Do my home work..

Writing fluency and quality in kindergarten and first grade: the role of attention, reading, transcription, and oral language

Shawn Kent • Jeanne Wanzek • Yaacov Petscher •

Stephanie Al Otaiba • Young-Suk Kim

Published online: 1 September 2013

� Springer Science+Business Media Dordrecht 2013

Abstract In the present study, we examined the influence of kindergarten compo- nent skills on writing outcomes, both concurrently and longitudinally to first grade.

Using data from 265 students, we investigated a model of writing development

including attention regulation along with students’ reading, spelling, handwriting

fluency, and oral language component skills. Results from structural equation mod-

eling demonstrated that a model including attention was better fitting than a model

with only language and literacy factors. Attention, a higher-order literacy factor

related to reading and spelling proficiency, and automaticity in letter-writing were

uniquely and positively related to compositional fluency in kindergarten. Attention

and higher-order literacy factor were predictive of both composition quality and flu-

ency in first grade, while oral language showed unique relations with first grade writing

quality. Implications for writing development and instruction are discussed.

Keywords Component skills � Beginning writing � Kindergarten � First grade � Writing development

S. Kent (&) � J. Wanzek � Y.-S. Kim Florida Center for Reading Research and School of Teacher Education, Florida State University,

2010 Levy Avenue, Suite 100, Tallahassee, FL 32310, USA

e-mail: [email protected]

Y. Petscher

Florida Center for Reading Research, Florida State University, 2010 Levy Avenue, Suite 100,

Tallahassee, FL 32310, USA

S. Al Otaiba

Simmons School of Education and Human Development, Southern Methodist University, University

Park, TX, USA

123

Read Writ (2014) 27:1163–1188

DOI 10.1007/s11145-013-9480-1

Introduction

The act of writing is a vehicle for the expression of knowledge and the transmission

of information across time and generations. Most would agree that the ability to

write is critical for success throughout the school years and into adulthood, where

poor writing skills can have a detrimental impact in the work place (National

Commission on Writing, 2004). Students in the United States have demonstrated

relatively poor writing ability over time (National Assessment of Educational

Progress; Salahu-Din, Persky, & Miller, 2008). The most recent National

Assessment of Education Performance writing test results reported only 33 % of

eighth and 24 % of twelfth grade students exhibited proficient writing skills, a result

essentially unchanged from 2002. While these data relate to adolescents, Juel (1988)

found that students exhibiting difficulty in the area of writing as early as the first

grade were highly likely to remain poor writers in the fourth grade, suggesting that

problems in writing may begin in the early grades.

In recent years there has been a notable increase in expectations for writing skills

of students, beginning in kindergarten and first grade. Among the most recent

writing standards for kindergarten and first grade are expectations for students to

write about experiences, stories, people, and events by composing informational/

explanatory texts with a topic and relevant facts and writing narratives that recount

events sequentially (Common Core State Standards Initiative, 2010). Thus, research

identifying the potential intractability of early writing difficulties, coupled with

heightened expectations for writing, serves to emphasize the need for examining

processes and components that influence writing development in the early

elementary grades in order to provide possible targets of instruction, intervention

and remediation. The present study sought to explore component skills of writing in

the earliest years of schooling, kindergarten and first grade, by specifically

investigating the relations of early literacy and language abilities, as well as

attention, to the development of students’ compositional fluency and quality.

Component skills in early writing development

Some researchers have posited that the ability to write consists of attaining lower-

level, more rudimentary skills such as spelling and handwriting (i.e., transcription

skills) while also being able to utilize higher-level proficiencies such as creating,

organizing, and elaborating ideas (The Simple View of Writing; Juel, Griffith, &

Gough, 1986). More recently, Berninger and colleagues’ Not-So-Simple View of

Writing (Berninger & Winn, 2006) suggests three primary components: (a) tran-

scription; (b) executive functions regulating focused attention, inhibitory control,

and mental shifting during planning, reviewing, and revising of text as well as

strategies for self-regulation; and (c) text generation, or transformation of ideas to

language representations in writing. In addition, reading component skills may also

impact early writing as the two have been called ‘‘two sides of the same coin’’ (Ehri,

2000, p. 19). Thus, one may postulate that early developmental differences in

transcription, language, or reading skills, or attention and self-regulation, could

impact the development of writing.

1164 S. Kent et al.

123

Transcription skills

There is converging evidence that spelling and handwriting are strong predictors of

writing fluency and to some degree, compositional quality as students learn to write

in the primary grades (Graham, Berninger, Abbott, Abbott, & Whitaker, 1997; Jones

& Christensen, 1999; Juel, 1988; Juel et al., 1986). Graham et al. (1997) found that

together, spelling and handwriting accounted for 25 and 66 % of the variance in

compositional quality and fluency, respectively, in Grades 1–3. Individually, after

controlling for language ability, first grader’s spelling accounted for 29 % of the

variance in story writing in comparison to only 10 % for students in fourth grade

(Juel, 1988). Meanwhile, handwriting fluency has been shown to account for just

over one-half of the variance in student’s quality of writing in first grade, even after

accounting for reading (Jones & Christensen, 1999). In their recent meta-analysis,

Graham, McKeown, Kiuhara, & Harris (2012) found that, on average, elementary

students specifically taught transcription skills perform about one-half of a standard

deviation (ES = .55) higher than comparison students on measures of writing

quality. Most recently, Puranik and Al Otaiba (2012) found that in kindergarten,

handwriting fluency and spelling added significant unique variance in predicting

writing fluency after accounting for language, reading, and IQ.

Oral language

Investigating the contribution of oral language, primarily verbal reasoning and fluency,

Abbott and Berninger (1993) found a statistically significant association with writing

quality in first grade and compositional fluency in second and third grade. The

structural relationships among the language systems of expressive language and

writing have been found to generally change across development in Grades 1–3 with

evidence suggesting strengthening relations over time (Berninger et al., 2006).

However, \25 % of the variance in writing achievement was accounted for by expressive language ability. Specific to kindergarten, Hooper et al. (2011) found that

expressive and receptive language proficiency longitudinally predicted narrative

writing skill in later elementary grades (third to fifth grades). Further, word and syntax-

level language skills in kindergarten have demonstrated unique concurrent relations to

compositional fluency after accounting for other literacy skills (Kim et al., 2011).

Reading

Given the multidimensional nature of both reading and writing processes,

researchers have posited their important and reciprocal relationship may be due

to shared knowledge (e.g., metaknowledge, knowledge of text attributes) and

cognitive processes such as phonological and orthographic systems and short and

long-term memory (Fitzgerald & Shanahan, 2000; Shanahan, 2006). It has been

demonstrated that word and sentence-level reading ability contribute directly to

spelling, handwriting, and compositional quality in Grades 1–6 and to composi-

tional fluency in Grades 1–3, with the strongest relationship for fluency evident in

first grade (Abbott & Berninger, 1993; Berninger, Abbott, Abbott, Graham, &

Predictors of kindergarten and first grade writing 1165

123

Richards, 2002). Further, Abbott, Berninger, and Fayol (2010) found a longitudinal

relationship between reading, specifically comprehension, and a measure of fluency

and quality of writing across Grades 2–5.

Two recent studies have investigated several of these potential component skills of

writing specifically for students in kindergarten (Kim et al., 2011; Puranik & Al

Otaiba, 2012). Kim et al. (2011) findings supported a model of oral language, spelling,

and letter-writing fluency, although reading had a non-significant relation to writing

fluency after accounting for other skills. However, overall, only 33 % of the variance

in writing production was accounted for with these variables. Similarly, Puranik and

Al Otaiba (2012) accounted for 39 % of the variance in writing fluency with

handwriting, spelling, reading, oral language, as well as cognitive and demographic

factors. Both findings suggest the need to examine other component skills.

Regulation of attention

As theorized in the Not-so-Simple-View of writing, executive functions, including

supervisory attention, goal setting and planning, reviewing and revising, and

strategies for self-monitoring and regulation, are critical to proficiency in text

generation and are posited to increase in importance throughout development as the

complexity of writing increases in schooling (Berninger & Winn, 2006). More

specifically, Berninger and Winn view the supervisory attention system as

responsible for selective attention during writing tasks; namely, focusing attention

on relevant aspects of the task while inhibiting attention to non-relevant aspects,

allowing shifting between mental sets, remaining on task, and metacognitive

awareness. Specific executive functions that contribute to selective attention have

been identified including inhibitory control, set-shifting, and updating of memory

(Lehto, Juuarvi, Kooistra, & Pulkkinen, 2003; Willcutt et al., 2001). Similarly,

cognitive research from Happaney, Zelazo, and Stuss (2004) identified cognitive

and attentional flexibility, and inhibitory control, along with working memory, as

important to individual self-regulation. Hooper, Swartz, Wakely, de Kruif, and

Montgomery (2002) have argued that problems with attentional control may

particularly interfere with executive processes that coordinate strategic writing-

planning, monitoring, and revising of writing. Barkley (1996) has hypothesized that

inattention is a symptom of difficulty with self-regulation of internal cognitive

processes. Given that the foundation for self-regulatory behavior occurs early in life,

most often in the first 5 years (Blair, 2002), one would expect to see the influence of

a self-regulatory skill such as attentional control, in the earliest grades.

