Read Writ

https://doi.org/10.1007/s11145-018-9822-0

The impact of supplemental handwriting and spelling

instruction with first grade students who do not acquire
transcription skills as rapidly as peers: a randomized
control trial

Steve Graham1,2 · Karen R. Harris1,2 ·

Mary Adkins3

© Springer Science+Business Media B.V., part of Springer Nature 2018

Abstract The impact of supplemental handwriting and spelling instruction on

learning to write was examined in an experimental study with first grade students
who were not acquiring these skills as rapidly as their classmates. Thirty students
(16 boys, 14 girls) were randomly assigned to a handwriting and spelling instruc-
tional condition or a phonological awareness instructional control condition.
Students in each condition received 16 h of one-on-one instruction. The treatment
condition was designed to improve children’s handwriting fluency, handwriting
legibility, spelling accuracy, and knowledge of spelling patterns. In comparison to
students in the phonological awareness control condition, students who received
supplemental handwriting and spelling instruction made greater gains in hand-
writing fluency, handwriting legibility, and spelling accuracy. The treatment
condition also resulted in greater gains in sentence construction fluency and com-
position vocabulary, but did not statistically enhance composition length or
composition quality. These findings provide partial support for the theoretical
proposition that text transcription skills are causally related to writing. Further, this
study demonstrates that explicit and supplemental handwriting and spelling
instruction can play an important role in teaching writing to young children who
acquire text transcription skills more slowly than their peers.

& Steve Graham

[email protected]
Karen R. Harris
[email protected]
Mary Adkins
[email protected]
1
Learning Sciences Institute Australia – Brisbane Campus, Brisbane, Australia
2
Mary Lou Fulton Teachers College, Arizona State University, 434 Bengamin Building, Tempe,
AZ, USA
3
Gouher College, 1021 Dulaney Valley Road, Baltimore, MD 1021, USA

123
 S. Graham et al.

Keywords Handwriting · Spelling · Writing

Writing is a complex and challenging task

There are many competing and potentially demanding cognitive actions that writers
must attend to while composing text (Graham, 2006; Kellogg, 1993). Take for
instance the seemingly simple task of writing a sentence (Graham, in press). An idea
must be generated and evaluated for its suitability, crafted into a grammatically
correct sentence, and transcribed into written text with correct spelling, capitaliza-
tion, and punctuation, resulting in a sentence that precisely reflects the author’s
intentions, will be understood by the intended audience, and is appropriate for the
writing community and the audience for which the sentence is created. If the
cognitive actions needed to produce this sentence exceed the capacity of the writer’s
processing system, due to inefficient skills such as slow handwriting or difficulties
with spelling, then cognitive overload is likely (McCutchen, 1988), resulting in a
disruption in one or more aspects of the sentence construction process. This is not
just limited to sentence writing though, as it applies to the larger act of composing
(we refer to this as writing in this article, whereas the term handwriting fluency and
handwriting legibility refers to how quickly and legibly letters of the alphabet are
produced, respectively).
The complexity and difficulty of writing is reflected in the challenges that many
children experience in learning to write. For instance, only 27% of grade 12 students
in the United States (U.S.) performed at or above the “proficient” writing level
(defined as solid academic performance by the National Center for Educational
Statistics, 2012). Unfortunately, such findings are not limited to the U.S., as a
majority of students in countries across the globe perform poorly on writing
assessments (e.g., Department of Education, 2012; Micahelowa, 2001; Research
Center for Psychological and Educational Testing, 2016).

Children simplify writing to address the challenges of writing

Writing can be especially challenging for young children, as they have not yet
mastered the knowledge, skills, and processes needed to write effectively (Graham,
2006). Not surprisingly, therefore, they typically apply a relatively simple approach
to writing, as they write by telling what they know about a topic (i.e., writing-as-
remembering; Scardamalia & Bereiter, 1986). Such an approach serves an adaptive
function for beginning writers. Writing requires that children transcribe their ideas
onto paper or into digital text. However, the handwriting and keyboarding skills of
beginning writers are slow and laborious, and they cannot spell most words
correctly. Because these text transcription skills require considerable cognitive
effort for young children (Graham & Harris, 2000; McCutchen, 1988), the
knowledge telling approach described above has the advantage of making it possible
to produce text while simultaneously minimizing the use of other demanding

123
The impact of supplemental handwriting and spelling…

cognitive activities, such as planning, monitoring, and evaluating when writing

(Kellogg, 1993), reducing the possibility of cognitive overload.

Transcription skills impact children’s writing in multiple ways

Until the transcription skills (e.g., handwriting and spelling) of beginning writers
become more automatic and fluent, they do more than just consume cognitive
resources; they can interfere with other writing processes (Berninger, 1999; Graham
& Harris, 2000). Slow and laborious handwriting or keyboarding, for instance, may
interfere with text generation, as ideas or plans a child is trying to hold in working
memory may be lost before they can be committed to paper or digital text. Likewise,
having to consciously think about how to spell a word while writing may tax
cognitive processing capacity, leading the child to forget ideas in working memory
not yet committed to paper. Further, uncertainty about how to spell a word may lead
to the selection of a different word the child knows how to spell, potentially
undermining the preciseness of the intended message.
Handwriting, keyboarding, and spelling can negatively impact beginning writers
in two other important ways. First, an analysis of the available literature by Graham
and his colleagues (Graham, Harris, & Hebert, 2011) found that readers form
negative judgments about students’ papers when spelling errors are evident or the
text is difficult to read because of issues with handwriting legibility. They also
found that teachers were more likely to judge text that was typed more harshly than
handwritten text when spelling or grammar errors are present, as they expected such
errors to be less evident in typed versus handwritten text.
Second, the impact of poor transcription skills may extend beyond a specific
writing task for young beginning writers. For example, Berninger, Mizokawa, and
Bragg (1991) reported that children who experienced difficulties mastering
handwriting avoided writing whenever possible and developed a mindset that they
could not write. This is not an insignificant issue, as 44% of elementary students
may experience difficulties with handwriting (Graham & Weintraub, 1996).

Young writers need to master transcription skills

An important goal in early writing instruction is to help beginning writers master

basic text transcription skills so that they can be executed correctly and with little
effort (Graham, 1999). This may be particularly critical for children who do not
master these skills as quickly or efficiently as their peers. The present study directly
tested this proposition by examining the impact of supplemental handwriting and
spelling instruction with first grade writers who scored at the bottom 25th percentile
or lower on screening measures for both handwriting fluency and spelling. The
study only tested the effectiveness of handwriting and spelling instruction, and not
keyboarding instruction, as students only wrote by hand at school according to their
teachers.
To our knowledge, this is the only study that examines the impact of a combined
supplemental handwriting and spelling program with young children who do not
acquire these skills as efficiently as their peers. Past intervention studies examined

123
 S. Graham et al.

the effectiveness of handwriting or spelling instruction separately (e.g., Alves et al.,

2016; Berninger et al., 1997, 1998, 2002; Case-Smith, Holland, & White, 2014;
Graham, Harris, & Fink, 2000; Graham, Harris, Fink-Chorzempa, 2002; Reiben,
Ntamakiliro, Gonthier, & Fayol, 2005; Uhry & Sheperd, 1993; Weintraub, Yinon,
Hirsch, & Parush, 2009). These and other studies (see Graham & Santangelo, 2014;
Santangelo & Graham, 2016) demonstrated that the handwriting and spelling of a
broad array of children can be improved by directly teaching each skill separately,
but it is not clear if combining such instruction improves both skills simultaneously.
Theoretically, it is possible that combined handwriting and spelling instruction
would be beneficial for young beginning writers. Handwriting instruction designed
to improve handwriting fluency and legibility may make it easier for young children
to practice writing words as well as check spelling accuracy during spelling
instruction, leading to greater spelling gains. Similarly, spelling instruction may
make it easier for beginning writers to remember the correct sequence of letters in
practice words used during handwriting instruction, facilitating greater handwriting
fluency (while effective handwriting instruction involves practice with individual
letters, it also involves practicing target letters within words; Graham, 1999).

