Recent evidence, however, suggests that ‘text’ reading fluency, fast and accurate reading of connected text, might be another critical component skill for reading comprehension over and above children’s word reading (Jenkins et al., 2003) and listening comprehension (Kim, Park, & Wagner, 2014; Kim, Wagner, & Lopez, 2012; Tilstra, McMaster, van den Broek, & Rapp, 2009).

According to the two-process expectancy theory (Posner & Snyder, 1975) and the interactive-compensatory model (Stanovich, 1980), inefficient word reading causes children to rely on both conscious and automatic activation processes for word identification. Efficient word reading, in contrast, enables readers to rely less on conscious processes (e.g., attention) for word identification. Thus, fast and accurate reading of connected text (text reading fluency) releases readers’ limited cognitive resources such as attention and working memory to be used for higher order comprehension processes, facilitating reading comprehension (LaBerge & Samuels, 1974; Samuels, 2006). A number of studies have shown that text reading fluency is strongly related to reading comprehension with bivariate correlations ranging from .67 to .91 for children in primary grades (e.g., Kim, Petscher, Schatschneider, & Foorman, 2010; National Institute of Child Health and Human Development [NICHD], 2000; Silverman et al., 2013) and somewhat weaker relations for children in upper elementary and secondary schools (e.g., Tilstra et al., 2009; Yovanoff, Duesbery, Alonzo, & Tindal, 2005). Given these relations, perhaps it is not surprising that text reading fluency is widely used in North America as a screening and progress monitoring tool for children in primary grades as a proxy for reading comprehension (Jenkins, Hudson, & Johnson, 2007).

Despite its widespread use, however, our understanding is limited about the nature of developmental relations between text reading fluency and reading comprehension. Therefore, the primary goal of the present study is to address this gap in our understanding. Note that in the present study we use the term, ‘text’ reading fluency, to refer to reading accuracy and rate in ‘connected texts’ or in context, and distinguish it from ‘word’ reading fluency, which is accurate and fast reading of words in isolation or out of context (Jenkins et al., 2003). Although we acknowledge that reading prosody, oral expression during connected text reading, is an important aspect of the definition of reading fluency (e.g., Hudson, Pullen, Lane, & Torgesen, 2009; Kuhn & Stahl, 2003; Kuhn, Schwanenflugel, & Meisinger, 2010), it is beyond the scope of the present study.

Text reading fluency has been characterized as “a bridge” between word reading and reading comprehension (Hudson, Pullen, Lane, & Torgesen, 2009; Kuhn, Schwanenflugel, & Meisinger, 2010; Pikulski & Chard, 2005). In other words, text reading fluency plays a mediating role between word reading and reading comprehension. However, at least two points need further clarification about the hypothesized mediation of text reading fluency. First, this hypothesis assumes that text reading fluency is an extension and consequence of only word reading (Pikulski & Chard, 2005; see NICHD, 2000, for a similar view). However, theoretically, the reason why text reading fluency is dissociable from context-free word reading fluency is because automaticity in word reading allows children to focus on meaning construction (Jenkins et al., 2003; Samuels, 2006; Wolf & Katzir-Cohen, 2001), which is in large part a function of their oral language comprehension. Then, text reading fluency is not just an outcome of word reading but also depends upon children’s language comprehension (Kim et al., 2012, 2014). In other words, according to the text reading fluency theories, text reading fluency should mediate not only the relation between word reading and reading comprehension but also between listening comprehension and reading comprehension at least to some extent. Recent emerging evidence from English-speaking children (Kim, Wagner, & Foster, 2011; Kim et al., 2012) and Korean-speaking children (Kim et al., 2014) indicates that text reading fluency is uniquely predicted by listening comprehension as well as word reading fluency after children have reached a certain level of word reading proficiency.

The second clarification about the mediating role of text reading fluency involves whether the mediation is static or developmental. Although previous studies have been informative, they were limited in examining developmental relations because many were cross-sectional (Kim et al., 2011; Jenkins et al., 2003; Schwanenflugel et al., 2006; Silverman et al., 2013; Tilstra et al., 2009). Furthermore, in some studies reading fluency was operationalized more broadly than text reading fluency, and did not differentiate text reading fluency from word reading fluency. For instance, Silverman et al. (2013) found reading fluency to be independently related to reading comprehension for children in Grade 4, but their reading fluency construct was composed of sublexical (rapid automatized naming), lexical (word reading fluency), and text level (text reading fluency) tasks. In the present study, our primary goal was to examine ‘text’ reading fluency as differentiated from ‘word’ reading fluency, and the relations of word reading fluency and text reading fluency to reading comprehension as children develop reading skills. Differentiating text vs. word reading fluency allows us to enhance our theoretical understanding about what text reading fluency is, including its mediating role.

A developmental perspective of text reading fluency hypothesizes the changing nature of text reading fluency as a function of children’s reading proficiency (Wolf & Katzir-Cohen, 2001).During the beginning phase, children’s primary focus is decoding, and children’s word reading fluency and text reading fluency are very highly related. Therefore, the influence of word reading fluency and text reading fluency on reading comprehension overlaps to a large extent such that text reading fluency does not mediate relations between word reading fluency and reading comprehension. However, as children develop their word reading proficiency, and listening comprehension can contribute to text reading fluency, and text reading fluency might begin to mediate relations of word reading fluency and listening comprehension to reading comprehension. In a recent study, text reading fluency was not independently related to reading comprehension for English-speaking children in first grade, but it was in second grade along with word reading fluency and listening comprehension (Kim et al., 2012). What remains unclear is the developmentally changing nature of relations for a longer developmental span. Therefore, in the present study, we sought to expand our understanding about the developmental nature of relations among word reading fluency, listening comprehension, text reading fluency, and reading comprehension, using longitudinal data from English-speaking children in Grades 1 to 4. The primary research question was whether text reading fluency mediates the relation of word reading fluency and listening comprehension to reading comprehension, and if so, how the mediating role changes over time.

Method

Participants

The sample students were 316 English-speaking children who participated in a four-year longitudinal study from Grade 1 to Grade 4 in Florida. Data in Grades 1 and 2 have been reported in previous studies (Kim et al., 2011, 2012), and the present study extends analysis from Grades 1 to 4. Students’ ages were 85 (SD = 5.67), 95.92 (SD = 5.26), 107.85 (SD = 5.42), and 119.97 (SD = 5.07) months in Grades 1, 2, 3, and 4, respectively. Of 316 (49% female), 270 (49% female) remained in Grade 2, 260 (51% female) in Grade 3, and 219 (52% female) in Grade 4. Most of the observed attrition resulted from several teachers not agreeing for their students to participate in the longitudinal study. Racial and ethnic backgrounds were as follows in the order of Caucasian, African American, Hispanic, Asian, and Others: 60%, 25%, 4%, 4%, and 7% in Grade 1; 62%, 21%, 5%, 4%, and 7% in Grade 2; 61%, 23%, 5%, 4%, and 7% in Grade 3; 65%, 19%, 6%, 5%, and 5% in Grade 4.