Several studies have demonstrated the impact of attention on early academic

outcomes, such as reading and math achievement, in preschool and early elementary

grades (Duncan et al., 2007; McClelland, Acock, Piccinin, Rhea, & Stallings, 2013;

Rhoades, Warren, Domitrovich, & Greenberg, 2011). One might argue that the ability

to regulate attention would also impact early writing, as it allows students engaged in a

writing task to specifically attend to relevant tasks and keep previous content written

in mind for use in subsequent composing, all while disregarding extraneous

information. With regards to writing, Hooper et al. (2002) found that less competent

writers in fourth and fifth grade, in comparison to more able writers, demonstrated less

1166 S. Kent et al.

123

proficiency with initiating and sustaining attention, inhibitory control, and set

shifting. Subsequently, Hooper et al. (2011) identified a model of writing in first and

second grade, including attention, memory, and executive functions, language, and

fine motor ability, that accounted for nearly 50 % of the variance in outcomes on a

standardized measure of writing achievement. Attention/executive function was the

only unique predictor of first grade writing while both language and attention/

executive function uniquely predicted second grade writing. Teacher ratings of

student attentiveness have also been shown to predict substantive quality of writing

and writing convention as early as first grade (Kim, Al Otaiba, Folsom, & Greulich,

2013). Meanwhile, Thomson et al. (2005) have suggested that attention plays an

indirect role in writing via influence on orthographic coding and rapid naming skills.

Intervention research has demonstrated that training student’s attentional processes

(e.g., sustained, selective, and alternating attention), coupled with writing instruction,

significantly improves compositional skills of students in fourth through sixth

(Chenault, Thomson, Abbott, & Berninger, 2006). Further, many self-regulatory

strategies utilized during writing, which focus student’s attention on the task, have

been postulated including goal setting and planning, organizing, self-monitoring, self-

verbalizing, and revising (Graham & Harris, 2000). The potential importance of such

strategies during writing provided impetus for the development of the self-regulated

strategy development approach, which subsequently has been proven effective across

both grade and writing performance levels (e.g., Graham, Harris, & McKeown, 2013;

Graham, McKeown, Kiuhara, & Harris, 2012; Graham & Perin, 2007).

Study purpose and research questions

Although research interest in the area of writing has grown, the extant literature is

still sparse at early developmental levels; only a few have specifically examined

students as young as kindergarten (Kim et al., 2011; Puranik & Al Otaiba, 2012).

While several component skills involved in writing development have been

investigated and their potential importance demonstrated, most studies have

examined only one or two of these component skills rather than the contribution

of multiple component skills on writing. Although recent multivariate research of

component skills at the early grades is encouraging (Hooper et al., 2011; Kim et al.,

2011, 2013; Puranik & Al Otaiba, 2012), no study has yet examined the unique and

shared role of attention, transcription, reading, and language ability in a model of

writing development at the kindergarten level nor has the contribution of these skills

been examined longitudinally in kindergarten and first grade.

Given the importance of language and literacy and attention skills to early

learning, the purpose of the present study was to build upon and extend the extant

literature on early writing development by examining four primary research

questions (RQ). First,

1. What are the shared and unique relations of component skills of writing fluency in

kindergarten by including attention, reading, transcription, and oral language skills?

2. Does a model of kindergarten writing development with an attention factor fit

better than one without?

Predictors of kindergarten and first grade writing 1167

123

3. What are the longitudinal relations of kindergarten component skills to writing

quality and fluency in first grade?

4. Do kindergarten component skills exhibit a direct effect on writing quality and/

or fluency in first grade or are these relationships mediated by kindergarten

writing fluency?

Method

Sample characteristics

The present study examined extant data from a larger, 5-year project examining a

response to intervention framework for reading, including examination of general

education classroom instruction for all students, within a school setting (Al Otaiba

et al., 2011). Participants in the present analysis included one cohort of students who

participated in the project in both kindergarten and first grade (n = 265).

In kindergarten, students represented 10 schools and 31 classrooms with a range of

1–15 students per participating classroom. The mean age in fall of kindergarten was

5.13 years of age. Sixty-one percent of the students qualified for free or reduced lunch

programs, indicating a large proportion of the sample may be classified as low socio-

economic status. The sample included 53.6 % male students. Students represented a

diversity of races/ethnicities, with 54 % African American, 31 % Caucasian, 8 %

Hispanic, 4 % multi-racial, and \1 % of Asian or Native American. Data on special education status were available for 253/265 (95 %) students with approximately 11 %

eligible for services; most (over 70 %) were identified with either a speech or

language impairment. As demographic data was collected from schools, we were

unable to obtain specific information relative to medical diagnoses such as Attention-

Deficit Hyperactivity Disorder (ADHD). However, in this representative sample of

students, only two (\1 %) students were eligible for special education as Other Health Impairment, a category often comprised of students with ADHD.

Measures

Transcription skills

Student’s spelling and letter writing automaticity/fluency (accuracy and fluency in

writing individual letters) were both assessed. On the spelling subtest from the WJ-III

students write the corresponding graphemic representation of orally presented letters or

words. Responses are scored dichotomously (correct or incorrect) on this measure.

Median split-half reliability is .90. On a separate, untimed spelling measure students

were presented 10 decodable or high-frequency words (e.g., dog, man, one, come) and 4

nonwords (e.g., sut, frot) used in previous literacy research (Byrne & Fielding-Barnsley,

1993). Using a standard protocol from Byrne et al. (2005), examiners explained to the

students that some presented words were real and some were nonwords. All real words

were presented orally, used in a sentence, and then repeated, while nonwords were

1168 S. Kent et al.

123

repeated three times. Student performance on each word was assigned a development

score ranging from 0 to 6 based on modification of a rubric from Tangel and Blachman

(1992). A developmental score, in contrast to a dichotomous score was chosen given the

age of the sample in order to capture differences in how students orthographically

represented phonological word features. This scoring method has been utilized in recent

studies with kindergarten students and in previous research (Jones & Christensen, 1999;

Kim et al., 2011; Puranik & Al Otaiba, 2012). The following guidelines were utilized: no

response or a random string of letters was assigned a 0, providing at least one letter that

was related to the target spelling word phonetically was scored as a 1 (e.g., writing an ‘‘o’’

or ‘‘g’’ for dog), writing the correct initial letter followed by random letters was scored as

a 2, including more than one correct phoneme resulted in a score of 3, a score of 4 was

given for words spelled with all letters represented and phonetically correct (e.g.,

‘‘dawg’’ for dog), a 5 was assigned when all requirements of a score of 4 were met and

student made an attempt to mark a long vowel (e.g., ‘‘bloo’’ for blue), and a word

received a score of 6 when spelled correctly.

Student’s ability to access, retrieve, and automatically write letter forms was also

measured. The task was modified from Berninger and Rutberg’s (1992) task of

handwriting automaticity. Students were given 1 min to write, quickly and

accurately, all lower case letters of the alphabet in order (Jones & Christensen,

1999; Wagner et al., 2011). Consistent with prior research, the letter writing fluency

(LWF) task was scored primarily on penmanship and correct letter formation with a

score of 1.0 given for a letter that was correctly formed and sequenced, .5 for letters

poorly formed yet recognizable and/or reversed, and 0 for illegible letters, cursive

letters, letters written out of order, or uppercase letters.

Oral language

Student’s oral language ability was assessed using measures of word and syntax

knowledge. Expressive vocabulary was assessed using the Picture Vocabulary

subtest of the WJ-III (Woodcock, McGrew, & Mather, 2001). Students were

presented with pictures of common objects and asked to say the word depicted in the

picture. The reported median split-half reliability was .77. Knowledge of syntax, or

grammar, was measured by the Grammatic Completion subtest of the Test of

Language Development—Primary, third edition (TOLD-P: 3; Newcomer & Hamill,

1997). This task, measures the ability to use various morphological forms found in

English (e.g., ‘‘Here is a cat. Over there are four more ___.’’). The Sentence

Imitation subtest of the TOLD-P: 3 (30 items), measuring auditory short-term

memory and grammatical understanding, was also administered. Students were

asked to repeat sentences that increase in length and complexity. Reliability was

reported to be .90 and .91 respectively, for these subtests.