Current study

The current study was designed to answer the following questions: (a) Does a
supplemental handwriting and spelling instructional program improve the hand-
writing and spelling skills of beginning writers who do not acquire these skills as
rapidly as their peers?, and (b) Does supplemental handwriting and spelling
instruction improve these children’s sentence construction fluency and performance
writing compositions? We applied a randomized control trial to answer these
questions. Students were randomly assigned to either the handwriting/spelling
instruction condition or a phonological awareness instruction control condition.

Predictions for improved handwriting and spelling performance

We anticipated that providing supplemental handwriting and spelling instruction to

first grade students who do not develop these skills as rapidly as their classmates
would enhance handwriting fluency, handwriting legibility, and spelling accuracy.
We expected that handwriting legibility and fluency would be enhanced, as students
received additional one-on-one instruction in how to form letters that were
challenging for them, and they were provided opportunities to practice these letters
in context. Similar forms of handwriting instruction enhanced handwriting legibility
and fluency in Santangelo and Graham’s meta-analysis (2016). We also expected
that the additional one-on-one spelling instruction students received would improve
their spelling accuracy, as they were taught to spell words that beginning writers
commonly apply when writing, received instruction designed to increase their
knowledge of spelling patterns, and were provided opportunities to apply newly
acquired spelling skills in context. In their meta-analysis, Graham and Santangelo

123
The impact of supplemental handwriting and spelling…

(2014) reported that similar forms of instruction enhanced the spelling performance
of poor spellers.

Predictions for improved sentence construction fluency and writing

performance

We further anticipated that handwriting and spelling instruction would have a

positive impact on students’ sentence construction fluency and their performance on
writing compositions (i.e., composition length, composition vocabulary, and
composition quality). Theoretically, instruction designed to make beginning writers’
handwriting more fluent and correct spelling more automatic should free up
cognitive resources for carrying out other writing process and reduce cognitive
overload and interference (Graham, in press; McCutchen, 1988). This theoretical
position is supported by Santangelo and Graham’s (2016) meta-analysis, as they
found that handwriting instruction statistically improved children’s sentence
construction fluency and two aspects of their writing: composition length and
composition quality. It must be noted, however, that similar positive results were
not obtained in Graham and Santangelo’s (2014) meta-analysis examining the
impact of spelling instruction.

Methods

Students

Screening

In November, we administered two screening measures to 336 children in 14 first

grade classrooms in four schools in a single school district in the Mid-Atlantic
region of the US. This district was located in a large metropolitan area, serving
mostly children who were African American (80%) and who were eligible for a free
or reduced lunch (70%).
One of the screening measures was the Spelling subtest from the Wechsler
Individual Achievement Test (WIAT; Wechsler, 2009), and the other screening
measure assessed handwriting fluency. The WIAT Spelling subtest assesses
children’s ability to spell individual words (M standard score = 100; SD = 15).
It contains 45 items, and internal consistency and test–retest reliabilities as reported
in the test manual exceed 0.80. With the she screening measure for handwriting
fluency, children were asked to copy the following sentence quickly and correctly:
A little boy lived with his father in a forest. Before copying the sentence, the
examiner read the sentence to students, and then they read it back to her. Students
were directed to copy the sentence until the examiner said stop (90 s after students
were told to start). The score for this test was the number of letters copied correctly
(i.e., clearly identifiable as the correct letter, with no missing parts, reversals, or
rotations).

123
 S. Graham et al.

Student responses to both tests were independently scored by trained graduate

students. Interrater reliability was 0.99 and 0.95 for spelling and the handwriting
fluency screening measures, respectively.
To qualify for participation in this study, a child had to fall two-thirds of a
standard deviation below the normative sample on WIAT Spelling (standard score
of 90 or less) and two-thirds of a standard deviation or more below their school
mean for all first graders in their school on the screening measure for handwriting
fluency (school means ranged from 15.94 to 23.0).
Thirty-eight children met both of these criteria. Parental consent and student
assent was obtained for 34 of these children (there were at least two of these
children in all 14 classrooms, with four students in three of the classes). Over the
course of the experiment, four students moved (two from each condition). The
children who moved did not appear to differ appreciably from the remaining 30
students. The average score on the spelling screening measure (WIAT Spelling) for
the 30 students who completed the study was 80.37 (SD = 16.62), whereas the
average score on the screening measure for handwriting fluency was 15.93 letters
(SD = 5.24).

Participating students

Students who met the criteria specified above were randomly assigned by class to
treatment (handwriting and spelling instruction) or control (phonological awareness
instruction). The phonological instruction condition controlled for attention and
maturation effects. The mean chronological age of the 30 students completing the
study was 76.5 months (SD = 3.94). Sixteen of them were boys, 22 of them were
African American (six students were white and two Hispanic), 10 students received
special education services (for speech and language problems or ADHD), three
children were left handed, and English was not the first language for two children
(both in the handwriting and spelling condition). The racial composition of these 30
students was similar to racial composition of the school district and the participating
schools.
There were no statistical differences between students in the two conditions (i.e.,
handwriting and spelling vs. phonological awareness) in terms of gender, race,
students with special needs, and performance on the spelling and handwriting
screening measures (all p’s [ 0.24). With two exceptions, there were also no
statistically significant differences between students in the treatment and control
conditions on all other measures administered before instruction started (all
p’s [ 0.24). These assessments are described in greater detail in the Measures
section. One of the exceptions involved our measure for sentence construction
fluency, the Woodcock Johnson Psychoeducational Battery III Writing Fluency
subtest (WJ Writing Fluency; Woodcock, McGrew, & Mather, 2001). The other
exception occurred for a paragraph handwriting fluency task administered at pretest
and posttest. With this measure students copied a short paragraph consisting of two
sentences. Students in the control condition had statistically higher (p = 0.002)
sentence construction fluency scores (M = 95.07; SD = 7.69) than treatment
students (M = 86.33; SD = 6.34). Control students also had statistically faster

123
The impact of supplemental handwriting and spelling…

(p = 0.023) handwriting fluency on the paragraph copying task (M = 26.13;

SD = 6.94) than students in the treatment condition (M = 20.93; SD = 3.77).