Multivariate Analysis of Variance (MANOVA) results showed that children who remained in the study through Grade 4 did not differ from those who left the study in all the measures in Grade 1 except in mean standard scores of WRMT-R and WJ-III Passage Comprehension measures (ps < .001; Wilks’ Lambda = 1.70, F [296, 15] = 1.70, p = .05). The following were mean standard scores of children who stayed in the study and left the study, respectively: M = 109.97 (SD = 10.01) and M = 105.31 (SD = 10.78) in the WRMT-R Passage Comprehension; M = 106.20 (SD = 12.76) and M = 100.30 (SD = 13.04) in the WJ-III Passage Comprehension. These suggest that children who left the study had lower standard scores in the Passage Comprehension in Grade 1 than those who stayed in the study. In the present study, we utilized all available data for analysis in each grade (i.e., N = 316 in Grade 1, 270 in Grade 2, 260 in Grade 3, and 219 in Grade 4), and the Mplus software which was used in the study can handle missing data. When we conducted analysis using only data from children who stayed in the study (stayers, N = 219), the pattern of the results (Appendix A) was essentially identical as that using all available data in each grade (Figure 1). However, the path coefficients of text reading fluency to reading comprehension tended to be slightly larger in Grades 2, 3, and 4 (βs = .47, .49, and .42, respectively) when using data from stayers. We acknowledge that the attrition is a limitation of the study, and therefore, findings should be taken with this caution in mind (see below for further discussion).

Figure 1.

Figure 1

Open in a new tab

Standardized structural regression weights among listening comprehension, word reading fluency, text reading fluency, and reading comprehension. Solid lines represent statistically significant relations and dotted lines statistically nonsignificant relations. In order to reduce the complexity of the illustrations, correlations among word reading fluency and listening comprehension at different time points are not shown in the figure although they were modeled.

Measures

Multiple measures were used to assess the following constructs: listening comprehension, word reading fluency, text reading fluency, and reading comprehension.

Listening comprehension

Three tasks were used: the Woodcock-Johnson III (WJ-III) Oral Comprehension subtest (Woodcock, McGrew, & Mather, 2001) and an experimental task. Oral Comprehension is a cloze task in which children complete orally presented sentences (e.g., People sit in____) and short passages. Cronbach’s alpha estimates ranged from .70 to .71. The Oral Comprehension subtest has been shown to be related to other language skills such as Verbal Comprehension (r = .59) and Story Recall (r = .47) for children six to eight years old (Woodcock et al., 2001). In order to assess the child’s oral language comprehension skills of extended passages such as stories, an experimental measure was developed and used. In the experimental tasks, children listened to two short passages (one narrative and one expository text) read aloud by the assessor and answered four open-ended questions for each passage (a total of eight questions in two passages). The passages were composed of 133 to 230 words. The comprehension questions involved children’s recall of details (e.g., What did Pierre take to the town square?) and inference skills (e.g., Do you think Pierre ever got a job? Why?). Children’s answers were scored dichotomously (1 = correct; 0 = incorrect) for each question. For the questions that had two parts (e.g., Do you think Pierre ever got a job? Why?), the child had to provide correct answer to both parts. The same passages were used in Grades 1 and 2 whereas passages in Grades 3 and 4 were different. The reliability estimates (α) across eight items were .50, .60, .37, and .58 in Grades 1, 2, 3, and 4, respectively. Although these reliability estimates are not ideal, it should be noted that in the latent variable approach, the challenge of somewhat low reliability is ameliorated given that only common variance among measures is used to estimate relations.

Word reading fluency

Two forms (Forms A and B) of the Sight Word Efficiency subtest of the Test of Word Reading Efficiency – 2nd Edition (TOWRE-2, Torgesen, Wagner, & Rashotte, 2012) were used. In these tasks, the child is asked to read aloud as many words as possible within 45 seconds. Words increase in difficulty progressively (e.g., from simple single-syllable to multisyllabic words). Total score is the number of correctly read words within 45 seconds. Test-retest reliabilities have been reported to range from .96 to .97 (Torgesen et al., 2012). The Sight Word Efficiency task has been shown to be highly related to other reading measures of Gray Oral Reading Test-4 reading fluency (r = .91), Test Of Silent Contextual Reading Fluency (r = .75), and WRMT-R Passage Comprehension (r = .88) (Torgesen et al., 2011).

Text reading fluency

A widely used ‘oral’ text reading fluency task were employed, using three grade-level mid-year passages from the Dynamic Indicators of Basic Early Literacy Skills (DIBELS) Text reading fluency (6th edition; Good & Kaminski, 2007). In these tasks, the child was asked to read the passages aloud for one minute and the number of words accurately read during the interval was calculated. Word omissions, substitutions, and hesitations of more than three seconds were scored as errors. Reliabilities have been reported to range from .92 to .97 (Shaw & Shaw, 2002). DIBELS text reading fluency has been shown to be highly related to other reading fluency tasks such as Gray Oral Reading Test-4 (.86 ≤ rs ≤ .88, Hudson et al., 2012) and to reading comprehension (Good, Simmons, & Kame’enui, 2001; Ridel, 2007; Roehrig et al., 2008).

Reading comprehension

Reading comprehension was measured by the WJ-III Passage Comprehension subtest (Woodcock, et al., 2001), Woodcock Reading Mastery Test – Revised Passage Comprehension subtest (WRMT-R; Woodcock, 1987), and an experimental task. Both Woodcock measures are cloze tasks in which the child is asked to read sentences and passages and fill in a correct word based on the texts read. Although the Woodcock tasks have the same format, the items are different in these tasks and are normed on different samples. Cronbach’s alpha estimates for WJ-III and WRMT-R Passage Comprehension tasks ranged from .84 to .93 across grades. Studies have shown that reading comprehension tasks vary in the extent to which they tap into component skills. For instance, cloze tasks such as WJ-III and WRMT-R Passage Comprehension tasks have been shown to tap into word reading to a greater extent than to oral language compared to other reading comprehension tasks (e.g., reading passages and answering questions; Cutting & Scarborough, 2006; Keenan, Betjemann, & Olson, 2008). Therefore, an experimental task was developed and used. In this task, children were asked to read two short passages (one narrative and one expository) and answer four open-ended questions in each passage that required children to recall details in the passage (e.g., Where does Harry live?) and make inferences (e.g., Why did Harry and Sally both see windows made of gold, but at different times of the day?). The passages ranged from 126 to 218 words with a total of eight questions per grade. Children’s answers were scored dichotomously (1 = correct; 0 = incorrect) for each question. Cronbach’s alpha estimates across eight items were .61, .62, .58, and .43 in Grades 1, 2, 3, and 4, respectively. Note that reliability estimates of experimental measures appear to decrease particularly in Grade 4. However, again, given the use of a latent variable approach, only common variance among standardized and experimental measures are used in estimating relations, and thus reducing the impact of lower reliability in Grade 4.