Reading skills

Three measures of letter and word reading and decoding were utilized as indicators of

student’s reading ability in kindergarten. On the Woodcock-Johnson (WJ-III) Letter-

Word Identification subtest (Woodcock, McGrew, & Mather, 2001), which includes

Predictors of kindergarten and first grade writing 1169

123

76 items increasing in difficulty, students were required to name individual letters, as

well as decode and/or identify real words presented. Median split-half reliability was

reported to be .94 for this measure. Decoding skill was measured using the WJ-III

Word Attack subtest. Utilizing pseudowords, items proceed from identification of a

few single letter sounds to decoding of complex letter combinations. Reported median

split-half reliability was .87. The ability to decode phonetically regular words fluently

and accurately was assessed using the Phonemic Decoding Efficiency (PDE) subtest of

the Test of Word Reading Efficiency (TOWRE; Torgesen, Wagner, & Rashotte,

1999). The PDE measures the number of pronounceable nonwords accurately decoded

within 45 s. Reported test–retest reliability was .90.

Attention

Student’s attention regulation was assessed through teacher report using the

Strengths and Weaknesses of ADHD-symptoms and Normal behavior scale

(SWAN; Swanson et al., 2006). This 30-item scale, assessing students’ attention-

based behavior was completed by classroom teachers. Some example items include

‘‘Engage in tasks that require sustained mental effort’’ and ‘‘Remain focused on

task’’. Students are rated in comparison to their peers along a 7-point continuum that

ranges from ‘‘far below’’ to ‘‘far above’’, based upon observations made over the

past month (Swanson et al., 2006). The Diagnostic and Statistical Manual (DSM-IV;

American Psychiatric Association, 1994) criteria used for identifying individuals

with Attention Deficit Hyperactivity Disorder was utilized in the development of the

SWAN. Although the SWAN is frequently used as one aspect in an ADHD

evaluation, it also provides a direct measure of student’s attention skills across a

continuum. Recent research (Saez, Folsom, Al Otaiba, & Schatschneider, 2012) has

supported the items on the SWAN as representing three separable factors of

selective attention, including attention-memory, attention-set shifting, and attention-

inhibitory control, which align with the specific components of self-regulation

described by previous research (Happaney et al., 2004; Lehto et al., 2003; Willcutt

et al., 2001). The mean score across the items representing each of these three

factors was calculated to form three indicators of selective attention. Cronbach’s

alpha for this measure across all 30 items was .99.

Writing skills

In the spring of kindergarten, students were asked to compose a writing sample in

response to an examiner-presented prompt. Examiners introduced the task and

facilitated a brief group discussion to orient students to expectations. Specifically,

examiners said, ‘‘You have been in kindergarten for almost a whole year. Today we

are going to write about kindergarten. Let’s think about what you enjoyed about

being in kindergarten. What did you learn in school? Did anything special happen to

you in kindergarten?’’ These questions and student responses were not written down

by examiners. Examiners also instructed students to write what they had learned in

kindergarten until instructed to stop and that they could not receive help with

spelling words. Students were allowed 15 min to complete the task. This specific

1170 S. Kent et al.

123

researcher-developed task was developed in response to the lack of validated

measures to assess students writing in kindergarten.

Students’ writing was scored for the number of words, sentences, and ideas using

a coding scheme developed by Puranik, Lombardino, and Altmann (2007). Due to

limited production, scoring of correct word sequences (CWS) was not suitable in

kindergarten. All actual words, explicitly related to the prompt, that students

composed were counted in the computation for the number of words written (e.g.,

‘‘The end’’ was not counted), while number symbols and random letters were not

counted. All complete sentences were counted in the total number of sentences,

regardless of punctuation. When punctuation was absent, raters broke the writing

into sentences with the criteria that a sentence had to express a complete thought,

feeling, or idea and have an explicit or implied subject and predicate with a verb

(e.g., ‘‘I eat and play [STOP] and go outside and laugh [STOP] and draw pictures.’’

counted as three sentences). For calculating the number of ideas in a student’s

writing sample, only ideas that could be identified in the writing were counted, Ideas

required a subject and a predicate, but could use a common subject/verb (e.g., ‘‘I

like playing’’ would be scored as one idea; ‘‘I like playing and writing’’ would count

as two ideas). Repeated ideas were only counted once. These scoring measures for

writing fluency (i.e., production) have been utilized in previous studies with students

in the earliest grades (Kim et al., 2011, in press).

In the spring of first grade, students were asked to compose a brief narrative text

when presented with a story prompt (i.e., One day when I got home from school …) developed by curriculum based writing researchers (McMaster, Du, & Petursdorrir,

2009). Brief 5 min prompts are widely used in writing research as global indicators

of writing performance in first grade and also through the elementary grades

(Lembke, Deno, & Hall, 2003). Students’ writing was evaluated in two different

ways. First, analytic scoring of student essays for the following components was

completed: organization of text structure (e.g., beginning, middle, and end), ideas

(e.g., development of main idea), word choice (e.g., use of specific/interesting

words), and sentence fluency (e.g., grammatical use of sentences and sentence flow).

These components of writing were adapted from the widely utilized 6 ? 1 Traits of

Writing Rubric for Primary Grades (NREL, 2011). These four components were

recently identified as representing a separable factor in students’ writing, capturing

substantive quality (Kim et al., in press). Second, writing was evaluated for the

number of CWS, a commonly used metric within curriculum-based writing

measures; in the present study, CWS serves as a proxy for writing fluency in first

grade. For writing quality each domain was assigned a score from 1 (‘‘experiment-

ing’’) through 5 (‘‘experienced’’) based on the degree of proficiency exhibited in the

writing. A score of 0 was assigned for each trait if the student’s writing sample was

unscorable due to being illegible, the student did not produce any writing, or the

student simply rewrote the prompt. For CWS, a word sequence was defined as the

sequence between two adjacent words or between a word and punctuation mark. For

a sequence to be scored as a CWS, the adjacent words must be spelled correctly and

be syntactically and semantically correct within the writing context. To take into

account the beginning and end of sentences, the beginning word must be capitalized

and the end of the sentence properly punctuated in order to receive a CWS for those

Predictors of kindergarten and first grade writing 1171

123

writing sequences. All CWS in the student’s writing were summed to create a total

score.

Procedures

Reading, spelling, and letter writing fluency assessments for the current study were

collected during spring of kindergarten as was the WJ-III Picture Vocabulary

measure. The additional oral language measures from the TOLD-P: 3 were part of

the fall of kindergarten assessment battery. Writing samples were collected at the

end of both kindergarten and first grade. Trained research assistants (RA) served as

examiners. Reading and oral language assessments were administered individually,

whereas spelling and writing assessments were done in small groups or entire

classrooms. Letter-writing fluency, spelling, and writing quality were all scored by

trained research assistants using specific rubrics.

For LWF, spelling, and writing scoring, RAs were trained to use each rubric on a

small subset of the sample through practice and discussion of scoring issues. A

randomly selected 15 % of the complete data set for LWF and spelling were scored

by each RA individually to calculate reliability. Kappa was .98 for letter-writing

fluency and .99 for spelling. For writing samples, interrater reliability for each

variable coded was calculated on a random selection of 20 % of writing samples.

For writing scoring in kindergarten, agreement averaged 88 % for total number of

words, sentences, and ideas. For first grade samples, inter-rater agreement was 92 %

for CWS and averaged 91 % across the quality components. All discrepancies in

scoring were resolved through discussion.

Data analysis

Confirmatory factor analysis (CFA) and structural equation modeling (SEM) were

utilized to answer the research questions. Initially, the appropriateness of the

measurement model was established utilizing CFA including latent factors of

reading, spelling, letter-writing fluency, oral language, attention, and writing

production in kindergarten and writing quality in first grade. While all observed

measures of most latent constructs were collected in spring, students’ scores from

the TOLD-P: 3 subtests collected in fall of kindergarten were utilized to create an

oral language factor. Correlations between the TOLD-P:3 subtests in fall and a

vocabulary measure in the spring ranged from .39 to .49 and were statistically

significant. Next, a series of SEMs were specified to address the research questions:

Model 1—Kindergarten component factors and a single-indicator of letter writing

fluency predicting kindergarten writing fluency (research question 1); Model 2—

Same as model 1 but with all covariances with, and direct paths from, the attention

factor constrained to 0 (RQ 2); Model 3—Kindergarten component skills predicting

first grade writing quality and writing fluency (RQ 3); and Model 4—kindergarten

factors predicting kindergarten writing production and direct and indirect effects on

first grade writing quality and fluency (RQ 4). Measurement error for single

indicators of letter-writing fluency and the writing production fluency in kinder-

garten was accounted for by fixing the residual variances of each observed variable

1172 S. Kent et al.

123

(i.e., [1 - reliability] 9 r2). Multiple indices were evaluated to assess model fit including Chi square, root mean square error of approximation (RMSEA),

comparative fit index (CFI), Tucker-Lewis Index (TLI), and standardized root

mean square residuals (SRMR). Given that Chi square values tend to be influenced

by sample size, RMSEA values below .085, CFI/TLI values greater than .95, and

SRMR below .05 indicate excellent model fit (Kline, 2011).