Writing instruction in participating students’ classes

Although students’ teachers did not deliver either the experimental or control
conditions (this was done by graduate assistants), it is important to describe the
writing instruction students received in their regular class, as it better contextualizes
our investigation (this was based on an interview and two observations of each
teachers’ class). All but three of the participating students 14 teachers indicated they
applied a process plus skills approach to teaching writing. They noted their students
wrote stories, letters, journal entries, paragraphs, responses to reading material, and
informational text. They further indicated they taught spelling and handwriting (this
usually involved the use of commercial materials). Sentence construction skills
were reportedly taught in three of these 11 classes. Two of these 11 teachers
indicated students in their classes spent only 20 min a week writing, with the rest of
the time spent teaching writing skills, whereas the other nine teachers indicated they
balanced time spent on writing and teaching writing skills.
The three teachers who did not apply a process plus skills approach to teaching
writing reported their students did very little writing, as they devoted most of their
“writing time” to teaching handwriting, spelling, and/or sentence construction. Two
of these teachers indicated they used a commercial handwriting and spelling
program.
The 14 teachers varied in how much time they devoted to writing and writing
instruction each day, with four teachers devoting an hour a day. The remaining
teachers indicated they spent 20–40 min a day on writing.
Thus, even though students who participated in this study evidenced a slow rate
of progress in acquiring handwriting and spelling skills, all of their teachers taught
handwriting and spelling. Most but not all students wrote stories as part of regular
classroom instruction, and students in about 40 of the classes received sentence
construction instruction.

Measures

Before instruction began and immediately after it ended, students were administered
tests to measure their handwriting fluency, handwriting legibility, spelling accuracy,
sentence construction fluency, and composition length, vocabulary, and quality. All
tests were administered by trained graduate students who were also involved in
providing instruction. However, assessments were administered to each student by a
different graduate student than the one who acted as their instructor. In addition, the
order in which assessments were administered was randomized for each student.
This occurred at both pretest and posttest. All assessments were independently
scored by two trained raters (their scores were averaged).

123
 S. Graham et al.

Compositions

Students’ skills in writing compositions were assed with two different writing tasks.
One writing task asked students to write a story in response to a black-and-white
line drawing (e.g., a frowning turtle sitting in a tree with his arms crossed or a girl
running in a field with a rabbit and a butterfly). A second writing task involved
writing a description (e.g., a favorite place or favorite person in your family).
Prior to the start the study, the story and descriptive prompts were judged to be
appropriate for use with first-grade children by four primary grade teachers. For
stories, the assignment of the two line drawings was counterbalanced so that each
student responded to a different picture at pretest and posttest and an equivalent
number of children wrote about each picture at each testing time. This was also done
for the two descriptive writing prompts.
For both story and descriptive writing, students were directed to write in response
to the assigned prompt, they had as long as they needed to complete the paper, and
the examiner could not provide assistance. For story writing, students were also told
not to describe the picture when writing, but to create a story to go with it.
The resulting compositions were scored for composition length, composition
vocabulary, and composition quality. Composition length was computed by
counting the number of words included in a paper, regardless of the how words
were spelled. Composition vocabulary was the proportion of unique words seven
letters or longer (when the word was spelled correctly) included in a students’ paper.
This was determined by counting the number of letters in each word (when spelled
correctly) in a paper. This was calculated by tallying the number of unique words
seven letters or longer, and dividing this sum by number of different words. This
measure developed by Hammill and Larsen (1996) provided an indication of
vocabulary complexity, as longer words are likely to occur less frequently in young
students’ writing and be more difficult to spell (Graham, Harris, & Loynachan,
1993).
Composition quality was assessed using a traditional holistic writing measure.
Raters were asked to read each story (or description) attentively, but not laboriously,
to obtain a general impression of overall quality. Examiners were told that
imagination, organization, aptness of word choice, grammar, and sentence structure
should all be taken into account when forming a judgement about story quality and
that no one factor should receive undue weight. Students’ papers were scored on a
9-point scale, with higher scores representing higher quality of writing. These same
directions were given to raters scoring descriptions, except the word “imagination”
was replaced with “ideation.”
To guide the composition quality scoring process, raters were provided with a
representative paper for a score of 2, 4, 6, and 8. One set of representative papers
was provided for scoring stories and another for descriptions. These representative
papers or anchor points were obtained from students in four first-grade classes not
participating in the study, but in the same school district. Students in these classes
wrote a story and description following the same procedures applied in this
investigation. After reading through all of the stories and descriptions written in
these classes, two former elementary school teachers selected anchor papers for the

123
The impact of supplemental handwriting and spelling…

scoring points identified above for each type of writing separately. Their ordering of
these anchor points were confirmed independently by two first grade teachers.
Before any pre- or post-test story and descriptive paper was scored, it was typed
(removing all identifying information) and corrected for spelling, punctuation, and
capitalization. Students’ handwriting, spelling, and usage errors can bias raters’
perceptions of writing quality (Graham et al., 2011). Scoring occurred after all data
had been collected. Interrater scoring reliability (based on correlations between the
scores assigned by the two reviewers) for composition length, composition
vocabulary, and composition quality was 0.99 (0.99 for both stories and
descriptions), 0.95 (0.96 for stories and 0.95 for descriptions), and 0.84 (0.86 for
stories and 0.81 for descriptions), respectively.

Sentence construction fluency

The WJ Writing Fluency subtest (Woodcock et al., 2001) was used to assess
sentence construction fluency. This is a timed measure where students compose
sentences from three written words that go with a picture. A sentence is scored as
correct if all three words are used without change in a grammatically complete
sentence. The test–retest reliability for primary grade students reported in the test
manual is greater than 0.80, and scoring reliability in this study was 0.97.

Handwriting fluency

Two timed tasks were used to assess handwriting fluency. For each task, students
used a pencil without an eraser and primary-lined paper. They were told to cross out
and rewrite if they made mistakes. One of the tasks, alphabet handwriting fluency,
was developed by Abbott and Berninger (1993), and asked students to print the
entire lowercase alphabet as quickly as possible from memory without making any
mistakes. After 15 s, a slash was made after the last letter completed. Students were
then asked to finish the task. The score for this task was the number of letters written
correctly during the 15-s interval. Reversals, substitutions, omissions, additions, and
letters written in the wrong case (i.e., upper-case) were not counted as correctly
written letters. Interrater scoring reliability was 0.96.
The second task, paragraph handwriting fluency, was used previously by Graham
et al. (2000) and involved copying a two sentence paragraph. This paragraph
included the sentence used on the handwriting fluency screening measure (A little
boy lived with his father in a large forest.), but it was followed by a second sentence
(Every day the father went out to cut wood.). This paragraph was displayed at the
top of a page of primary-lined paper, and students copied the paragraph directly
below the display. The directions were identical to the directions for the handwriting
fluency screening measure. The score for this task was the number of letters copied
correctly. Again, reversals, substitutions, additions, and omissions were not counted
as correct letters. Interrater scoring reliability was 0.98.

123
 S. Graham et al.

Handwriting legibility

Students’ stories, descriptive compositions, and texts produced via the paragraph
handwriting fluency task were scored for handwriting legibility. They were scored
using the scale from the Test of Handwriting (Larsen & Hammill, 1989). The
reported test–retest reliabilities exceed 0.80 in the test manual. This legibility scale
provides scored samples of students’ handwriting that illustrate more to less legible
handwriting (higher scores represent more legible handwriting). The examiner
matches the sample of text produced by the student to the scored samples on the
scale to derive a legibility score. Interrater scoring reliability was 0.70.