Procedures

All the assessments were individually administered in quiet areas by trained research assistants. Children were assessed in two 30-minute sessions. The assessments were administered at the end of the fall semester and during the spring semester. To minimize time-sampling error, multiple measures of each construct were administered during different testing sessions to the extent possible. In all grades, assessments were administered in the following order: Text reading fluency, WJ-III Passage Comprehension, WJ-III Oral Comprehension; Researcher-developed reading comprehension, TOWRE Sight Word Efficiency, Researcher-developed listening comprehension, and WRMT Passage Comprehension.

Data Analysis

Confirmatory factor analysis and structural equation modeling were employed as the primary analytic strategies using MPLUS 7.0 (Muthen & Muthen, 2013). Preliminary analyses confirmed that univariate and multivariate normality assumptions were met overall despite some floor effects in Grade 1 (see below), and all the measurement models were appropriate. Model fits were evaluated by the following multiple indices: Chi-square statistics, comparative fit index (CFI), the Tucker-Lewis index (TLI), root mean square error of approximation (RMSEA), and standardized root mean square residuals (SRMR). RMSEA values below .08, CFI and TLI values equal to or greater than .95, and SRMR equal to or less than .05 indicate an excellent model fit (Hu & Bentler, 1999). TLI and CFI values greater than .90 are considered to be acceptable (Kline, 2005).

Because the analysis involved a longitudinal sample, measurement invariance was examined following procedures described in Brown (2006) and Thompson and Green (2006). Establishing measurement invariance is important because the indicators that measure the constructs in one time point should be the same as another time point (i.e., factor loadings from observed measures to hypothesized latent variables across time points) to ensure unbiased effects of the measures across time points (Byrne & Watkins, 2003; Kline, 2011).

Results

Descriptive statistics and preliminary analyses

Table 1 presents descriptive statistics (i.e., means, standard deviations, minimum, and maximum) of measures in Grades 1 to 4. The participating children’s mean performances in the language and reading measures were in the average range compared to the norms (e.g., WJ-III Oral Comprehension, TOWRE Sight Word Efficiency, and Woodcock passage comprehension measures). Although not relevant to addressing the research questions, we also report children’s performances on word reading accuracy, measured by the WJ-III Letter Word Identification Task for descriptive purposes. In this task, the child is asked to read aloud increasingly difficult words. Children’s mean standard scores on the Letter Word Identification task tended to be high average ranging from 112.54 in Grade 1 to 107.99 in Grade 4. This compares to average performances on the Sight Word Efficiency tasks with mean standard scores with an exception in Grade 2. Similar trend of higher mean standard scores in the WJ-III Letter Word Identification compared to the Sight Word Efficiency has been reported in previous studies (e.g., Al Otaiba, Kim, Wanzek, Petscher, & Wagner, 2014). Tables 2 (Grades 1 and 2) and 3 (Grades 3 and 4) present bivariate correlations among the observed measures. Relations were all in the expected directions.

Table 1.

Descriptive statistics

First grade Second grade Third grade Fourth grade
Mean (SD) Min - Max Mean (SD) Min - Max Mean (SD) Min-
Max		 Mean (SD) Min-
Max
Listening comprehension
 WJ Oral Comprehension 14.10 (3.31) 5 – 23 17.13 (3.44) 5 – 25 19.36 (3.85) 9 – 28 21.35 (3.86) 8 – 28
 WJ Oral Comprehension (SS) 107.09 (11.07) 82 – 137 108.89 (11.42) 76 – 136 108.68 (12.33) 78 – 136 109.50 (12.05) 71 – 130
 Experimental task 2.50 (1.53) 0 – 7 3.68 (1.80) 0 – 8 5.39 (1.43) 1 – 8 3.93 (1.80) 0 – 8
* Word reading accuracy
 WJ Letter Word Identification 35.78 (7.85) 17 – 57 45.69 (6.78) 26 – 66 51.55 (6.80) 35 – 69 55.19 (6.22) 33 – 70
 WJ Letter Word Identification (SS) 112.54 (13.44) 70 – 143 109.62 (10.63) 72 – 136 107.55 (10.65) 76 – 136 107.99 (10.51) 72 – 136
Word reading fluency
 SWE Form 1 34.98 (16.56) 3 – 70 57.73 (11.29) 12 – 81 64.20 (10.81) 26 – 88 69.23 (9.65) 34 – 92
 SWE Form 1 (SS) 99.72 (17.19) 55 – 144 109.52 (14.00) 11– 138 100.70 (14.32) 56 – 138 101.04 (13.00) 61 – 133
 SWE Form 2 34.94 (16.65) 3 – 70 56.32 (12.08) 12 – 83 64.82 (10.68) 38 – 84 68.72 (10.02) 34 – 97
 SWE Form 2 (SS) 99.46 (17.50) 36 – 142 108.43 (14.21) 12 – 137 101.43 (14.04) 58 – 133 100.48 (13.48) 61 – 138
Text reading fluency
 ORF Passage 1 56.53 (37.14) 0 – 176 108.81 (37.35) 11 – 207 113.66 (37.07) 22 – 235 112.29 (35.09) 27 – 221
 ORF Passage 2 52.85 (35.06) 3 – 142 100.77 (38.03) 9 – 206 102.79 (38.42) 8 – 217 118.55 (30.58) 24 – 205
 ORF Passage 3 52.62 (36.06) 2 – 184 92.72 (30.00) 10 – 197 118.30 (32.40) 19 – 242 145.30 (35.58) 32 – 233
Reading comprehension
 WRMT-R Passage Comp 21.06 (8.52) 1 – 38 29.95 (6.13) 8 – 47 35.28 (5.97) 16 – 51 38.40 (6.01) 16 – 52
 WRMT-R Passage Comp (SS) 108.54 (10.46) 75 – 132 107.89 (10.64) 77 – 193 106.06 (9.78) 81 – 134 105.29 (10.12) 77 – 130
 WJ Passage Comp 18.01 (5.13) 6 – 31 24.46 (4.65) 12 – 35 27.96 (4.25) 14 – 37 29.98 (4.09) 15 – 55
 WJ Passage Comp (SS) 104.38 (13.12) 68 – 136 103.26 (10.40) 72 – 125 102.03 (9.83) 72 – 127 101.72 (9.34) 68 – 126
 Experimental task 3.26 (1.82) 0 – 8 5.05 (1.80) 1 – 8 4.72 (1.88) 0 – 8 3.49 (1.58) 0 –7
Open in a new tab

Note:

*

Children’s performances on the WJ-III Letter Word Identification is reported for descriptive purposes, and were not relevant and thus not used in the analysis. SS = Standard Score; WJ = Woodcock Johnson Tests of Achievement – Third Edition; SWE = Sight Word Efficiency subtest of the Test of Word Reading Efficiency; ORF = Oral Reading Fluency; WRMT-R = Woodcock Reading Mastery Test – Revised

Table 2. Correlations among observed variables for English-speaking children in grades 1 and 2.

