Effects of stimulus distance on children's handwiting copying performance


Material Information

Effects of stimulus distance on children's handwiting copying performance
Physical Description:
xii, 180 leaves : ill. ; 28 cm.
Walker, Kay Frances, 1942-
Publication Date:


Subjects / Keywords:
Penmanship   ( lcsh )
Children -- Writing   ( lcsh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1990.
Includes bibliographical references (leaves 166-179).
Statement of Responsibility:
by Kay Frances Walker.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 22579330
System ID:

Full Text








The writer thanks all those who assisted and supported

her efforts in the completion of this dissertation.

Specifically, thanks are due to colleagues, supervisory

committee members, research and manuscript preparation

assistants, and personal friends and family.

Colleagues in the Department of Occupational Therapy

and the College of Health Related Professions are thanked

for their support of my leave from work that was necessary

to allow time to complete this dissertation. Jane

Slaymaker, M.S., O.T.R./L., Associate Professor, made the

personal sacrifice of postponing her retirement and move to

Arizona in order to serve as Acting Chairperson during my

absence from the University. Richard R. Gutekunst, Ph.D.,

Dean, College of Health Related Professions, provided

unwavering support of my doctoral studies. The faculty and

staff in the Department of Occupational Therapy gave me

encouragement and carried on the functions of the Department

in my absence.

The supervisory committee members gave me excellent

feedback and guidance in the proposal, research, analysis,

and write-up of the project. Mary Kay Dykes, Ph.D.,

Professor of Special Education, served as my chairperson.

She was consistently challenging, fair, supportive, and

insightful. Committee member, Vivian Correa, Ph.D.,

Associate Professor, was especially helpful in the

generation of the graphics for the figures. William

Wolking, Ph.D., Professor, shared his expertise in single

subject research analysis. Cecil Mercer, Ed.D., Professor,

helped me to examine the classroom and clinical implications

of the project. Aside from these professors in the

Department of Special Education, Floyd Thompson, Ph.D.,

Associate Professor from the Department of Neuroscience,

provided a link to the medical science aspects of my

project. Many thanks go to these persons for their

patience, efforts, and support of my work.

Without the technical assistance of several persons,

the completion of this project would not have been possible.

At Terwilliger Elementary School, numerous persons provided

assistance. R. M. Craig, Ed.D., principal, was supportive

of the project. Beverly Haynes, librarian, and Deloros

Utley, curriculum resource teacher, generously adjusted

their schedules to provide space for conducting the project.

Mr. Sinclair Holmes, custodian, helped to set up the

furniture needed for the data collection. Classroom

teachers, Joanne Blair, Mary Ellen Lehman, Ellen Abramson,

and M. E. Harper, and varying exceptionalities teacher,


Kathryn Harvin, cooperated in referring subjects for the

study and in scheduling them for the data collection.

Research assistants, Sylvia Bamburg, Richard Schreidell,

Lisa Elmhurst, and Jennifer Boynton, collected the data.

Jennifer Staer scored the word spacing and Carolyn Harris

and Jolena Stoutimore did the scoring for the interrater

agreements. The names of the 13 children in the study will

not be mentioned here to protect their confidentiality, but

thanks are due to the children and to their parents.

Finally, the typing skills of Leila Cantara transformed my

written drafts into a polished product. All of these

persons are thanked for their time and help and for the part

they had in completing parts of this project that were

beyond my efforts.

My friends and family provided unlimited support and

interest in this endeavor. My friend, Horace Sawyer, Ed.D.,

provided proofreading and shared the wisdom of his

experience with the dissertation process. Sandra Adams, my

friend and business partner, was understanding and

congratulatory. My children, Katie and Suzanne Sieg, were

understanding when I needed to work on this project and

could not spend time with them. My siblings, Faye Anderson-

Copier, Imogene Walker, Jamie Walker, Mona Hurst, Marcia

Hurst, and Diane Buss, were loving, interested, and

thoroughly supportive of my efforts. Finally, my mother's

pride in this goal for me was an enduring source of

inspiration. With admiration, gratitude, and love, this

dissertation is dedicated in loving memory to my mother,

Katie Helen Clemons Walker.



ACKNOWLEDGEMENTS ........................... ........... ii

LIST OF TABLES ..... .............................. ix

LIST OF FIGURES .................................... x

ABSTRACT ........................................... xi



Introduction .............................. 1
Statement of the Problem ....................... 4
Purpose of the Study ....................... 5
Rationale of the Study ...................... 5
Handwriting Problems ...................... 6
Handwriting Instruction ................... 7
Analysis of Copying ....................... 8
Definition of Terms ........................ 13
Delimitations of the Study ................. 16
Limitations of the Study ..................... 18
Summary ...................................... 18

II REVIEW OF THE LITERATURE .................... 20

Introduction ................................. 20
Selection of Relevant Literature ............ 20
Handwriting as a Neuropsychological Process .. 23
Neuropsychological Model of Handwriting ... 25
Motor Aspects of Graphomotor Skills ....... 30
Visual Aspects of Graphomotor Skills ...... 44
Attentional Mechanisms for Graphomotor
Skills ................................. 48
Qualitative Investigation of Copying
Behavior ............ ........... ..... 51
Handwriting Evaluation ...................... 53
Validity .................................... 53
Reliability ............................... 55
Usability ............. ..... ....... ......... 56
Handwriting Instruction ....................... 59

Background Information .....
Single Subject Research ....
Summary ..... ...............

III METHODS .....................................

Introduction ..................
Questions .....................
Subjects ......................
Setting .......................
Personnel .....................
Dependent Variables ...........
Independent Variables .........
Design ........................
Treatment of the Data .........
Summary ......................