Spelling

Three measures assessed students’ spelling accuracy. Spelling achievement was

assessed with the Test of Written Spelling—(TWS; Larsen & Hammill, 1999). The
TWS asks students to spell words that are consistently governed by the rules of
standard American English spelling as well as words that do not follow these rules.
Internal consistency reliability for this test, as reported in the TWS manual, is
greater than 0.90 in most studies. Interrater scoring reliability for this study was
0.99.
The second spelling measure, spelling in context, was derived from students’
compositions (stories and descriptions). In each paper, the number of words spelled
correctly was divided by the number of words written. Interrater scoring reliability
was 0.95.
The third spelling measure, spelling common words, asked student to spell 94
words from a list of spelling items compiled by Graham et al. (1993). The 94 words
represented words mostly commonly used by elementary grade students and were
appropriate for first grade writers; these words were taken from a running list of 10
million words children used in their writing. Students studied some, but not all, of
these words as part of instruction. All 94 words were tested at pre- and posttest.
Interrater scoring reliability was 0.99.

General instructional procedures

Instruction was delivered to students in the treatment and control conditions by six
graduate education majors blind to the purpose and hypotheses for the present study.
All instruction was delivered individually, with each instructor teaching students in
both conditions. Instruction was delivered as a supplement to the regular literacy
program.
Before the start of the study, instructors were taught how to implement both the
treatment and control conditions. For each condition, they were provided with a
notebook containing detailed directions for implementing all activities. The
materials for each condition were printed in different colors. Training took place
over a 2 week period (1 week for treatment and 1 week for control in that order). In
both conditions, training involved the first author providing an overview of the
material and stressing the value of instruction for helping students become better

123
The impact of supplemental handwriting and spelling…

writers and readers. It also involved the first author modeling lessons, and
instructors practicing them with each other until they could implement them
correctly. As they practiced lessons, they had a checklist for the elements of each
lesson, checking them as they were completed.
Instructors taught each of their assigned students three times a week for 20 min a
session. Instruction was delivered in a quite area in the school (e.g., library,
lunchroom).

Handwriting and spelling treatment condition

The program included eight units. In each unit, handwriting instruction focused on
two alphabet letters, whereas spelling instruction centered on two or more spelling
patterns (see below). There were six 20 min lessons in each unit. Each unit was
delivered over a 2 week period, with three lessons each week. Both handwriting and
spelling were taught in the first five lessons in a unit. Lesson six only involved
spelling instruction
Handwriting was taught in the first five lessons in each unit using the following
procedures. Each child was initially asked to write each lower-case letter of the
alphabet. Based on their performance on this assessment, the student was taught
how to form letters he or she produced incorrectly, inefficiently, or both. In each
unit, the student worked on two of these letters (e.g., i, t). During lessons one and
two in a unit, students were provided with a card for each letter that included
numbered arrows visually showing how to form the letter. After watching the
instructor trace and describe how to form each letter on the card, the student then
traced it and described how to form it. The student then discussed with the instructor
how the two letters were similar and different. After the student practiced tracing,
copying, and writing the two letters, the child circled his or her best-formed letters.
During lessons three and four, the student continued to practice tracing, copying,
and writing the two letters, but the child now practiced copying words that
contained each target letter (e.g., adding for the letter d). In lesson five, practice was
changed from copying words to copying “hinky-pinkys” that contained the target
letter (e.g., fuddy-duddy for the letter d). In lessons three to five, students circled
their best-formed word or hinky-pinky.
Each unit also included instruction on two or more spelling patterns involving
short vowel sounds, long vowel sounds, or both (the only exception was unit 1,
which just focused on short /a/). The sequence of skills taught was as follows: short /
a/ (unit 1); short /o/ and /e/ (unit 2); short /i/ and /u/ (unit 3); short and long /a/ (unit
4); short and long /i/ (unit 5); short and long /o/ (unit 6); short and long /e/ (unit 7),
and long /a/ in words like made, may, and maid (unit 8).
The first five lessons in each unit followed a set pattern that included four
different activities: phonics warm-up, word building, word study, and writing. The
sixth lesson in each unit involved a word sorting activity.
Phonics warm-up was designed to improve students’ skills in identifying letters
that correspond to sounds for consonants, blends, digraphs, and short vowels. Each
of the 46 target sound and letter combination(s) was depicted on a card with a
picture on one side of the card (e.g., a picture of a hat representing the sound /h/ at

123
 S. Graham et al.

the start of the word) and a corresponding letter (e.g., the letter h) on the other side
of the card. Students practiced identifying which letters made the sound represented
on the card (the target sound could be represented at the beginning, middle, or end
of the word), receiving feedback and additional practice when their response was
incorrect.
The word building activity was designed to improve spelling knowledge about
onset and rimes in words (a rime is the string of letters following the onset). In
lesson one, a rime (e.g., ate) that corresponded to one of the spelling patterns
emphasized in that unit (i.e., long /a/) was introduced on a card. After hearing the
sound made by the rime and observing how to combine the rime card with onset
cards containing a consonant (e.g., letter l) or a blend (e.g., Letters pl) to form
words, the student practiced forming words by combining the rime card with various
onset cards. In lesson two, the student repeated this process with the same rime, but
now built words on a worksheet instead of with cards. Lessons three and four were
identical to the first two lessons, except a new rime was introduced for the other
spelling pattern emphasized in that unit. During lesson 5, the student reviewed the
rimes introduced in the first four lessons, by using these rimes to build words on a
worksheet.
The purpose of the word study activity was to ensure students could correctly
spell words they were likely to use when writing. The words a student practiced
were either misspelled on the pretest measure, spelling common words, or
misspelled in the writing activity they completed as part of instruction (described
below). At the start of the study, each student was given five words to study. Each
word was written on a card and they were attached together with a ring. During
lessons one to five in each unit, the child studied the words attached to the ring by
saying it, studying its letters, writing it from memory, and checking whether it was
spelled correctly (and correcting it, if it was not). After a student spelled a word
correctly six times in a row over two lessons, the word was removed from the ring
and another previously misspelled word was added to the ring. Periodically,
students spelling of the words removed from the ring were retested, and if a word
was missed the student restudied it as needed.
With the writing activity, students completed writing tasks that encouraged them
to use words that fit the spelling patterns they were studying as they wrote a short
composition. For example, in unit 1 where the short vowel sound for /a/ was
emphasized, students were asked to write a paper about “A fat cat who is trying to
catch Mat the rat.” After writing the story, the child with help from the instructor
placed a star next to any word in the paper that fit the targeted spelling patterns. The
primary instructional purposes of the writing activity were to promote handwriting
fluency and provide an opportunity to apply spelling skills learned in context.
The word-sorting activity (based on Graham et al., 1996) was designed to help
students to learn the rule for each spelling pattern emphasized in the unit. For
instance, in unit 3 (short /i/ and /u/), the student participated in a word sort where
were placed into two categories: CVC-type words containing the short vowel sounds
of /i/ (e.g., sit) and /u/ (e.g., fun).
The word sorting activity began with the student observing the instructor placing
master word cards for each spelling pattern next to each other (e.g., sit and fun).