1 2 3 4 5 6 7 8 9 10
1. WJ-III Oral Comprehension --- .55 .39 .38 .43 .41 .42 .51 .50 .44
2. Ex listening comprehension .55 --- .22 .23 .27 .27 .26 .35 .39 .37
3. Sight Word Efficiency Form 1 .32 .30 --- .96 .94 .94 .93 .86 .84 .62
4. Sight Word Efficiency Form 2 .33 .20 .92 --- .92 .94 .93 .85 .83 .65
5. DIBELS ORF passage 1 .43 .40 .82 .83 --- .96 .97 .84 .84 .65
6. DIBELS ORF passage 2 .37 .39 .85 .84 .94 --- .96 .83 .83 .64
7. DIBELS ORF passage 3 .41 .38 .84 .85 .92 .92 --- .83 .83 .62
8. WRMT-R Passage comp .56 .50 .61 .59 .69 .65 .66 --- .87 .62
9. WJ-III Passage comp .59 .50 .64 .63 .75 .71 .72 .81 --- .66
10. Ex reading comprehension .52 .45 .39 .37 .49 .44 .46 .65 .62 ---
Open in a new tab

Note: In the top panel, above diagonal in grade one and below diagonal in grade two. In the bottom panel, above diagonal in grade three and below diagonal in grade four. All coefficients are statistically significant at .01 level.

WJ-III = Woodcock Johnson Tests of Achievement – Third Edition; ORF = Oral Reading Fluency; WRMT-R = Woodcock Reading Mastery Test – Revised; Ex = Experimental

Table 3. Correlations among observed variables for English-speaking children in grades 3 and 4.

1 2 3 4 5 6 7 8 9 10
1. WJ-III Oral Comprehension --- .55 .36 .41 .45 .44 .43 .59 .62 .62
2. Ex listening comprehension .56 --- .19 .26 .31 .30 .27 .46 .45 .38
3. Sight Word Efficiency Form 1 .37 .39 --- .89 .80 .77 .78 .58 .56 .43
4. Sight Word Efficiency Form 2 .38 .42 .90 --- .81 .79 .81 .62 .60 .45
5. DIBELS ORF passage 1 .50 .49 .78 .78 --- .93 .92 .72 .69 .55
6. DIBELS ORF passage 2 .44 .46 .79 .80 .92 --- .93 .72 .70 .53
7. DIBELS ORF passage 3 .46 .48 .80 .80 .90 .92 --- .69 .67 .55
8. WRMT-R Passage comp .68 .53 .53 .53 .68 .67 .65 --- .79 .62
9. WJ-III Passage comp .61 .45 .46 .46 .62 .60 .61 .72 --- .59
10. Ex reading comprehension .50 .39 .31 .34 .48 .47 .42 .50 .49 ---
Open in a new tab

Note: Above diagonal in grade three and below diagonal in grade four. All coefficients are statistically significant at .01 level.

WJ-III = Woodcock Johnson Tests of Achievement – Third Edition; ORF = Oral Reading Fluency; WRMT-R = Woodcock Reading Mastery Test – Revised; Ex = Experimental

Using the observed variables, measurement models were fitted for the following latent variables: listening comprehension, word reading fluency, text reading fluency, and reading comprehension. Measurement models were all appropriate and standardized loadings and p values are presented in Table 4. In order to examine whether word reading fluency and text reading fluency are best described as a single construct or they are related but dissociable constructs, confirmatory factor analyses were conducted. Across grades, the models with word reading fluency and text reading fluency as separate latent variables had statistically significantly better fit according to chi-square difference tests (ps < .01). To examine measurement invariance, a baseline model of non-invariance was first specified in which the loadings were allowed to vary completely. This model demonstrated a good fit to the data: (χ2 [620] = 860.90, p < .001; CFI = .98; TLI = .98; RMSEA = .035; SRMR = .035). When a full invariance model was fitted, it had a statistically poorer fit (χ2 [640] = 1973.09, p < .001; CFI = .91; TLI = .89; RMSEA = .081; SRMR = .178) (Δχ2 [Δdf = 20] = 1112.19, p < .001). Thus, we fitted partial invariance models in subsequent analysis by examining the loadings of each observed variable on its corresponding latent variable, and relaxing equal loading constraints for the following variables: the researcher developed listening comprehension measures in all four years, and the researcher developed reading comprehension measures in all four years. Although full measurement invariance is desirable, partial measurement invariance is most common because of subtle changes in measurement properties of indicators (e.g., reliability estimates) when given across multiple years.

Table 4. Standardized loadings and residuals for children in grades 1, 2, 3, and 4.

First grade Second grade Third grade Fourth grade
Loading
(SE)		 Residuals
(SE)		 Loading
(SE)		 Residuals
(SE)		 Loading
(SE)		 Residuals
(SE)		 Loading
(SE)		 Residuals
(SE)
Listening comprehension
 WJ Oral Comprehension .78 (.03) .40 (.05) .81 (.03) .34 (.05) .88 (.03) .23 (.05) .90 (.03) .20 (.05)
 Ex listening comprehension .71 (.03) .50 (.05) .68 (.04) .53 (.05) .63 (.04) .60 (.05) .64 (.05) .59 (.06)
Word reading fluency
 SWE 1 .98 (.003) .03 (.005) .97 (.006) .05 (.01) .94 (.009) .13 (.02) .96 (.008) .08 (.02)
 SWE 2 .98 (.003) .04 (.006) .95 (.008) .11 (.02) .96 (.008) .09 (.02) .94 (.009) .11 (.02)
Text reading fluency
 ORF Passage 1 .99 (.002) .03 (.004) .96 (.006) .08 (.01) .97 (.005) .06 (.009) .95 (.007) .10 (.01)
 ORF Passage 2 .98 (.003) .04 (.005) .95 (.008) .11 (.01) .95 (.007) .10 (.01) .97 (.005) .05 (.009)
 ORF Passage 3 .98 (.002) .03 (.005) .96 (.005) .08 (.01) .96 (.005) .08 (.01) .93 (.009) .13 (.02)
Reading comprehension
 WRMT-R Passage Comp .93 (.009) .13 (.02) .91 (.01) .17 (.02) .90 (.01) .19 (.02) .89 (.02) .33 (.04)
 WJ Passage Comp .93 (.009) .13 (.02) .88 (.02) .22 (.03) .87 (.02) .24 (.03) .82 (.03) .22 (.03)
 Ex reading comprehension .69 (.03) .52 (.04) .71 (.03) .50 (.05) .70 (.03) .51 (.05) .63 (.04) .61 (.05)
Open in a new tab

WJ = Woodcock Johnson-III; Ex = Experimental; SWE = Sight Word Efficiency; ORF = Text reading fluency; WRMT-R = Woodcock Reading Mastery Test-Revised; Ex = Experimental

Correlations among latent variables are presented in Table 5. Overall, the relation between listening comprehension and reading comprehension became stronger from the beginning to the more advanced stage in reading with correlation coefficients increasing from .65 to .89 in Grades 1 to 4. In contrast, the relations between both word reading fluency and text reading fluency and reading comprehension decreased as children advanced in their reading skills from .93 in Grade 1 to .62 in Grade 4. Finally, word reading fluency remained strongly related to text reading fluency across developmental phases (.87 ≤ rs ≤ .96).