To account for the non-independence of observations, cluster-corrected standard

errors using CLUSTER option and TYPE = COMPLEX in Mplus 6.1 were derived.

This approach is an appropriate method for accounting for the nested nature of data

without specifically answering questions about variance components at different

levels (Asparouhov, 2006). However, for analyses purposes, clusters (i.e.,

classrooms) with \5 students were removed resulting in removal of seven clusters and 10 students. There were no statistically significant differences in sample means

for any variable with removal of these 10 cases.

As the extant data utilized in these analyses was collected on students in school,

missing data was to be expected. Complete data were available for 195 students, or

close to 75 % of participants. Missing data was treated as missing at random, which

Mplus handles using full-information maximum likelihood estimation.

Results

Descriptive statistics

Descriptive statistics are presented in Table 1 and where available, standard scores

are provided in addition to raw scores. Students in the sample demonstrated mean

scores within the average range for word reading, decoding, spelling, and expressive

vocabulary, while grammar/syntax skills were low average to below average.

Sample means for each area of attention were consistent (M = 4.59–4.69).

Kindergarten writing production averaged almost 13 words (0–50). Approximately

one in five (22 %) kindergarten writing samples were either unscorable due to

illegibility or the student was unable to compose any actual words suggesting the

presence of a floor effect on this measure. In first grade, students averaged 32 words

written (SD = 14.5) and 21 CWS (SD = 12.4) for the narrative writing prompt.

Mean ratings on each of the first grade writing quality components ranged from a

low of 2.45 on Word Choice to a high of 2.95 for Ideas with only three prompts

deemed unscorable. Table 2 presents bivariate correlations among observed

variables, all of which were statistically significant (p \ .01).

Measurement model

Initial evaluation of the fit indices for the proposed measurement model indicated

excellent data fit: v2 (119) = 234.62, p = .000; CFI = .963; TLI = .952; RMSEA = .060 (CI [.049, .072]); and SRMR = .048. All predictor factors were

significantly and positively related to one another (.33 B r B .95) and with writing

in kindergarten and first grade, with generally moderate correlations (see Table 3).

Predictors of kindergarten and first grade writing 1173

123

Reading and spelling factors were very highly related (r = .95). Given approxi-

mately 90 % of the variance in one factor was accounted for by the other suggested

that the two constructs may be captured by a second-order factor (e.g., global early

literacy skills). Previous research has suggested a unitary language and literacy

construct at first grade (Mehta, Foorman, Branum-Martin, & Taylor, 2005).

Table 1 Descriptive statistics

M SD Range

Kindergarten writing fluency

Total number of words a

12.64 11.3 0–50

Total number of ideas a

2.23 2.4 0–11

Total number of sentences a

1.7 2.1 0–11

Writing quality

Ideas b

2.95 .72 0–4

Structure b

2.77 .66 0–4

Word choice b

2.45 .73 0–4

Grammar b

2.78 .62 0–4

First grade writing fluency

Correct writing sequences c

20.80 12.4 1–61

Reading

Letter word identification-raw score d

23.99 7.8 11–52

Letter word identification-standard score d

108.81 14.2 80–149

Word attack-raw score e

7.64 5.2 2–27

Word attack-standard score e

112.25 13.0 78–141

Phonemic decoding efficiency-raw score e

7.05 7.9 0–43

Spelling

WJ spelling-raw score e

16.28 3.9 2–29

WJ spelling-standard score e

104.92 14.6 31–137

Byrne spelling task f

49.06 18.3 0–82

Letter writing fluency g

10.46 5.5 .5–25

Oral language

Picture vocabulary-raw score d

18.19 2.8 12–28

Picture vocabulary-standard score d

101.41 9.7 76–140

TOLD sentence imitation-raw score h

8.35 5.9 0–28

TOLD sentence imitation-standard score h

8.29 3.0 2–20

TOLD grammatic completion-raw score h

6.60 5.5 0–24

TOLD grammatic completion-std. score h

7.61 2.9 1–17

Attention

SWAN-memory i

4.59 1.6 1–7

SWAN-set shifting i

4.70 1.7 1–7

SWAN-inhibitory control i

4.49 1.6 1–7

a n = 238,

b n = 265,

c n = 264,

d n = 253,

e n = 252,

f n = 232,

g n = 214,

h n = 257,

i n = 249

1174 S. Kent et al.

123

T a

b le

2 C

o rr

e la

ti o

n s

b e tw

e e n

o b

se rv

e d

v a ri

a b

le s

1 2

3 4

5 6

7 8

9 1

0 1

1 1

2 1

3 1

4 1

5 1

6 1

7 1

8 1

9 2

0

R e a d

in g

1 .

L W

ID –

2 .

W A

.8 1

–

3 .

P D

E .8

4 .8

4 –

T ra

n sc

ri p ti

o n

4 .

W J

sp e ll

in g

.8 4

.7 5

.7 3

–

5 .

S p

e ll

in g :

in fo

rm a l

.7 7

.6 9

.6 5

.7 7

–

6 .

L W

F .4

6 .4

1 .4

1 .5

0 .4

8 –

O ra

l la

n g

u a g

e

7 .

P V

.4 2

.4 7

.4 1

.3 7

.3 8

.2 4

–

8 .

T O

L D

-S I

.3 6

.3 9

.3 6

.3 2

.4 0

.2 2

.3 9

–

9 .

T O

L D

-G C

.4 2

.4 5

.3 8

.3 4

.4 2

.3 6

.4 9

.6 0

–

A tt

e n

ti o

n

1 0

. S

W A

N :

m e m

o ry

.4 7

.4 2

.4 0

.4 4

.5 4

.4 6

.1 9

.2 6

.3 7

–

1 1

. S

W A

N :

se t

sh if

t .3

8 .3

3 .3

4 .3

2 .3

8 .3

2 .1

5 .1

8 .2

4 .7

1 –

1 2

. S

W A

N :

in h

ib it

o ry

c o

n tr

o l

.3 7

.3 1

.3 1

.3 3

.4 4

.3 7

.1 5

.2 1

.3 0

.8 3

.8 3

–

K fl

u e n

c y

1 3

. W

o rd

s .6

0 .5

0 .5

3 .5

9 .6

2 .4

8 .1

6 .2

9 .2

7 .4

4 .4

1 .3

9 –

1 4

. S

e n

t. .4

4 .3

7 .4

0 .4

4 .5

1 .4

1 .1

5 .2

1 .2

6 .4

2 .3

7 .3

9 .8

2 –

1 5

. Id

e a s

.5 2

.4 7

.4 9

.5 2

.5 8

.4 7

.2 0

.2 6

.2 9

.4 2

.3 6

.3 8

.8 7

.9 0

–

F ir

st g

ra d

e q

u a li

ty

1 6

. Id

e a s

.2 8

.2 7

.2 6

.3 1

.3 4

.2 4

.1 7

.2 3

.2 4

.2 3

.1 6

.2 1

.2 6

.2 3

.2 5

–

1 7

. S

tr u

c tu

re .3

3 .2

6 .2

8 .3

4 .3

4 .2

0 .2

0 .1

5 .1

9 .2

7 .1

7 .2

6 .2

7 .2

0 .2

2 .5

6 –

1 8

. W

o rd

c h

o ic

e .2

8 .2

7 .2

7 .3

3 .3

2 .2

1 .2

3 .2

0 .2

4 .2

6 .1

8 .2

6 .2

4 .2

3 .2

1 .5

0 .4

6 –

Predictors of kindergarten and first grade writing 1175

123

T a

b le

2 c o

n ti

n u

e d

1 2

3 4

5 6

7 8

9 1

0 1

1 1

2 1

3 1

4 1

5 1

6 1

7 1

8 1

9 2

0

1 9

. G

ra m

m a r

.3 9

.3 2

.3 5

.4 1

.4 4

.2 8

.1 7

.2 9

.3 5

.4 0

.2 9

.3 5

.3 3

.2 7

.3 0

.5 7

.6 1

.4 6

–

F ir

st g

ra d

e fl

u e n c y

2 0

. C

W S

.6 0

.4 4

.4 9

.5 4

.5 3

.3 3

.1 6

.3 0

.2 5

.4 6

.4 0

.4 2

.5 1

.4 6

.4 7

.3 7

.3 6

.2 7

.5 8

–

A ll

c o

e ffi

c ie

n ts

a re

st a ti

st ic

a ll

y si

g n

ifi c a n

t a t

.0 5

le v

e l

L W

ID le

tt e r-

w o rd

id e n ti

fi c a ti

o n ,

W A

w o

rd a tt

a c k

, P

D E

p h

o n e m

ic d

e c o

d in

g e ffi

c ie

n c y

, L

W F

le tt

e r

w ri

ti n

g fl

u e n c y

, P

V p

ic tu

re v

o c a b

u la

ry ,

T O

L D

-S I

te st

o f

la n

g u

a g

e

d e v

e lo

p m

e n

t se

n te

n c e

im it

a ti

o n

, T

O L

D -G

C te

st o

f la

n g

u a g

e d

e v

e lo

p m

e n

t g

ra m

m a ti

c c o

m p

le ti

o n

, C

W S

c o rr

e c t

w ri

ti n g

se q u e n c e s

1176 S. Kent et al.

123

For this reason, CFA was used to examine whether an alternative model with a

second-order ‘‘literacy’’ might better explain the data than the existing correlated

factors model. The alternative model also demonstrated good fit [v2 (123) = 243.21, p = .000; CFI = .961; TLI = .952; RMSEA = .061 (CI [.049, .072]); and