123
The impact of supplemental handwriting and spelling…

Each master word was then pronounced twice, with emphasis placed on the target
spelling feature in the word (e.g., short vowel sound for /i/ in sit and /u/ in fun).
Next, the student considered how the master words were similar and different. The
student’s attention was focused on critical features in both words, such as how the
letter “i” is pronounced in each master word and the combination of consonants and
vowels in each master word. The word sorting activity continued with the student
working with the instructor to decide under which master word they should place an
additional 12 words, with the idea of figuring out the rule for why the letter “i”
makes the short /i/ sound in sit, and the letter “u” makes the short /u/ sound in fun.
This continued until all words were sorted. The child (with help from the instructor
if needed) generate a rule for each spelling pattern emphasized in that word sort.
The student then generated words of their own, writing them on blank word cards
and placing them into the appropriate category.

Phonological awareness control condition

The phonological awareness control condition also involved eight units. There were
also six 20 min lessons in each unit. All phonological awareness activities were
taken from the Ladders to Literacy Program (O’Connor, Notari-Syverson, &
Vadasy, 1998) or based on Rosner’s (1979) Auditory Analysis Program. The only
exception was the Reading activity which involved common folktales. Each lesson
in a unit involved five activities: Guess What Word, Sound Play, Rhyming Triplets,
Sound Songs, and Reading. Starting in lesson 34, Magic Squares took the place of
Rhyming Triplets.
With Guess What Word, the student listened as a target word was spoken aloud
in the following three ways: (1) stretching the syllable (e.g., hat as haaat), (2) saying
the onset and the rime separately (e.g., /h /at/), and (3) saying each phoneme
separately (/h a t/). The child then identified the word (from a list of four words).
Next, the student was directed to say the word, phoneme-by-phoneme.
During Sound Play, the student played four sound games. Sound games in the
first two units involved syllables. In Find the Hidden, the student was asked if a
specific syllable was hidden in a word (e.g., “Is the word ant hidden in the word
anthill?”). With Say the Missing, the child figured out which syllable a word was
missing (e.g., “Say Carelessly; now say lessly: what is missing?”). For Say the Word
Without, the child repeated a word without one of its syllables (e.g., “Say
friendliness; now say it without ness.”). With Substitute, one syllable was
substituted for another in a word (e.g., “Say party; now don’t say it with y say it
with ner.”).
Sound Play in the remaining six units applied the same four tasks, but focused on
phonemes, including finding where specific phonemes were hidden (e.g., “Does the
word big begin with a b sound?”), figuring out which sound was missing from a
word (e.g., “Say tin; now say in; what is missing?”), repeating a word without one
of its sounds (e.g., “Say tin; now say it without /t/.”), and substituting one phoneme
for another in a word (e.g., “Say tin; now don’t say it with /t/ say it with /k/.”).
With Rhyming Triplets, a pair of rhyming words were spoken aloud (e.g., cake-
snake) and the child was asked to repeat them and identify other words that rhymed

123
 S. Graham et al.

with them. Starting in lesson unit six, Rhyming Triplets was replaced by Magic
Squares. With this new activity, a word was said aloud (e.g., dog), and then said
phoneme by phoneme (e.g., /d /o/ /g/). The student then said the word phoneme by
phoneme with the instructor and identified how many sounds were in the word. The
student and instructor then said the sounds in the words again, with the student
sequentially touching a square in an Elkonian box as each sound was produced.
With Sound Songs, students substituted the letter sound of the day (from the
second activity) with one of the sounds in well-known songs (i.e., Old MacDonald,
Jimmy Cracked Corn, and the Birthday Song).
With the final activity, Reading, the student listened as a folktale was read aloud.
The reading activity was interactive as the student was asked questions about the
cover of the book (e.g., What do you see on the cover? What do you think the book
will be about? Why do you think so?) and material presented within it. Once the
folktale had been read, the student described what they liked best about the book
and if they would recommend it to a child their age.
It should be noted that none of the activities in the phonological awareness
condition involved presenting written letters or words. This would have moved the
control condition into the realm of spelling instruction. The spelling activities,
however, were likely to promote phonological awareness, as they involved
matching letters to the sounds in words, likely making students more aware that
words are composed of separate sounds. Spelling instruction did enhance
phonological awareness in Graham and Santangelo (2014).

Fidelity of instructional implementation

To ensure both conditions were delivered as planned, the following safeguards were
implemented. One, instructors received intensive practice applying each condition
(as noted). Two, instructors met with the first author weekly to discuss any glitches
that occurred during implementation. Reported glitches were rare, and when they
did occur usually involved an inadvertent mistake (e.g., skipping a step of
instruction). Three, instructors checked each step of a lesson on a recording sheet as
it was completed. For handwriting and spelling instruction 99.1% of steps were
checked (ranging from 98.0 to 99.7% across instructors), whereas 99.8% of steps
were checked for phonological awareness instruction (ranging from 99 to 100%
across instructors). Four, in both the treatment and control condition 8 lessons (17%
of lessons) were tape recorded for each child. A trained graduate student listed to all
taped sessions (240 sessions) and checked which steps of instruction had been
completed (using the same checklist instructors applied during instruction). A
second graduate student independently scored 60 sessions (25%). Interrater
reliability was 97%. The percentage of steps completed on the tape recorded
lessons for handwriting and spelling was 97.4% (93.2–100%) across instructors. For
phonological awareness, percentage of steps completed was 98.1% (93.8–100%
across instructors).

123
The impact of supplemental handwriting and spelling…

Results

Because a single piece of writing provides a less reliable estimate of performance

than multiple pieces (Graham, Hebert, Sandbank, & Harris, 2016), we calculated an
average score for composition length, composition vocabulary, and composition
quality for the two writing tasks (i.e., story and description) at each testing point.
Similarly, handwriting legibility was the average score for legibility across the two
writing tasks (story and description) and the paragraph handwriting fluency task
administered at each testing point. In addition, for each measure we conducted a
One-Way ANOVA using gain scores. Despite randomization, there were some
statistical differences between students in the two groups at pretest (see the
Participant section), and gain scores allowed us to adjust for these and non-
statistical differences on all other measures. An effect size (d) is reported for all
statistically significant findings. For spelling achievement (measured using the TWS)
and sentence construction fluency (measured via the WJ—Writing Fluency subtest),
standard scores were used in all analyses. Raw scores were used for all other
analyses, as measures involved research designed tests (Table 1).

Does supplemental handwriting and spelling instruction improve students’

handwriting and spelling?