Table 5. Correlations between latent variables.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 G1 Listening comp 1
2 G1 Word RF 0.43 1
3 G1 Text RF 0.48 0.97 1
4 G1 Reading comp 0.65 0.93 0.91 1
5 G2 Listening comp 0.99 0.44 0.47 0.65 1
6 G2 Word RF 0.35 0.78 0.74 0.72 0.44 1
7 G2 Text RF 0.47 0.81 0.82 0.78 0.56 0.91 1
8 G2 Reading comp 0.76 0.75 0.75 0.9 0.81 0.71 0.8 1
9 G3 Listening comp 0.92 0.46 0.45 0.64 0.91 0.38 0.49 0.74 1
10 G3 Word RF 0.37 0.68 0.66 0.64 0.47 0.88 0.66 0.66 0.46 1
11 G3 Text RF 0.48 0.73 0.78 0.75 0.56 0.84 0.93 0.79 0.51 0.87 1
12 G3 Reading comp 0.75 0.73 0.73 0.87 0.79 0.69 0.76 0.98 0.79 0.72 0.83 1
13 G4 Listening comp 0.88 0.48 0.5 0.66 0.94 0.45 0.56 0.76 0.89 0.47 0.55 0.75 1
14 G4 Word RF 0.37 0.64 0.62 0.58 0.50 0.85 0.62 0.63 0.41 0.89 0.83 0.64 0.49 1
15 G4 Text RF 0.51 0.73 0.74 0.75 0.59 0.84 0.92 0.78 0.54 0.87 0.95 0.80 0.59 0.88 1
16 G4 Reading comp 0.84 0.62 0.64 0.8 0.87 0.65 0.72 0.95 0.82 0.66 0.76 0.99 0.89 0.62 0.79
Open in a new tab

Note: G = Grade; RF = Reading Fluency; all coefficients are statistically significant at .001 level.

Developmental Relations among Listening Comprehension, Word Reading Fluency, Text Reading Fluency, and Reading Comprehension

Structural equation modeling was conducted to examine the relations of word reading, listening comprehension, text reading fluency, and reading comprehension from Grades 1 to 4. Word reading fluency and listening comprehension were hypothesized to have direct paths to text reading fluency. In addition, text reading fluency was hypothesized to have a direct path to reading comprehension, mediating the relations between word reading fluency and listening comprehension, and reading comprehension. As noted above, partial invariance model was employed, and the model fit was good: χ2 (700) = 1291.56, p < .001, CFI= .96, TLI=.96, RMSEA=.05, SRMR=.055. The results are shown in Figure 1. Note that although all the predictors were allowed to covary across years as well, in order to reduce the complexity of the figure, those correlations are not presented in the figure. Correlation coefficients are as follows across years: .66 ≤ rs ≤ .93 for word reading fluency; .92 ≤ rs ≤ .96 for listening comprehension; and .34 ≤ rs ≤ .81 for text reading fluency. Correlations among reading comprehension across years were constrained to be 1 because they were high ranging from .87 to .99 in bivariate correlations (Table 5).

As shown in Figure 1, word reading fluency was strongly related to text reading fluency in Grades 1, 2, 3, and 4, with modest decline in magnitudes as children developed their reading skills (γ = .93 in Grade 1 to γ = .72 in Grade 4, ps < .001). Listening comprehension was uniquely, but weakly related to text reading fluency in Grade 1 (γ = .08, p = .001), and became more strongly related was in Grades 2, 3, and 4 (γs = .23, .23, and .30, respectively, ps < .001). When it comes to the mediational role of text reading fluency, in Grade 1, word reading fluency (γ = .71, p < .001) and listening comprehension (γ = .33, p < .001) were uniquely related to reading comprehension whereas text reading fluency was not (β = .07, p = .44). In grade 2, listening comprehension was fairly strongly related to reading comprehension (γ = .62, p < .001) whereas word reading fluency was not (γ = .09, p = .32). Text reading fluency was somewhat moderately related to reading comprehension (γ = .37, p < .001) over and above word reading fluency and listening comprehension. In Grades 3 and 4, the pattern of relations was similar to Grade 2 such that text reading fluency was somewhat moderately related to reading comprehension (βs = .35 & .36, ps <.001), and listening comprehension was fairly strongly and strongly related to reading comprehension (γs = .65 & .79, ps < .001).

Discussion

The primary goal of the present study was to investigate the developmental nature of a mediating relation of text reading fluency to reading comprehension. Overall findings support the hypothesis of text reading fluency being “a bridge” to reading comprehension – playing a mediating role in the relation between word reading and listening comprehension, and reading comprehension. However, the results highlight developmental nature of the relations.

Component skills of reading comprehension (word reading fluency, text reading fluency, and listening comprehension) and their relations to reading comprehension changed over time. In Grade 1, reading comprehension was largely explained by word reading fluency, and moderately by listening comprehension, but not by text reading fluency. In Grade 2, listening comprehension became more strongly related to reading comprehension and text reading fluency became independently related to reading comprehension. In fact, this pattern of relations remained in Grades 3 and 4 although listening comprehension became more strongly related to reading comprehension in Grade 4. The increasing role of listening comprehension is convergent with the theoretical hypothesis that word reading has a strong influence on reading comprehension during the beginning phase of reading development while the influence of listening comprehension becomes stronger over time (Hoover & Gough, 1990; Francis et al., 2006). Previous studies showed a similar pattern of increasing influence of listening comprehension on reading comprehension with English-speaking children (Adlof, Catts, & Little, 2006; Joshi et al., 2012; Schatschneider et al., 2004) as well as Korean-speaking children (Kim et al., 2014) and Spanish-speaking children (Joshi et al., 2012).