SRMR = .049]. Results from a Chi square difference test (Dv2 = 8.84, df = 4, p = .065) suggest this more parsimonious model would be preferred and thus, was

used in subsequent analyses. 1

Figure 1 displays factor loadings for each latent factor.

Research questions 1 and 2

The first research question addressed the unique and shared relations of component

skills of writing fluency in kindergarten. Standardized parameter estimates and

standard errors for the hypothesized SEM of writing production in kindergarten, are

presented in Fig. 2a. Results suggested good model fit: v2 (79) = 186.53, p = .000; CFI = .965; TLI = .954; RMSEA = .076 (CI [.062, .090]); and SRMR = .050.

Literacy (i.e., reading and spelling), letter-writing fluency, oral language, and

attention were all positively related to one another (u range = .35-.57, ps \ .01). Attention-related skills (c = .16, p = .001) exhibited a unique and statistically significant relation to kindergarten composition fluency after controlling for

literacy, handwriting fluency, and oral language. Early literacy skill in reading

and spelling (c = .58, p \ .001), as well as letter-writing fluency (c = .13, p = .047) were also uniquely and positively related to students’ composition

fluency, while oral language (c = -.10, p = .237) demonstrated no relationship when accounting for the other factors. This model accounted for approximately

49 % of the variance in compositional fluency in kindergarten.

For the second research question, we further sought to confirm the appropriate-

ness of including attention in the model in comparison to a model with only literacy,

Table 3 Correlations among reading, spelling, letter-writing fluency, oral language, self-regulation, and writing

Reading Spelling Letter-

writing

fluency

Oral

language

Attention K writing

production

First

writing

quality

Spelling .95 –

Letter-writing fluency .51 .54 –

Oral language .54 .57 .40 –

Attention .45 .47 .42 .35 –

K writing production .64 .67 .47 .35 .45 –

First writing quality .50 .53 .31 .42 .40 .37 –

First writing production .60 .64 .34 .33 .47 .52 .61

All coefficients are statistically significant at the .01 level

K kindergarten, First first grade

1 A CFA with a single factor ‘‘literacy’’ variable was also conducted resulting in significantly worse fit

than either the correlated factor or higher-order factor model.

Predictors of kindergarten and first grade writing 1177

123

letter-writing fluency, and oral language as predictors of writing production

(Fig. 2b). The model including attention had a statistically significant better fit than

a model without this factor (Dv2 = 73.5, df = 4, p \ .001) supporting its inclusion.

Research question 3

The third research question considered the longitudinal relations of kindergarten

component skills to first grade writing. As presented in Fig. 3, a SEM predicting

compositional fluency and quality in first grade from kindergarten component skills

exhibited excellent fit: v2 (104) = 203.32, p = .000; CFI = .964; TLI = .953; RMSEA = .061 (CI [.049, .074]); and SRMR = .047. In this longitudinal model,

attention in kindergarten was uniquely related to both compositional fluency

(c = .23, p \ .001) and quality (c = .19, p = .001) in first grade after accounting for the other factors. Literacy skills in kindergarten were also uniquely and

positively related to fluency (c = .60, p \ .001) and quality (c = .36, p \ .001) of writing in the spring of first grade, while kindergarten letter-writing fluency

exhibited no statistically significant relationship to either first grade outcome

(c = -.05 and -.03, ps [ .41). Oral language skills in kindergarten were not uniquely related to compositional fluency in first grade (c = -.07 p = .50) but did exhibit a unique relation to quality of writing in first grade (c = .16 p = .05). Compositional fluency and quality were moderately related (u = .42, p \ .001). Overall, the model accounted for 33 and 45 % of the variance in compositional

quality and fluency respectively, in first grade.

Fig. 1 Factor loadings for kindergarten and first grade latent variables

1178 S. Kent et al.

123

Research question 4

The final question examined whether kindergarten component skills had a direct or

indirect effect on first grade writing quality and fluency. The resulting model (see

Fig. 4) again demonstrated excellent model-data fit: v2 (149) = 289.34, p = .000; CFI = .962; TLI = .951; RMSEA = .061 (CI [.050, .071]); and SRMR = .046.

After accounting for both compositional fluency in kindergarten and all other

component skills, attention (c = .19, p \ .01), literacy skills (c = .36, p \ .001)

Fig. 2 Standardized structural regression weights (standard errors in parentheses) for SEM of kindergarten component skills and writing (a) and model with attention constrained to 0 (b). Solid lines represent p \ .01; dashed lines p [ .05; dotted lines represent paths constrained at 0

Predictors of kindergarten and first grade writing 1179

123

and oral language (c = .16, p = .05) in kindergarten exhibited statistically significant direct paths to first grade writing quality. Direct effects to first grade

compositional fluency were only statistically significant for attention (c = .22, p \ .01) and early literacy (c = .53, p \ .01), while oral language had no unique relation (c = -.06, p = .55). Letter-writing fluency demonstrated no unique

Fig. 3 Standardized SEM coefficients (standard errors) for kindergarten component skills predicting 1st grade writing quality and fluency. Solid lines represent p \ .01, dashed lines represent p [ .05

Fig. 4 Full SEM with standardized structural regression weights (standard errors) of direct and indirect effects of kindergarten component skills on 1st grade writing quality and fluency. Solid lines represent p \ .01, dashed lines represent p [ .05. Model separated by outcome for presentation purposes

1180 S. Kent et al.

123

relationship with composition fluency or quality in this model. Despite the moderate

factor correlations between fluency of written composition in kindergarten and first

grade compositional quality (r = .37) and fluency (r = .52), kindergarten writing

was not predictive of either writing quality (c = .00, p = .99) or fluency (c = .12, p = .24) in first grade, after accounting for the direct effects of kindergarten

component skills. Thus, kindergarten writing fluency does not appear to mediate the

relationship between kindergarten component skills and writing quality or

production 1 year later. This model accounted for approximately 33 % of the

variance in writing quality and 45 % of the variance in first grade writing

production. A summary of model-fit statistics for all CFA and SEMs is presented in

Table 4.

Discussion

The present study provides preliminary findings regarding early predictors of

writing development. Based on longitudinal data collected on a diverse sample of

students in their kindergarten and first grade years, evidence supports a model of

early writing including attention as a component factor given its unique relation to

compositional fluency and quality above and beyond early literacy and language

ability. An early literacy factor, related to word reading and spelling proficiency,

accounted for statistically significant variation in concurrent and future writing

outcomes. Student’s handwriting automaticity in kindergarten showed a unique,

concurrent relation to fluency of composition but not to writing quality and fluency

1 year later. Finally, after accounting for other component skills, kindergarten oral

language was only related to quality of writing in first grade. Of particular note, all

statistically significant relations of component skills, both concurrently and

longitudinally, to writing outcomes were direct effects.