Handwriting

When compared to students receiving phonological awareness instruction, students

who received supplemental handwriting and spelling instruction made greater gains
on the alphabet handwriting fluency and paragraph handwriting fluency measures.
On the paragraph handwriting fluency measure, treatment students made statistically
greater gains (M = 15.93; SD = 7.44) than control students (M = 6.67; SD = 12.98)
from pretest to posttest in the number of letters copied correctly, F(1,28) = 5.76,
p = 0.023, d = 0.81. Likewise, treatment students (M = 2.07; SD = 1.71) made
statistically greater gains from pretest to posttest than controls (M = 0.13;
SD = 1.92) on the alphabet handwriting fluency measure, F(1,28) = 8.47,
p = 0.007, d = 0.95.
Improved handwriting fluency for the treatment students was also accompanied
by better handwriting legibility. Treatment students made statistically greater gains
from pretest to posttest (M = 0.27; SD = 1.33) than control students (M = − 0.58;
SD = 0.83) in handwriting legibility, F(1,28) = 4.34, p = 0.046, d = 0.72.
However, the statistically significant difference is likely due to the decline in the
legibility of control students’ written products, as there was not a statistically
significant gain in handwriting legibility from pretest to posttest when only the
performance of treatment students was analyzed (p = 0.55).

123
 S. Graham et al.

Table 1 Means and standard deviations for measures by condition and time
Measures Pretest Posttest Gain

Alphabet handwriting fluency

Treatment 3.73 (1.98) 5.80 (1.15) 2.07 (1.71)
Control 4.40 (1.68) 4.53 (2.03) 0.13 (1.92)
Paragraph handwriting fluency
Treatment 20.93 (3.77) 36.87 (6.32) 15.93 (7.44)
Control 26.13 (6.94) 32.80 (16.51) 6.67 (12.98)
Handwriting legibility
Treatment 2.82 (1.32) 3.09 (1.21) 0.27 (1.33)
Control 3.71 (1.10) 3.13 (0.94) − 0.50 (0.83)
Spelling achievement
Treatment 81.40 (4.93) 93.67 (6.64) 13.60 (4.39)
Control 83.33 (5.77) 91.93 (4.77) 8.93 (7.05)
Spelling common words
Treatment 27.73 (12.26) 51.13 (4.69) 23.40 (10.83)
Control 34.13 (8.38) 48.20 (5.25) 14.07 (10.07)
Spelling in context (errors)
Treatment 0.32 (0.18) 0.19 (0.12) 0.13 (0.20)
Control 0.35 (0.14) 0.28 (0.22) 0.08 (0.18)
Sentence construction fluency
Treatment 86.33 (6.34) 91.40 (9.10) 5.07 (7.69)
Control 95.07 (7.69) 94.27 (3.13) − 0.80 (6.66)
Composition length
Treatment 28.20 (19.90) 58.60 (34.52) 30.40 (41.11)
Control 28.20 (20.25) 53.12 (39.30) 24.93 (39.18)
Composition vocabulary
Treatment 0.05 (0.06) 0.11 (0.07) 0.06 (0.08)
Control 0.08 (0.07) 0.07 (0. 04) − 0.01 (0.06)
Composition quality
Treatment 2.53 (1.04) 3.63 (1.32) 1.10 (1.31)
Control 2.97 (0.97) 3.80 (1.18) 0.83 (0.98)

The spelling achievement and sentence construction fluency measures involved norm-referenced tests and
standard scores are reported; spelling in context gain score represents improvement in reduced proportion
of spelling errors

Spelling

Treatment students made greater gains than control students on two of the three
spelling measures. Students who received supplemental handwriting and spelling
instruction made statistically greater pretest to posttest gains (M = 13.60; SD = 4.39)
than control students in the phonological awareness control condition (M = 8.93;
SD = 7.05) on spelling achievement measures (i.e., the TWS), F(1,28) = 5.76,
p = 0.038, d = 0.75. Further, treatment students learned the correct spelling of more
words (M = 23.40; SD = 10.83) from pretest to posttest than control students

123
The impact of supplemental handwriting and spelling…

(M = 14.07; SD = 10.07) on the spelling common words measure, F(1,28) = 5.97,

p = 0.021, d = 0.82. Treatment (M = 0.13; SD = 0.20) and control students
(M = 0.08; SD = 0.18) did not differ statistically on the spelling in context measure,
with both groups reducting the proportion of spelling errors they made in their
writing.

Does supplemental handwriting and spelling instruction improve students’

sentence construction fluency?

Transfer effects from supplemental handwriting and spelling instruction were

obtained for sentence construction fluency (as measured by the WJ Writing Fluency
test). Treatment students made statistically greater gains (M = 5.07; SD = 7.69)
than control students (M = − 0.80; SD = 6.66) from pretest to posttest on this
measure, F(1,28) = 4.99, p = 0.034, d = 0.77.

Does supplemental handwriting and spelling instruction improve students’

composition length, composition vocabulary, and composition quality?

Transfer effects from supplemental handwriting and spelling instruction were

obtained for one of the composition measures. Treatment students (M = 0.06;
SD = 0.08) made statistically greater gains from pretest to posttest than controls
(M = − 0.01; SD = 0.06) on the composition vocabulary measure, F(1,28) = 8.10,
p = 0.008, d = 0.91). Contrary to predictions, transfer effects from supplemental
handwriting and spelling instruction to composition length and composition quality
were not obtained. Treatment (M = 30.40; SD = 41.11) and control students
(M = 24.93; SD = 39.18) did not differ statistically in pretest to posttest gains in
composition length, nor did treatment students (M = 1.10; SD = 1.31) and controls
(M = 0.83; SD = 0.98) differ statistically in composition quality gains.

Discussion

Writing is a complex, demanding, and challenging task (Kellogg, 1993). Writers

must become familiar with the purposes, norms, values, and practices of the various
writing communities in which they participate as well as master the specialized
writing knowledge and beliefs needed to be successful in these communities
(Graham, in press). Young children who are just starting their journey in learning
how to write at school face an especially difficult challenge. Writing skills such as
handwriting and spelling that more advanced writers apply with little effort are
highly demanding and effortful for beginning writers (Berninger, 1999; Graham &
Harris, 2000). This influences not only how young children write, as they minimize
their use of other attention demanding writing processes such as planning,
monitoring, and evaluating (McCutchen, 1988; Scardamalia & Bereiter, 1986), but
can result in cognitive overload and interference. This may be particularly
problematic for young children who do not acquire text transcription skills as
quickly as their peers. This study directly tested if the handwriting and spelling of

123
 S. Graham et al.

these children can be improved by providing supplemental instruction to improve

these skills, and if such instruction resulted in better sentence construction fluency
as well as improvements in composition length, composition vocabulary, and
composition quality.