Importantly, the present study showed that text reading fluency is a differentiated construct from word reading fluency, and is an additional factor that influences reading comprehension over and above word reading fluency and listening comprehension after children have developed a certain level of word reading proficiency – Grade 2 in the present study. We hypothesized that text reading fluency largely develops from word reading fluency (Ehri, 2002; NICHD, 2000), but it also involves meaning-related processes associated with context, and thus involves children’s oral language comprehension. This hypothesis was confirmed such that initially, text reading fluency was highly related to word reading fluency (r = .96 in Grade 1), but with further reading development, text reading became more influenced by listening comprehension (γ = .08 in Grade 1 to γ = .30 in Grade 4) and increasingly dissociated from word reading fluency. This finding is in line with the hypothesis that word reading constrains reading to a large extent in the beginning phase, and as children develop word reading skill, more cognitive resources are available for meaning-related processes (Jenkins et al., 2003; Stanovich, 1980). Coupled with recent studies with English-speaking and Korean-speaking children (Kim et al., 2011, 2012, 2014), the present study provides empirical support that text reading fluency is a separate construct from word reading fluency because it is a function of not only word reading fluency but also listening comprehension. Thus, text reading fluency acts as a bridge not only between word reading fluency and reading comprehension as hypothesized (Pikulski & Chard, 2005), but also between listening comprehension and reading comprehension. Note, however, that the nature of mediation differed for word reading fluency and listening comprehension – text reading fluency completely mediated the relation between word reading fluency and reading comprehension in Grades 2 to 4. In comparison, text reading fluency partially mediated the relation between listening comprehension and reading comprehension across Grades 2 to 4. This indicates that different language and cognitive skills captured in listening comprehension are likely to be involved in text reading fluency vs. reading comprehension (see Kim, under review). Overall, the present findings support the theoretical account of text reading fluency as a component skill of reading comprehension (Fuchs, Fuchs, Hosp, & Jenkins, 2001; Hudson et al., 2012; Kuhn et al., 2010; LaBerge & Samuels, 1974), and the importance of considering developmentally changing nature of relations (Wolf & Katzir-Cohen, 2001). Note, however, that the present findings should be interpreted in the context of how text reading fluency was operationalized – reading accuracy and rate in connected texts. Reading prosody, an important part of the definition of the text reading fluency construct, has been suggested to be implicated in reading comprehension (e.g., Kuhn et al., 2010; Ravid & Mashraki, 2007; Schwanenflugel et al., 2004), but was not included in the present study.

It is worth noting that the relation of text reading fluency to reading comprehension remained stable from Grades 2 to 4 whereas listening comprehension became more strongly related to reading comprehension. These compare interestingly to a recent study with Korean-speaking children, which showed a similar pattern of unique contributions of text reading fluency and listening comprehension for children at a relatively early phase of reading development (i.e., kindergarten), but at a more advanced phase in Grade 1, listening comprehension dominated the relation (γ = .90) and text reading fluency did not make an independent contribution to reading comprehension (Kim et al., 2014). A stronger effect of listening comprehension compared to word reading skill has been hypothesized at an advanced phase of reading development (Hoover & Gough, 1990). However, in English a very strong, dominant relation of listening comprehension to reading comprehension with correlations above .90 were found for readers in 8th grade and college students (Adlof et al., 2006; Gernsbacher, Varner, & Faust, 1990).

One potential explanation for this discrepancy in Korean and English might be differences in orthographic depth as Korean has a relatively transparent orthography whereas English has a deep orthography. Because word reading acquisition occurs at a faster rate in transparent orthographies – Seymour and colleagues’ (2003) study showed that word reading accuracy development takes more than twice the time in English than transparent orthographies -, the changing nature of relations among word reading fluency, listening comprehension, text reading fluency, and reading comprehension might occur at an expedited rate in a transparent orthography compared to English. This speculation can be examined by modeling the nature of relations among word reading fluency, text reading fluency, listening comprehension, and reading comprehension with English-speaking students beyond Grade 4, and with children learning to read in transparent orthographies (see Kim, under review). This would enhance our understanding about generalizability of the present findings and our understanding about orthography-specific vs orthography-general processes and patterns.

Note that text reading fluency was a dissociable construct from word reading fluency in the present study and previous studies (Kim et al., 2014) is discrepant from Adlof et al.’s (2006) study with children in Grades 2, 4, and 8 and Schwanenflugel et al.’s (2006) study with children in Grades 1 to 3. However, these discrepant findings might be due to the number of measures of text reading fluency vs. word reading fluency as Adlof et al.’s and Schwanenflugel’s studies included a single measure of text reading fluency, and thus text reading fluency measures were combined with word reading fluency to represent reading fluency whereas in the present study we had sufficient number of measures (i.e., 3 tasks; Kline, 2011) to test whether word reading fluency and text reading fluency are separate constructs.

The present findings offer some practical implications, albeit preliminary due to correlational nature of the present study. First, the finding that text reading fluency is an important skill for reading comprehension in addition to word reading and listening comprehension suggests that instructional attention is warranted to text reading fluency. Previous studies have shown that instruction on text reading fluency (e.g., repeated reading) improved reading comprehension (Vadasy & Sanders, 2008). Furthermore, the finding that listening comprehension was consistently related to text reading fluency indicates text reading fluency is not just a function of word reading, but oral language comprehension. Previous studies of text reading fluency focused on word reading skill such as repeated reading to improve text reading fluency (see NICHD, 2000), but the present finding suggests that oral language comprehension skill merits instructional attention not only for reading comprehension but also for text reading fluency.

A few limitations and future directions are worth noting. First, some measures suffered from low reliabilities. Although their impacts on path coefficients are reduced in a latent variable approach, a future study with higher reliability of estimates is needed to replicate the present findings. Furthermore, children who did not stay through the study had lower standard scores in the two standardized reading comprehension measures in Grade 1, which limits generalizability of the present findings. As noted above, when analysis was conducted with those who stayed in the study (stayers), the patterns of results were the same as reported in the study. However, the coefficients from text reading fluency to reading comprehension tended to slightly larger for the stayers, compared to when using all the available data in each grade. These results suggest that although the pattern of developmental relations might be similar if “leavers” stayed in the study, the reading ability of the sample might influence the strength of relations, particularly that between text reading fluency to reading comprehension. That is, if leavers stayed in the study, and their reading comprehension was lower than “stayers” in Grades 2, 3, and 4, the relation of text reading fluency to reading comprehension might have been somewhat weaker than what is reported in the present study. We believe that this speculation is partially in line with our developmental hypothesis that the nature of relations change as a function of children’s reading skill level. The stronger relation of text reading fluency to reading comprehension with only “stayers” indicates that as children’s reading skills advance, an independent relation of text reading fluency to reading comprehension is expected. However, future replications with lower attrition rate and/or studies with samples of different reading skill levels are necessary to examine this speculation.

It is also important to note that the present findings were from children whose reading skills were largely in the average range, although differences in mean standard scores were found across measures – performances on the WJ-III Letter Word Identification were in the high average range whereas those on the Sight Word Efficiency were in the average range. Therefore, the present findings may not generalize to children with lower reading abilities, those with reading disability, and/or English-language learners. For instance, Kim (2012) found that for Spanish-speaking English-language learners, despite their equivalent levels of word reading skills to English native speakers in Kim et al. (2011), their listening comprehension was not independently related to text reading fluency whereas listening comprehension was uniquely related to text reading fluency for English native speakers. It was speculated that English-language learners’ lower level of listening comprehension might have constrained utilizing language comprehension in their text reading.