Substantiating the important role of attention, an aspect of self-regulation, in

writing development serves to bolster previous research findings of this relationship

in older grades (Graham and Harris, 2000; Hooper et al., 2002, 2011) and extend

Table 4 Model fit statistics for CFA and SEM

Model v2 (df) RMSEA (CI) CFI TLI SRMR

CFA

Correlated factors 234.62 (119) .060 (.049–.072) .963 .952 .048

2nd order literacy factor 243.21 (123) .061 (.049–.072) .961 .952 .049

SEM

Concurrent 186.53 (79) .074 (.060–.088) .965 .954 .050

Concurrent without attention factor 260.03 (83) .093 (.008–.106) .942 .927 .193

Longitudinal: direct effects 203.32 (104) .061 (.049–.074) .964 .953 .047

Longitudinal: direct and indirect effects 289.34 (149) .061 (.050–.071) .962 .951 .046

RMSEA root mean square error of approximation, CFI comparative fit index, TLI Tucker-Lewis index,

SRMR standardized root mean square residuals, CI confidence interval

Predictors of kindergarten and first grade writing 1181

123

recent evidence with students at this young age (e.g., Kim et al., 2013). Difficulties

in working memory and attention, both assessed in the present analyses, have been

linked to poor writing outcomes in both lower and upper elementary grades

(Chenault et al., 2006; Kim et al., 2013), and this study provides emerging evidence

of this role in writing development as early as kindergarten. To our knowledge, the

present results are the first to examine a component skill model of early writing in

kindergarten, and longitudinally to first grade, that includes attention with literacy

and language-related skills. Given constraints in self-regulatory processes such as

attention during writing, novice writers have been previously described as engaging

in ‘‘knowledge-telling’’, with a primary focus on the act of putting thoughts to words

on paper (Bereiter & Scardamalia, 1987; McCutchen, 2000). Our findings suggest

that higher levels of attention regulation at this emergent level may free cognitive

resources to assist not only in efficient text production by allowing students to

remain engaged, abstain from competing demands, and transfer ideas and thoughts

to the written word, but also engage in self-regulatory strategies during writing that

promote higher quality compositions. While strategies such as goal-setting,

planning, and revising has been attributed to older and more skilled writers

(Bereiter & Scardamalia, 1987), it is possible that students, particularly those with

better attention skills, may begin to develop these strategies at an earlier age.

The individual relations of reading and spelling could not be modeled due to the

extreme overlap with these two factors. As previously stated, this finding is not

without precedent at this age level (Mehta et al., 2005). Nonetheless, the important

role of early literacy on kindergarten writing fluency, as well as both composition

fluency and quality 1 year later, was clearly evident. Both reading and spelling are

influenced by phonological, orthographic, and morphological knowledge (Berninger

et al., 2002). One could reason that stronger knowledge in these areas facilitates

access to written text via reading and subsequently, better understanding of the

written language system, potentially aiding the generation of written text. Further,

greater phonological, orthographic, and morphological knowledge may allow

students to form a lexicon of letter/word forms that can be accessed quickly and

accurately (Berninger et al., 2006) and thus, allows ideas to be represented in text at

the word, sentence, and discourse levels through efficient encoding.

Automaticity in handwriting also had a small, yet statistically significant relation

to the efficient production of words, sentences, and ideas in writing in kindergarten,

clearly supporting the extant research with students in the earliest grades (Jones &

Christensen, 1999; Kim et al., 2011; Puranik & Al Otaiba, 2012). Although a

separable construct, letter-writing fluency, which was moderately correlated with

spelling (u = .54), likely operates similarly to automaticity in spelling skill in that without automatic retrieval of letter forms, generating text becomes slow and

effortful and the strategic thought processes required for writing are impeded,

particularly on timed measures of writing (Graham et al., 1997; McCutchen, 2000).

Of note was the absence of relation between handwriting fluency in kindergarten

and writing outcomes 1 year later. Although evidence exists supporting the role of

handwriting fluency on writing in first grade (e.g., Jones & Christensen, 1999; Kim

et al., 2013), the present study did not include concurrent measures of these skills in

first grade.

1182 S. Kent et al.

123

The relations of oral language to writing outcomes were mixed, with a small, yet

statistically significant finding for writing quality but not compositional fluency.

This supports research at the earliest grades demonstrating individual differences in

oral language were not related to writing fluency when accounting for other factors

(Abbott & Berninger, 1993; Puranik & Al Otaiba, 2012). Our findings do however

differ from the Kim et al. (2011) finding of oral language as a unique predictor of

writing fluency in kindergarten. The act of writing requires the development and

elaboration of ideas and therefore, limitations in children’s vocabulary and

knowledge of language structures may serve to constrain the quality of text

generated (McCutchen, 2000). Prior research has established the relations between

oral language and writing increase across the early grades (Abbott & Berninger,

1993; Berninger et al., 2006), it is plausible that the importance of oral language to

compositional quality grows across the grades much like it does to reading

comprehension after the earliest grades (Storch & Whitehurst, 2002). Furthermore,

Juel (1988), Juel et al. (1986) have also found that student’s ideation, likely related

to language ability, becomes more important to writing after first grade. It may be

that larger relations are apparent after first grade because writing production for

students becomes less constrained, and thus, individual differences in language

skills are more evident in the quality of composition.

In general, findings suggest that students’ literacy and language skills may work

in tandem with self-regulatory functions such as attention to influence writing at this

early level. Our findings clearly dovetail with results from other studies (Kim et al.,

2011, 2013; Puranik & Al Otaiba, 2012) demonstrating the role of early component

skills of writing including language and literacy skills, as well as regulation of

attention. However, the present study further adds to our understanding of their

influence over time (i.e., longitudinally). So, while knowledge of the writing system

(e.g., Fitzgerald & Shanahan, 2000), automaticity with transcription, and oral

language proficiency appear necessary for early writing development, they may not

be sufficient. Individual differences in level of attention also play a role efficient text

production (fluency) and qualitative elements of writing such as ideation,

organization, structure, and word choice for these early writers. Berninger and

Winn’s (2006) view of text generation posits that executive functions, particularly

supervisory attention, do play a role in beginning writing along with transcription

skills, Thus, we believe that the results of this present study do indeed lend support

to their theory at the stage of beginning writing. The Not-So-Simple View

(Berninger & Winn, 2006) also stresses the increased role of multiple, more

complex functions, such as the link between working memory and long-term

memory and the reliance on strategies for self-monitoring and self-regulation for

older students; however these particular executive functions were not assessed in the

current study. This may represent an area of further exploration at the younger

grades.

Despite the prevailing focus on early literacy skills in schools, particularly in

kindergarten through third grade, concerns remain. Namely, the amount of explicit

instruction in many of these component skills, as well as in writing instruction, in

kindergarten may be lacking. Graham et al. (2012) have recommended students in

kindergarten spend at least 30 min daily writing and developing writing skills.

Predictors of kindergarten and first grade writing 1183

123

However, Kent, Wanzek, and Al Otaiba (2012) have observed that only 10 %

(9 min) of an appropriated 90 min kindergarten literacy block was allocated to

writing instruction, including spelling and handwriting instruction and only 3 %

explicitly devoted to vocabulary and language development. Additional observa-

tional research in kindergarten reported similar amounts of writing instruction, with

the majority of time devoted to independent writing rather than teacher instruction

such as modeling and group instruction (Puranik, Al Otaiba, Folsom & Greulich, in

press). Further, handwriting instruction was observed for \2 min across fall and spring observations. If such skills have a clear link to writing development (e.g.,

Graham et al., 2012), increased instructional attention would be warranted at this

early level.

Limitations

Several limitations to the present study should be mentioned. The sample for this

study comes from a single school district in the southeast. Although relatively

diverse, there were few English Language Learners. In order to have more

confidence in the conclusions drawn from these analyses, cross-validation with a

different sample would be warranted. Second, although several aspects of writing

were assessed, both within fluency and quality, this data was drawn from only a

single writing sample in the spring of kindergarten and first grade. Moreover, these

writing prompts and the time students were allowed to respond differed. In

kindergarten, there was little research guidance to inform the choice of prompt, but

for first grade, a CBM-W prompt used in prior research was used. Kindergarten

students had more time to respond to the writing prompt. Further, at kindergarten, it

was not possible to code correct word sequences; thus differences in the prompts,

the time, or scoring methods could have impacted the correlation between writing at

both times. The inclusion of multiple samples of student writing, via both authentic

and direct assessment, may have increased measurement reliability. Additionally,

the presence of a floor effect on the writing sample in kindergarten (about 20 %

deemed unscorable), although attributable to developmental constraints, may have

served to decrease the resulting relationships among component skills and writing.

Third, only a teacher rating of attention/self-regulation was included in the study.

Although the SWAN appears to capture distinct factors related to attention

regulation (Saez et al., 2012) further research using additional, more direct measures

of executive functioning and self-regulation and their relationship to writing

production and quality would be warranted. This might include direct observations

of student and teacher behavior during writing tasks (e.g., teacher scaffolding and

student planning, reviewing, revising) or the utilization of new technology that

allows researchers to record student verbalizations (i.e., self-talk) when engaged in

the process of writing which may reflect early attempts at self-regulation.