The impact of supplemental instruction on students’ handwriting

and spelling

Students participating in this study all scored at or below the bottom 25th percentile
on screening measures of spelling and handwriting fluency competence. Each
student in the experimental condition received one-on-one instruction from a trained
tutor using instructional procedures for teaching handwriting and spelling that were
successful in previous studies with beginning writers (e.g., Graham, 1999; Graham
et al., 2000, 2002). This supplemental instruction was in addition to the handwriting
and spelling instruction provided by their regular classroom teachers, which
generally involved the use of commercial programs to teach these skills.
The combined supplemental handwriting and spelling instruction these children
received had a positive impact on their handwriting and spelling. In comparison to
the control condition where students received one-on-one phonological awareness
instruction, students in the experimental treatment condition made greater gains on
two handwriting fluency measures (i.e., measures of alphabet and paragraph
handwriting fluency), a measure of handwriting legibility (although these gains may
have mainly been a consequence in the decline of control students’ scores), and two
measures of spelling accuracy (i.e., spelling achievement as measured by a norm-
referenced measures and spelling common words). The observed gains were
relatively large, with effect sizes of 0.81, 0.95, and 0.72 for paragraph handwriting
fluency, alphabet handwriting fluency, and handwriting legibility, respectively.
Similar effects were obtained for the two statistically significant spelling accuracy
measures, with and effect size of 0.75 and 0.82 for the spelling achievement and
spelling common words measures, respectively.
These spelling and handwriting gains compared favorably to gains made by
students who were taught handwriting separately or spelling separately in recent
meta-analyses. Santangelo and Graham (2016) found handwriting instruction
produced an average ES of 0.63 for handwriting fluency and 0.59 for handwriting
legibility. Graham and Santangelo (2014) reported spelling instruction produced an
ES of 0.54 for spelling. While caution must be exercised in comparing the results of
a single study such as ours to findings from a meta-analysis (e.g., studies may differ
in terms of types of students or how handwriting and spelling are taught), the
findings from the two published meta-analyses provide a general benchmark for
thinking about the effectiveness of our study.
It is important to note that statistically greater gains were not obtained for the
handwriting and spelling treatment on one of the spelling measures: spelling in
context. Treatment students did not make greater reductions in proportion of
spelling errors made when writing stories and descriptions when compared to
phonological awareness controls. One possible reason for this is that the average
length of students’ compositions were generally short at both pretest and posttest

123
The impact of supplemental handwriting and spelling…

(28 and 55 words, respectively), and the number of different words was even less
(fewer than 10 and 20 different words, respectively). As a result, the opportunity to
showcase newly acquired spelling skills was limited. We obtained some support for
this explanation by conducting a post hoc analysis (reported here), where we found
that treatment students had lower spelling error scores than controls (p = 0.045),
once pretest spelling errors and number of unique words at posttest were statistically
controlled.
A second possible explanation for why treatment students did not make greater
statistical gains in reducing spelling errors is that phonological awareness
instruction improved spelling skills (Graham et al., 2017). This was borne out in
this investigation, as control students evidenced improved scores on all spelling
measures from pretest to posttest. Consequently, they may have made enough
spelling improvement to blunt any possible statistical difference with treatment
students with regards to spelling in context.

Impact of supplemental handwriting and spelling instruction on other

writing measures

Until handwriting and spelling are mastered, they can theoretically lead to cognitive
overload and interference when writing (Berninger, 1999; Graham, in press). For
young writers who are just beginning to learn these skills, this can result in the
development of a restricted approach to writing that minimizes the use of other
attention demanding writing processes (McCutchen, 1988). It can also lead to
students restricting the words they use in writing to ones they already know how to
spell correctly. Problems with cognitive overload, interference, and restriction may
be especially problematic for beginning writers who do not master handwriting and
spelling as quickly as their peers, leading to difficulties with sentence construction
as well as shorter compositions, with less diverse vocabulary, and poorer overall
writing quality.
The results of the current study provided partial support for these theoretical
propositions. When compared to peers who received phonological awareness
instruction, treatment students made greater gains in sentence construction fluency
(as measured by the WJ—Writing Fluency subtest) and composition vocabulary (i.e.,
proportion of words seven letters or longer in their compositions). The observed
gains were relatively large, with effect sizes of 0.77 and 0.91 for sentence
construction fluency and composition vocabulary, respectively. The transfer gain for
sentence construction fluency compares favorably with gains reported in prior
studies; Santangelo and Graham (2016) reported in their meta-analysis that
handwriting instruction produced an ES of 0.48 for writing fluency. No previous
studies have examined the impact of teaching handwriting or spelling on
composition vocabulary, so comparable effects from the literature for this variable
were not available.
We did not find statistically significant transfer effects from handwriting and
spelling instruction to composition length or composition quality, contrary to our
predictions. In contrast, Graham and Santangelo (2014) reported that handwriting
instruction enhanced composition length (ES = 1.33) and composition quality

123
 S. Graham et al.

(ES = 0.84). So why did handwriting and spelling instruction not enhance these two
aspects of writing in this study? One possible reason is that students in this study
differed from students in prior studies where positive results were obtained. This
seems unlikely, as prior studies by Berninger et al. (1997) and Graham et al. (2000)
worked with similar students and obtained stronger results. A more likely
explanation is that transfer effects to composition length and composition quality
are due mostly to handwriting and not spelling instruction. Spelling instruction did
not improve writing performance in prior studies in the meta-analysis conducted by
Graham and Santangelo (2014). Thus, the inclusion of spelling instruction in this
study may have weakened transfer effects to these two measures of writing
performance, as it reduced the amount of handwriting instruction provided.

Future research and implications for practice

We found that supplemental handwriting and spelling instruction improved the

handwriting and spelling skills of first grade students who did not acquire these
skills as rapidly as their peers. Further, this instruction had a significant and
meaningful impact on students’ sentence construction fluency and composition
vocabulary, but did not statistically enhance the length and quality of their
compositions. Additional research is needed to replicate and extend the findings
from the current study. For example, will supplemental instruction that combines
the teaching of handwriting and spelling produce similar results with another sample
of students similar to those who participated in this study? Will increasing the
amount of time in handwriting instruction in such combined instruction result in
meaningful impacts on composition length and quality? Is such instruction effective
with older students? Is it better to combine supplemental handwriting and spelling
instruction versus just concentrating on one or the other? How can supplemental
handwriting and spelling instruction for beginning writers be improved so that it
positively reduces spelling errors in context and increases writing output and
quality?
Practically, this study demonstrated that providing 1 h a week of supplemental
handwriting and spelling instruction can have positive effects on first grade
students’ transcription and writing skills. As a result, we strongly recommend that
teachers provide such instruction to students in their class who do not acquire these
skills as easily as their peers.
It is important to note that our study was not designed to determine if a combined
handwriting and spelling instructional program is better than separate instruction in
these two skills. We hope that such an investigation will be undertaken in the future.
When such a study is conducted, we recommend that it include four conditions: a
handwriting treatment, a spelling treatment, a combined handwriting and spelling
treatment, and a control treatment that receives useful instruction in an area like
math that is unlikely to enhance any aspect of students’ writing performance. It is
critical that such a study provides an equal amount of instruction in each condition.
In closing, we would like to draw attention to several limitations that should be
taken into consideration when interpreting the findings from our investigation. This

123
The impact of supplemental handwriting and spelling…

includes a relatively small sample of students, limiting the power of the study to
detect possible statistical differences between conditions, especially for transfer
measures to writing. The small sample size may have also limited the power of
randomization, as we found that performance on the paragraph handwriting fluency
measure and the sentence construction fluency measure favored students in the
control condition at the start of the study. It is not clear how these differences
affected the outcomes obtained. Further, the study did not provide equal time to
handwriting and spelling instruction. This should be corrected in future studies.
Lastly, it is not clear if the effects of handwriting and spelling instruction linearly
impacted the outcome measures. This determination is dependent on future research
in this area.