Another limitation includes selection of tasks for target constructs in the study. For the word reading fluency and text reading fluency constructs, we employed widely used tasks for these constructs such as DIBELS oral text reading fluency tasks, but they had the same task formats. Other text reading fluency measures exist (e.g., GORT), and it appears that curriculum based measures of oral text reading fluency such as those in the present study are highly related to GORT (Hudson et al., 2012). In contrast to word reading fluency and text reading fluency, listening comprehension and reading comprehension were measured by various types of tasks. This is partly due to nature of the constructs of listening comprehension and reading comprehension. That is, there is relatively less consensus about how to best measure reading comprehension than word reading fluency and text reading fluency (oral reading fluency in particular), and studies have shown that reading comprehension tasks vary in the extent to which they measure various component skills (Cutting & Scarborough, 2006; Keenan et al., 2008). Therefore, in the present study we attempted to complement standardized measures with experimental measures. Given that latent variables are a function of indicators, future studies could replicate the present study using a greater variety of measures than those in the study.

In summary, despite limitations noted above, the present findings suggest that text reading fluency is an important, independent construct to consider in reading development, and it makes an independent contribution to reading comprehension over and above word reading fluency and listening comprehension. Future studies are warranted to deepen our understanding of the nature and role of text reading fluency in reading development.

Acknowledgements

Part of the research reported here was supported by Grant P50 HD052120 from the National Institute of Child Health and Human Development and by Grant R305A120147 from the Institute of Education Sciences, U.S. Department of Education. The opinions expressed are those of the authors and do not represent views of the Institute. The authors wish to thank participating children, teachers, and schools.

Appendix A

Standardized structural regression weights among listening comprehension, word reading fluency, text reading fluency, and reading comprehension using data from children who stayed through the project (N = 219). Solid lines represent statistically significant relations and dotted lines statistically nonsignificant relations. In order to reduce the complexity of the illustrations, correlations among word reading fluency and listening comprehension at different time points are not shown in the figure although they were modeled.