Future directions

In the present study, a model of early components of writing that included attention

as well as early literacy and language skills accounted for more variance in writing

1184 S. Kent et al.

123

fluency than in studies without this component (Kim et al., 2011; Puranik & Al

Otaiba, 2012). Nonetheless, there is still much unexplained variance. It is possible

that early writing models could be advanced by examining the role of instructional

factors alongside student-level factors. While Kim et al. (2013) recently found that

instructional quality during reading and writing instruction in first grade was not

uniquely predictive of writing outcomes after accounting for student-level factors,

specific research has demonstrated that writing outcomes can be influenced by time

allocated to writing instruction (Mehta et al., 2005) and quality of instruction in

writing (Moats, Foorman, & Taylor, 2006). However, research has also demon-

strated the relative dearth of time allocated to writing instruction in the early grades

(Puranik et al., in press; Kent et al., 2012; Mehta et al., 2005). Thus, there is a

continued need for research examining specific instructional ‘‘ingredients’’ that

promote early writing skills, particularly when examined in a model accounting for

individual differences in student skills. The examination of additional student-level

factors may also be warranted, such as student attitude or self-efficacy regarding

writing. Given its’ complex nature, it reasons that individuals who have greater

belief and judgment regarding their ability to complete given writing tasks may

demonstrate increased willingness to engage in the task and be more persistent. To

date, studies have shown the positive relationship of self-efficacy to writing for

students in upper grades (Shell, Colvin, & Bruning, 1995) but we know little about

the impact on writing at the earliest grade levels.

In conclusion, the results from the present study offer preliminary findings

substantiating the role of attention, as well as early literacy and language skills, in

the development of writing fluency and quality in the earliest grades. While further

validation is necessary, these findings help provide additional evidence as the field

moves toward a more complete understanding of writing development and ways in

which such development can be promoted.

Acknowledgments This research was supported by Grant P50HD052120 from the National Institute of Child Health and Human Development, and Grant R305B04074 from the Institute of Education Sciences.

Dr. Petscher’s time was also supported by Grant R305F100005 from the Institute of Education Sciences.

The content is solely the responsibility of the authors and does not necessarily represent the official views

of the National Institute of Child Health and Human Development, the National Institutes of Health, or

the Institute of Education Sciences.

References

Abbott, R. D., & Berninger, V. W. (1993). Structural equation modeling of relationships among

development skills and writing skills in primary- and intermediate-grade writers. Journal of

Educational Psychology, 85, 478–508. doi:10.1037/0022-0663.85.3.478.

Abbott, R. D., Berninger, V. W., & Fayol, M. (2010). Longitudinal relationships of levels of language in

writing and between writing and reading in grades 1 to 7. Journal of Educational Psychology,

102(2), 281–298. doi:10.1037/a0019318.

Al Otaiba, S., Connor, C. M., Folsom, J. S., Greulich, L., Meadows, J., & Li, Z. (2011). Assessment data-

informed guidance to individualize kindergarten reading instruction: Findings from a cluster-

randomized control field trial. Elementary School Journal, 111, 535–560. doi:10.1086/659031.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th

ed.). Washington, DC: American Psychiatric Association.

Predictors of kindergarten and first grade writing 1185

123

Asparouhov, T. (2006). General multi-level modeling with sampling weights. Communications in

Statistics: Theory and Methods, 35, 439–460.

Barkley, R. (1996). Linkages between attention and executive functions. In G. R. Lyon & N. Krasnegor

(Eds.), Attention, memory, and executive functions (pp. 307–326). Baltimore, MD: Brookes.

Bereiter, C., & Scardamalia, M. (1987). The psychology of written composition. Hillsdale, NJ: Lawrence

Erlbaum.

Berninger, V. W., Abbott, R. D., Abbott, S. P., Graham, S., & Richards, T. (2002). Writing and reading:

Connections between language by hand and language by eye. Journal of Learning Disabilities, 35,

39–56. doi:10.1177/002221940203500104.

Berninger, V. W., Abbott, R., Jones, J., Wolf, B., Gould, L., Anderson-Youngstrom, M., et al. (2006).

Early development of language by hand: Composing, reading, listening, and speaking connections;

three letter-writing modes; and fast mapping in spelling. Developmental Neuropsychology, 29(1),

61–92. doi:10.1207/s15326942dn2901_5.

Berninger, V. W., & Rutberg, J. (1992). Relationship of finger function to beginning writing: Application

to diagnosis of writing disabilities. Developmental Medicine and Child Neurology, 34, 155–172.

doi:10.1111/j.1469-8749.1992.tb14993.x.

Berninger, V. W., & Winn, W. D. (2006). Implications of advancements in brain research and technology

for writing development, writing instruction, and educational evolution. In C. A. MacArthur, S.

Graham, & J. Fitzgerald (Eds.), Handbook of Writing Research (pp. 96–114). New York, NY: The

Guilford Press.

Blair, C. (2002). School readiness: Integrating cognition and emotion in a neurobiological conceptu-

alization of children’s functioning at school entry. American Psychologist, 57(2), 111–127. doi:10.

1037/0003-066X.57.2.111.

Byrne, B., & Fielding-Barnsley, R. (1993). Evaluation of a program to teach phonemic Awareness to

young children: A one year follow up. Journal of Educational Psychology, 85, 104–111. doi:10.

1037/0022-0663.85.1.104.

Byrne, B., Wadsworth, S., Corley, R., Samuelsson, S., Quain, P., DeFries, J., et al. (2005). Longitudinal

twin study of early literacy development: Preschool and kindergarten phases. Scientific Studies of

Reading, 9, 219–235. doi:10.1207/s1532799xssr0903_3.

Chenault, B., Thomson, J., Abbott, R. D., & Berninger, V. W. (2006). Effects of prior attention training

on child dyslexics’ response to composition instruction. Developmental Neuropsychology, 29(1),

243–260. doi:10.1207/s15326942dn2901_12.

Common Core State Standards Initiative. (2010). Common core state standards for English language arts

and literacy in history/social studies, science, and technical subjects. Retrieved from http://www.

corestandards.org/assets/CCSSI_ELA%20Standards.pdf.

Duncan, G. J., Dowsett, C. J., Claessens, A., Magnuson, K., Huston, A. C., Klebanov, P., et al. (2007).

School readiness and later achievement. Developmental Psychology, 43, 1428–1446. doi:10.1037/

0012-1649.43.6.1428.

Ehri, L. C. (2000). Learning to read and learning to spell: Two sides of a coin. Topics in Language

Disorders, 20, 19–36. doi:10.1006/ceps.1999.1002.

Fitzgerald, J., & Shanahan, T. (2000). Reading and writing relations and their development. Educational

Psychologist, 35, 39–50. doi:10.1207/S15326985EP3501_5.

Graham, S., Berninger, V. W., Abbott, R. D., Abbott, S. P., & Whitaker, D. (1997). Role of mechanics in

composing of elementary school students: A new methodological approach. Journal of Educational

Psychology, 89, 170–182. doi:10.1037/0022-0663.89.1.170.

Graham, S., Bollinger, A., Booth Olson, C., D’Aoust, C., MacArthur, C., McCutchen, D., et al. (2012a).

Teaching elementary school students to be effective writers: A practice guide (NCEE 2012-4058).

Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of

Education Sciences, US Department of Education. Retrieved from http://ies.ed.gov/ncee/wwc/

publications_reviews.aspx#pubsearch.

Graham, S., & Harris, K. R. (2000). The role of self-regulation and transcription skills in writing and

writing development. Educational Psychologist, 35, 3–12. doi:10.1207/S15326985EP3501_2.

Graham, S., Harris, K. R., & McKeown, D. (2013). The writing of students with learning disabilities,

meta-analysis of self-regulated strategy development writing intervention studies, and future

directions: Redux. In L. Swanson, K. R. Harris, & S. Graham (Eds.), Handbook of learning

disabilities (2nd ed., pp. 405–438). New York: Guilford Press.

1186 S. Kent et al.

123

Graham, S., McKeown, D., Kiuhara, S., & Harris, K. R. (2012b). A meta-analysis of writing instruction

for students in the elementary grades. Journal of Educational Psychology, 104, 879–896. doi:10.

1037/a0029185.

Graham, S., & Perin, D. (2007). A meta-analysis of writing instruction for adolescent students. Journal of

Educational Psychology, 99, 445–476. doi:10.1037/0022-0663.99.3.445.

Happaney, K., Zelazo, P. D., & Stuss, D. T. (2004). Development of orbitofrontal function: Current

themes and future directions. Brain and Cognition, 55(1), 1–10. doi:10.1016/j.bandc.2004.01.001.

Hooper, S. R., Costa, L., McBee, M., Anderson, K. L., Yerby, D. C., Knuth, S. B., et al. (2011).

Concurrent and longitudinal neuropsychological contributors to written language expression in first

and second grade students. Reading and Writing, 24, 221–252. doi:10.1007/s11145-010-9263-x.