References
Abbott, R., & Berninger, V. (1993). Structural equation modeling of relationships among developmental
skills and writing skills in primary and intermediate-grade writers. Journal of Educational
Psychology, 85, 478–508.
Alves, R., Limpo, T., Fidalgo, R., Carvalhais, L., Pererira, L., & Carlio, S. (2016). The impact of
promoting transcription on early text production: Effects on bursts and pauses, levels of written
language, and writing performance. Journal of Educational Psychology, 108, 665–679.
Berninger, V. (1999). Coordinating transcription and text generation in working memory during
composing: Automatic and constructive processes. Learning Disability Quarterly, 22, 99–112.
Berninger, V., Mizokawa, D., & Bragg, R. (1991). Scientific practitioner: Theory-based diagnosis and
remediation of writing disabilities. Journal of School Psychology, 29, 57–79.
Berninger, V. W., Vaughan, K. B., Abbott, R. D., Abbott, S. P., Rogan, L. W., Brooks, A., et al. (1997).
Treatment of handwriting problems in beginning writers: Transfer from handwriting to composition.
Journal of Educational Psychology, 89, 652–666.
Berninger, V. W., Vaughan, K., Abbott, R. D., Brooks, A., Abbott, S., Rogan, L., et al. (1998). Early
intervention for spelling problems: Teaching functional spelling units of varying sizes with a
multiple-connection framework. Journal of Educational Psychology, 90, 587–605.
Berninger, V., Vaughan, K., Abbott, R., Begay, K., Byrd, K., Curtain, G., Minnich, J., & Graham, S.
(2002). Teaching spelling and composition alone and together: Implications for the simple view of
writing. Journal of Educational Psychology, 94, 291–304.
Case-Smith, J., Holland, T., & White, S. (2014). Effectiveness of a co-taught handwriting-writing
program for first grade students. Physical & Occupational Therapy in Pediatrics, 34, 30–43.
Department for Education. (2012). The research evidence on writing. London, England: Department for
Education. Retrieved From https://www.gov.uk/government/uploads/system/uploads/attachment_.
Graham, S. (1999). Handwriting and spelling instruction for students with learning disabilities: A review.
Learning Disability Quarterly, 22, 78–98. https://doi.org/10.2307/1511268.
Graham, S. (2006). Writing. In P. Alexander & P. Winne (Eds.), Handbook of educational psychology
(pp. 457–478). Mahwah, NJ: Erlbaum.
Graham, S. (in press). A writer(s) within community model of writing. In C. Bazerman, V. Berninger, D.
Brandt, S. Graham, J. Langer, S. Murphy, P. Matsuda, D. Rowe, & M. Schleppegrell (Eds.), The
lifespan development of writing. Urbana, IL: National Council of English.
Graham, S., & Harris, K. (2000). The role of self-regulation and transcription skills in writing and writing
development. Educational Psychologist, 35, 3–12.
Graham, S., Harris, K. R., & Fink, B. (2000). Is handwriting causally related to learning to write?
Treatment of handwriting problems in beginning writers. Journal of Educational Psychology, 92,
620–633.
Graham, S., Harris, K. R., & Fink-Chorzempa, B. (2002). Contributions of spelling instruction to the
spelling, writing, and reading of poor spellers. Journal of Educational Psychology, 94, 669–686.

123
 S. Graham et al.

Graham, S., Harris, K. R., & Hebert, M. (2011). It is more than just the message: Presentation effects in
scoring writing. Focus on Exceptional Children, 44(4), 1–12.
Graham, S., Harris, K. R., & Loynachan, C. (1993). The basic spelling vocabulary. Journal of
Educational Research, 86, 363–368.
Graham, S., Harris, K. R., & Loynachan, C. (1996). The directed spelling thinking activity: Application
with high frequency words. Learning Disabilities Research and Practice, 11, 34–40.
Graham, S., Hebert, M., Sandbank, M., & Harris, K. R. (2016). Credibly assessing the writing
achievement of young struggling writers: Application of generalizability theory. Learning Disability
Quarterly, 39, 72–82.
Graham, S., Liu, K., Bartlett, B., Ng, C., Harris, K. R., Aitken, A., et al. (2017). Reading for writing: A
meta-analysis of the impact of reading and reading instruction on writing. Review of Educational
Research. https://doi.org/10.3102/0034654317746927.
Graham, S., & Santangelo, T. (2014). Does spelling instruction make students better spellers, readers, and
writers? A meta-analytic review. Reading & Writing: An Interdisciplinary Journal, 27, 1703–1743.
Graham, S., & Weintraub, N. (1996). A review of handwriting research: Progress and prospect from l980
to l994. Educational Psychology Review, 8, 7–87.
Hammill, D., & Larsen, S. (1996). Test of written language—3. Austin, TX: Pro-ED.
Kellogg, R. (1993). The psychology of writing. New York: Oxford University Press.
Larsen, S., & Hammill, D. (1989). Test of legible handwriting. Austin, TX: Pro-ED.
Larsen, S., & Hammill, D. (1999). Test of written spelling—4. Austin, TX: Pro-ED.
McCutchen, D. (1988). “Functional automaticity” in children’s writing: A problem of metacognitive
control. Written Communication, 5, 306–324.
Micahelowa, K. (2001). Primary education quality in Francophone Sub-Saharan Africa: Determinants of
learning achievement and efficiency considerations. World Development, 29, 1699–1716.
National Center for Education Statistics. (2012). The Nation’s Report Card: Writing 2011 (NCES 2012–
470). Washington, D.C.: Institute of Education Sciences, U.S. Department of Education.
O’Connor, R., Notari-Syverson, A., & Vadasy, P. (1998). Ladders to literacy: A kindergarten activity
book. Baltimore: Brookes.
Research Center for Psychological and Educational Testing (RCPET). (2016). 105年國中教育會考各科
計分與閱卷結果說明 [2016 comprehensive assessment program for junior high school students:
Explanation of grading results]. Retrieved from http://www.cap.ntnu.edu.tw/exam/105/
Pressrelease1050603.pdf.
Rieben, L., Ntamakiliro, L., Gonthier, B., & Fayol, M. (2005). Effects of various early writing practices
on reading and spelling. Scientific Studies of Reading, 9, 145–166.
Rosner, J. (1979). Helping children overcome learning disabilities (2nd ed.). New York: Walker.
Santangelo, T., & Graham, S. (2016). A comprehensive meta-analysis of handwriting instruction.
Educational Psychology Review, 28, 225–265.
Scardamalia, M., & Bereiter, C. (1986). Written composition. In M. Wittrock (Ed.), Handbook of research
on teaching (3rd ed., pp. 778–803). New York: MacMillan.
Uhry, J. K., & Sheperd, M. J. (1993). Segmentation/spelling instruction as part of a first-grade reading
program: Effects on several measures of reading. Reading Research Quarterly, 28, 219–233.
Wechsler, D. (2009). Wechsler individual achievement test—III. San Antonio, TX: Psychological
Corporation.
Weintraub, N., Yinon, M., Hirsch, I., & Parush, S. (2009). Effectiveness of sensorimotor and task-
oriented handwriting intervention in elementary school-aged students with handwriting difficulties.
OTJR: Occupation Participation, and Health, 29, 125–134.
Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). Woodcock-Johnson III tests of achievement.
Itasca, IL: Riverside Publishing.

123