graphic file with name nihms-657455-f0002.jpg

References

  1. Adams MJ. Beginning to read: Thinking and learning about print. MIT Press; Cambridge, MA: 1990. [Google Scholar]
  2. Al Otaiba S, Kim Y-S, Wanzek J, Petscher Y, Wagner R. Long term effects of first grade multi-tier intervention. Journal of Research on Educational Effectiveness. 2014;7:250–267. doi: 10.1080/19345747.2014.906692. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Brown TA. Confirmatory factor analysis for applied research. Guilford Press; New York, NY: 2006. [Google Scholar]
  4. Cain K, Oakhill J, Bryant P. Children’s reading comprehension ability: Concurrent prediction by working memory, verbal ability, and component skills. Journal of Educational Psychology. 2004;96:31–42. [Google Scholar]
  5. Cutting LE, Scarborough HS. Prediction of reading comprehension: Relative contributions of word recognition, language proficiency, and other cognitive skills can depend on how comprehension is measured. Scientific Studies of Reading. 2006;10:277–299. [Google Scholar]
  6. Johnston TC, Kirby JR. The contribution of naming speed to the simple view of reading. Reading and Writing: An Interdisciplinary Journal. 2006;19:339–361. [Google Scholar]
  7. Joshi RM, Aaron PG. The component model of reading: Simple view of reading made a little more complex. Reading Psychology. 2000;21:85–97. [Google Scholar]
  8. Daane MC, Campbell JR, Grigg WS, Goodman MJ, Oranje A. Fourth-grade students reading aloud: NAEP 2002 special study of oral reading (NCES 2006-469). US Department of Education. Institute of Education Sciences, National Center for Education Statistics. Government Printing Office; Washington, DC: 2005. [Google Scholar]
  9. Fuchs LS, Fuchs D, Hosp MK, Jenkins JR. Text reading fluency as an indicator of reading competence: A theoretical, empirical, and historical analysis. Scientific Studies of Reading. 2001;5:239–256. [Google Scholar]
  10. Gernsbacher MA, Varner KR, Faust ME. Investigating differences in general comprehension skill. Journal of Experimental Psychology: Learning, Memory, & Cognition. 1990;16:430–445. doi: 10.1037//0278-7393.16.3.430. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Good RH, Simmons DC, Kame’enui EJ. The importance and decision-making utility of a continuum of fluency based indicators of foundational reading skills for third-grade high-stakes outcomes. Scientific Studies of Reading. 2001;5:257–288. [Google Scholar]
  12. Hoover WA, Gough PB. The simple view of reading. Reading and Writing: An Interdisciplinary Journal. 1990;2 [Google Scholar]
  13. Hu L-T, Bentler PM. Cutoff criteria for fit indices in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling. 1999;6:1–55. [Google Scholar]
  14. Hudson RF, Pullen PC, Lane HB, Torgesen JK. The complex nature of reading fluency: A multidimensional view. Reading and Writing Quarterly. 2009;25:4–32. [Google Scholar]
  15. Hudson RF, Torgesen JK, Lane HB, Turner SJ. Relations among reading skills and sub-skills and text-level reading proficiency in developing readers. Reading and Writing: An Interdisciplinary Journal. 2012;25:483–507. [Google Scholar]
  16. Jenkins JR, Fuchs LS, van den Broek P, Espin C, Deno SL. Sources of individual differences in reading comprehension and reading fluency. Journal of Educational Psychology. 2003;95:719–729. [Google Scholar]
  17. Jenkins JR, Hudson RF, Johnson ES. Screening for at-risk readers in a response to intervention framework. School Psychology Review. 2007;36:582–600. [Google Scholar]
  18. Joshi RM, Tao S, Aaron PG, Quiroz G. Cognitive component of componential model of reading applied to different orthographies. Journal of Learning Disabilities. 2012;45:480–486. doi: 10.1177/0022219411432690. [DOI] [PubMed] [Google Scholar]
  19. Keenan JM, Betjemann RS, Olson RK. Reading comprehension tests vary in the skills they assess: Differential dependence on decoding and oral comprehension. Scientific Studies of Reading. 2008;12:281–300. [Google Scholar]
  20. Kendeou P, van den Broek P. The effects of prior knowledge and text structure on comprehension processes during reading of scientific texts. Memory and Cognition. 2007;35:1567–1577. doi: 10.3758/bf03193491. [DOI] [PubMed] [Google Scholar]
  21. Kim Y-S. The relations among L1 (Spanish) literacy skills, L2 (English) language, L2 text reading fluency, and L2 reading comprehension for Spanish-speaking ELL first grade students. Learning and Individual Differences. 2012;22:690–700. [Google Scholar]
  22. Kim Y-S, Petscher Y, Schatschneider C, Foorman B. Does growth rate in oral reading fluency matter in predicting reading comprehension? Journal of Educational Psychology. 2010;102:652–667. [Google Scholar]
  23. Kim Y-S, Wagner RK, Foster L. Relations among oral reading fluency, silent reading fluency, and reading comprehension: A latent variable study of first-grade readers. Scientific Studies of Reading. 2011;15:338–362. doi: 10.1080/10888438.2010.493964. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kim Y-S, Wagner R, Lopez D. Developmental relations between reading fluency and reading comprehension: a longitudinal study from grade 1 to grade 2. Journal of Experimental Child Psychology. 2012;113:93–111. doi: 10.1016/j.jecp.2012.03.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kim Y-S, Park C, Wagner RK. Is oral/text reading fluency a “bridge” to reading comprehension? Reading and Writing: An Interdisciplinary Journal. 2014;27:79–99. doi: 10.1007/s11145-013-9434-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kline RB. Principles and practice of structural equation modeling. 2nd. Ed. Guilford; New York: 2005. [Google Scholar]
  27. Kuhn MR, Schwanenflugel PJ, Meisinger EB. Aligning theory and assessment of reading fluency: Automaticity, prosody, and definitions of fluency. Reading Research Quarterly. 2010;45:232–253. [Google Scholar]
  28. Kuhn MR, Stahl SA. Fluency: A review of developmental and remedial practices. Journal of Educational Psychology. 2003;95:3–21. [Google Scholar]
  29. LaBerge D, Samuels SJ. Toward a theory of automatic information processing in reading. Cognitive Psychology. 1974;62:293–323. [Google Scholar]
  30. Mancilla-Martinez J, Kieffer MJ, Biancarosa G, Christodoulou JA, Snow CE. Investigating English Reading Comprehension Growth in Adolescent Language Minority Learners: Some Insights from the Simple View. Reading and Writing. 2011;24:339–354. [Google Scholar]
  31. Muthen LK, Muthen BO. Mplus 7. Muthen and Muthen; Los Angeles: 2013. [Google Scholar]
  32. National Institute of Child Health and Human Development . Report of the National Reading Panel. Teaching children to read: An evidence-based assessment of the scientific research literature on reading and its implications for reading instruction. U.S. Government Printing Office; Washington, DC: 2000. NIH Publication No. 00-4769. [Google Scholar]
  33. Primor L, Pierce ME, Katzir T. Predicting reading comprehension of narrative and expository texts among Hebrew-speaking with and without a reading disability. Annals of Dyslexia. 2011;61:242–268. doi: 10.1007/s11881-011-0059-8. [DOI] [PubMed] [Google Scholar]
  34. Ravid D, Mashraki YE. Prosodic reading, reading comprehension and morphological skills in Hebrew-speaking fourth graders. Journal of Research in Reading. 2007;30:140–156. [Google Scholar]
  35. Ridel BW. The relation between DIBELS, reading comprehension, and vocabulary in urban first-grade students. Reading Research Quarterly. 2007;42:546–567. [Google Scholar]
  36. Roehrig AD, Petscher Y, Nettles SM, Hudson RF, Torgesen JK. Not just speed reading: Accuracy of the DIBELS text reading fluency measure for predicting high-stakes third grade reading comprehension outcomes. Journal of School Psychology. 2008;46:343–366. doi: 10.1016/j.jsp.2007.06.006. [DOI] [PubMed] [Google Scholar]
  37. Samuels SJ. Toward a model of reading fluency. In: Samuels SJ, Farstrup AE, editors. What research has to say about fluency instruction. International Reading Association; Newark, Del.: 2006. pp. 24–46. [Google Scholar]
  38. Schatschneider C, Buck J, Torgesen J, Wagner R, Hassler L, Hecht S, et al. A multivariate study of individual differences in performance on the reading portion of the Florida comprehensive assessment test: A brief report. Florida State University, Florida Center for Reading Research; Tallahassee, FL: 2004. [Google Scholar]
  39. Schwanenflugel PJ, Meisinger EB, Wisenbaker JM, Kuhn MR, Strauss GP, Morris RD. Becoming a fluent and automatic readers in the early elementary school years. Reading Research Quarterly. 2006;41:496–522. doi: 10.1598/RRQ.41.4.4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Schwanenflugel PJ, Hamilton AM, Kuhn MR, Wisenbaker JM, Stahl SA. Becoming a fluent reader: Reading skill and prosodic features in the oral reading of young readers. Journal of Educational Psychology. 2004;96:119–129. doi: 10.1037/0022-0663.96.1.119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Seymour PHK, Aro M, Erskine JM. Foundation literacy acquisition in European orthographies. British Journal of Psychology. 2003;94:143–174. doi: 10.1348/000712603321661859. [DOI] [PubMed] [Google Scholar]
  42. Shaw R, Shaw D. DIBELS oral reading fluency-based indicators of third grade reading skills for Colorado State Assessment Program (CSAP) University of Oregon; Eugene: 2002. [Google Scholar]
  43. Stanovich KE. Toward an interactive-compensatory model of individual differences in the development of reading fluency. Reading Research Quarterly. 1980;16:32–71. [Google Scholar]
  44. Storch SA, Whitehurst GJ. Oral language and code-related precursors to reading: Evidence from a longitudinal structural model. Developmental Psychology. 2002;38:934–947. [PubMed] [Google Scholar]
  45. Thompson MS, Green SB. Evaluating between-group differences in latent variable means. In: Hancock GR, Mueller RO, editors. Structural equation modeling: A second course. Information Age; Greenwich, CT: 2006. pp. 119–170. [Google Scholar]
  46. Torgesen JK, Wagner RK, Rashotte CA. Test of word reading efficiency-2. Pro-Ed.Good &; Austin, TX: 2012. [Google Scholar]
  47. Vadasy PF, Sanders EA. Repeated reading intervention for low-achieving fourth- and fifth-grade students. Remedial and Special Education. 2008;29:235–249. [Google Scholar]
  48. van den Broek P, Fletcher CR, Risden K. Investigations of inferential processes in reading: A theoretical and methodological integration. Discourse Processes. 1993;16:169–180. [Google Scholar]
  49. Woodcock RW. Woodcock Reading Mastery-Revised. American Guidance Services; Circle Pines, MN: 1987. [Google Scholar]
  50. Woodcock RW, McGrew K, Mather N. Woodcock Johnson Tests of Achievement – Third Edition. Riverside Publishing; Itasca, IL: 2001. [Google Scholar]
  51. Wolf M, Katzir-Cohen T. Reading fluency and its intervention. Scientific Studies of Reading. 2001;5:211–239. [Google Scholar]
  52. Yovanoff P, Duesbery L, Alonzo J, Tindal G. Grade-level invariance of a theoretical causal structure predicting reading comprehension with vocabulary and oral reading fluency. Educational Measurement: Issues and Practice. 2005;24:4–12. [Google Scholar]

RESOURCES