Hooper, S. R., Swartz, C. W., Wakely, M. B., de Kruif, R. E. L., & Montgomery, J. W. (2002). Executive

functions in elementary school children with and without problems in written expression. Journal of

Learning Disabilities, 35(1), 57–68. doi:10.1177/002221940203500105.

Jones, D., & Christensen, C. A. (1999). Relationship between automaticity in handwriting and students’

ability to generate written text. Journal of Educational Psychology, 91, 44–49. doi:10.1037/0022-

0663.91.1.44.

Juel, C. (1988). Learning to read and write: A longitudinal study of 54 children from first through fourth

grades. Journal of Educational Psychology, 80, 437–447. doi:10.1037/0022-0663.80.4.437.

Juel, C., Griffith, P. L., & Gough, P. B. (1986). Acquisition of literacy: A longitudinal study of children in

first and second grade. Journal of Educational Psychology, 78, 243–255. doi:10.1037/0022-0663.78.

4.243.

Kent, S. C., Wanzek, J., & Al Otaiba, S. (2012). Print reading in general education kindergarten

classrooms: What does it look like for students at-risk for reading difficulties? Learning Disabilities

Research and Practice, 27(2), 56–65. doi:10.1111/j.1540-5826.2012.00351.x.

Kim, Y.-S., Al Otaiba, S., Folsom, J. S., Folsom, J. S., & Greulich, L. (2013). Language, literacy,

attentional behaviors, and instructional quality predictors of written composition for first graders.

Early Childhood Research Quarterly, 28, 461–469. doi:10.1016/j.ecrsq.2013.01.001.

Kim, Y.-S., Al Otaiba, S., Folsom, J. S., Greulich, L., & Puranik, C. (in press). Evaluating the

dimensionality of first grade written composition. Journal of Speech, Hearing, and Language

Research.

Kim, Y., Al Otaiba, S., Puranik, C., Folsom, J. S., Greulich, L., & Wagner, R. K. (2011). Componential

skills of beginning writing: An exploratory study. Learning and Individual Differences, 21,

517–525. doi:10.1016/j.lindif.2011.06.004.

Kline, R. B. (2011). Principles and practice of structural equation modeling (2nd ed.). New York, NY:

Guilford Press.

Lehto, J. E., Juuiärvi, P., Kooistra, L., & Pulkkinen, L. (2003). Dimensions of executive functioning:

Evidence from children. British Journal of Developmental Psychology, 21, 59–80. doi:10.1348/

026151003321164627.

Lembke, E., Deno, S. L., & Hall, K. (2003). Identifying an indicator of growth in early writing

proficiency for elementary school students. Assessment for Effective Intervention, 28, 23–35. doi:10.

1177/073724770302800304.

McClelland, M. M., Acock, A. C., Piccinin, A., Rhea, S. A., & Stallings, M. C. (2013). Relations between

preschool attention span-persistence and age 25 educational outcomes. Early Childhood Research

Quarterly, 28(2), 314–324. doi:10.1016/j.ecresq.2012.07.008.

McCutchen, D. (2000). Knowledge, processing, and working memory: Implications for a theory of

writing. Educational Psychologist, 35, 13–23. doi:10.1207/S15326985EP3501_3.

McMaster, K. L., Du, X., & Pestursdottir, A. L. (2009). Technical features of curriculum-based measures

for beginning writers. Journal of Learning Disabilities, 42, 41–60. doi:10.1177/0022219408326212.

Mehta, P. D., Foorman, B. R., Branum-Martin, L., & Taylor, W. P. (2005). Literacy as a unidimensional

multilevel construct: Validation, sources of influence, and implications in a longitudinal study in

grades 1 to 4. Scientific Studies of Reading, 9(2), 85–116. doi:10.1207/s1532799xssr0902_1.

Moats, L., Foorman, B., & Taylor, P. (2006). How quality of writing instruction impacts high-risk fourth

graders’ writing. Reading and Writing, 19(4), 363–391. doi:10.1007/s11145-005-4944-6.

National Commission on Writing. (2004, September). Writing: A ticket to work… Or a Ticket Out. New York, NY: College Entrance Examination Board. Retrieved from: http://www.collegeboard.com/

prod_downloads/writingcom/writing-ticket-to-work.pdf.

Newcomer, P. L., & Hamill, D. D. (1997). Test of language development-primary (Vol. 3). Austin, TX:

Pro-Ed.

Predictors of kindergarten and first grade writing 1187

123

Northwest Regional Educational Laboratory. (2011). 6 ? 1 Trait �

Writing. Retrieved from http://

educationnorthwest.org/traits.

Puranik, C. S., & Al Otaiba, S. (2012). Examining the contribution of handwriting and spelling to written

expression in kindergarten children. Reading and Writing: An Interdisciplinary Journal, 25,

1523–1546. doi:10.1007/s11145-011-9331-x.

Puranik, C. S., Al Otaiba, S., Sidler, J. F., & Greulich, L. (in press). Exploring the amount and type of

writing instruction during language arts instruction in kindergarten classrooms. Reading and

Writing: An Interdisciplinary Journal. doi:10.1007/s11145-9441-8.

Puranik, C. S., Lombardino, L. J., & Altmann, L. J. (2007). Writing through retellings: An exploratory

study of language-impaired and dyslexic populations. Reading and Writing: An Interdisciplinary

Journal, 20, 251–272. doi:10.1007/s11145-006-9030-1.

Rhoades, B. L., Warren, H. K., Domitrovich, C. E., & Greenberg, M. T. (2011). Examining the link between

preschool social-emotional competence and first grade academic achievement: The role of attention

skills. Early Childhood Research Quarterly, 26(2), 182–191. doi:10.1016/j.ecresq.2010.07.003.

Saez, L., Folsom, J. S., Al Otaiba, S., & Schatschneider, C. (2012). Relations among student attention

behaviors, teacher practices, and beginning word reading skill. Journal of Learning Disabilities, 45,

418–432. doi:10.1177/0022219411431243.

Salahu-Din, D., Persky, H., & Miller, J. (2008). The nation’s report card: Writing 2007 (NCES

2008–468). National Center for Education Statistics, Institute of Education Sciences, U.S.

Department of Education, Washington, DC.

Shanahan, T. (2006). Relations among oral language, reading, and writing development. In C.

A. MacArthur, S. Graham, & J. Fitzgerald (Eds.), Handbook of writing research (pp. 83–95). New

York, NY: The Guilford Press.

Shell, D. F., Colvin, C., & Bruning, R. H. (1995). Self-efficacy, attribution, and outcome expectancy

mechanisms in reading and writing achievement: Grade-level and achievement-level differences.

Journal of Educational Psychology, 87(3), 386–398. doi:10.1037/0022-0663.87.3.386.

Storch, S. A., & Whitehurst, G. J. (2002). Oral language and code-related precursors to reading: Evidence

from a longitudinal structural model. Developmental Psychology, 38(6), 934–947. doi:10.1037/

0012-1649.38.6.934.

Swanson, J., Shuck, S., Mann, M., Carlson, C., Hartman, K., Sergeant, J., et al. (2006). Categorical and

dimensional definitions and evaluations of symptoms of ADHD: The SNAP and SWAN Rating

Scales. Unpublished manuscript. University of California Irvine, CA.

Tangel, D. M., & Blachman, B. A. (1992). Effect of phoneme awareness instruction on kindergarten

children’s invented spelling. Journal of Reading Behavior, 24, 233–261. doi:10.1080/

10862969209547774.

Thomson, J. B., Chenault, B., Abbott, R. D., Raskind, W. H., Richards, R., Aylward, E., et al. (2005).

Converging evidence for attentional influences on the orthographic word form in child dyslexics.

Journal of Neurolinguistics, 18, 93–126. doi:10.1016/j.jneuroling.2004.11.005.

Torgesen, J. K., Wagner, R. K., & Rashotte, C. A. (1999). Test of word reading efficiency. Austin, TX:

Pro-Ed.

Wagner, R. K., Puranik, C. S., Foorman, B., Foster, E., Tschinkel, E., & Kantor, P. T. (2011). Modeling

the development of written language. Reading and Writing: An Interdisciplinary Journal, 24,

203–220. doi:10.1007/s11145-010-9266-7.

Willcutt, E. G., Pennington, B. F., Boada, R., Ogline, J. S., Tunick, R. A., Chhabildas, N. A., et al. (2001).

A comparison of the cognitive deficits in reading disability and attention-deficit/hyperactivity

disorder. Journal of Abnormal Psychology, 110, 157–172. doi:10.1037/0021-843X.110.1.157.

Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). Woodcock-Johnson III tests of achievement.

Itasca, IL: Riverside Publishing.

1188 S. Kent et al.

123

Copyright of Reading & Writing is the property of Springer Science & Business Media B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.