Introduction ..................
Analysis of Data .............
Question 1 ....................
Level ......................
Variability ................
Trend ......................
Summary ................
Question 2 ..................
Level ......................
Variability ................
Trend .....................
Summary ...................
Summary .......................

V DISCUSSION ...................................

Introduction ................................
Review of Purpose, Literature, and Methods ...
Review of Purpose ........................
Review of Literature .....................
Review of Methods .........................
Review of Questions and Results ..............
Discussion and Implications ..................
Summary of Findings, Interpretation, and
Literature Support ........................
Summary of Findings ......................
Interpretation ...........................
Literature Support ........................
Problems and Limitations .....................
Classroom ................................
Neuropsychological Model .................


............... o
. o o oo **
. ... ....o..... .









IV RESULTS ................ .................... ..



. ............ ..


Suggestions for Further Research ............. 143
Subjects .................................. 143
Stimuli ................................... 144
Follow-Up Analysis ..................... 145
General Recommendations .................. 146


(EACH) ....................................... 149

B STIMULI FOR COPYING .... .......................... 153

C SUBJECTS' SCORES ............................. 162

REFERENCES ....................................... 166

BIOGRAPHICAL SKETCH ............................... 180



Table Pa2e

1 Subject Ages, Race, Vision, Services,
Intelligence Quotient, Handwriting Grades,
Achievement Test Percentiles, and District
Stanines .................................... 78

2 Percentage of Agreement of Rater Pairs Prior
to and During the Study ...................... 85

3 Alternating Treatments Design (ATD): Counter-
balancing Six Combinations of Order for the
Three Stimulus Distances, Nine Sentences, and
12 Days of the Study ......................... 98

4 Medians and Interquartile Ranges of
Letter Formation Scores for 0.5 m, 3 m,
and 6 m Stimulus Distances .................. 114

5 Medians and Interquartile Ranges of
Word Spacing Scores for 0.5 m, 3 m, and
6 m Stimulus Distances ...................... 115

6 Intercept Coefficients (y) and Slope
Coefficients (x) and t-ratios for Letter
Formation Trends for 0.5 m, 3 m, and 5 m
Stimulus Distances .......................... 116

7 Intercept Coefficients (y) and Slope
Coefficients (x) and t-ratios for Word
Spacing Trends for 0.5 m, 3 m, and 6 m
Distances ................................. 117


Figure Page

1 Neuropsychological model for graphomotor
skills ................... ....... ............ 26

2 Boxplots for letter formation ............... 104

3 Lineplots for letter formation .............. 106

4 Boxplots for word spacing ................. 109

5 Lineplots for word spacing .................. 111

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Kay Frances Walker

May, 1990

Chairman: Mary K. Dykes
Major Department: Special Education

Copying tasks are routinely used for handwriting

instruction. In a neuropsychological model of graphomotor

skills, handwriting copying tasks are shown as requiring

complex language, visual acuity, and motor skills. Although

handwriting has been studied in diverse fields, the

parameters for optimal copying performance have not been


An alternating treatments single subject design was

used to compare the effects of three stimulus distances on

handwriting legibility. The effects on level, variability,

and trend of letter formation and word spacing scores when

the copying stimulus was placed in front of the subject at

distances of 0.5 m, 3 m, and 6 m were investigated in 13

second grade boys who were identified by their classroom

teachers as having handwriting problems. Three subjects had

emotionally handicapping conditions, 4 subjects were

identified as at-risk for learning problems, 5 subjects were

not receiving special services, and 1 subject was in a

program for the gifted.

A criterion-based handwriting checklist, Error Analysis

of Children's Handwriting (EACH), was devised to assess

letter formation and word spacing. Stimuli for copying

consisted of 50-letter, nonword sentences. Stimulus

sentences were written in black ink on white paper the size

of regular writing paper and a chalkboard. Data from each

subject were graphically displayed and analyzed for level

(medians), variability (interquartile ranges), and trend

(slope) through visual inspection of boxplots and line


Investigation of level, variability, and trend of

letter formation and word spacing data did not identify 0.5

m, 3 m, or 6 m stimulus distances as being optimal for

letter formation or word spacing in any of the subjects.

Attentional, motor, visual, or unknown factors other than

stimulus distance may affect copying performance. Further

research is needed to determine if the clarity of the ink-

printed words on the white paper compared to ordinary chalk-

written words on a chalkboard may have enhanced handwriting

legibility for 3 m and 6 m distances. Improvements in

letter formation with 6 minutes of practice per day for 12

days for one subject has implications for the efficacy of

systematic handwriting instruction and the development of a

nonword vocabulary in the graphemic lexicon.




Legible handwriting is important to academic

achievement and to occupational role performance. Students

need to take notes, write answers to test questions, copy

information, and express ideas in writing. These abilities

translate into adult occupational behaviors such as writing

a check, making a grocery list, completing an application

form, and scribing a telephone message. Although an "age of

technology" exists in which computers have revolutionized

communication, basic handwriting skills are needed for

communication in situations where the use of a computer is

impractical (Furner, 1985).

Since handwriting is a basic skill that is required to

some degree for all occupations, it is not surprising that

handwriting has been studied in diverse fields of education,

rehabilitation, psychology, and neurology. In education,

teaching paradigms and handwriting proficiency and

legibility have been the focus of study (Peck, Askov, &

Fairchild, 1980). Through the use of applied behavior

analysis, special educators have sought to improve

handwriting rate and legibility in children with

handicapping conditions (Bijou, Birnbrauer, Kidder, & Tague,

1966; Hopman & Glynn, 1988; Kerr & Lambert, 1982; Sulzer-

Azaroff & Mayer, 19860. Retraining and compensatory efforts

in the rehabilitation of children and adults have been based

upon the development of handwriting ability (Benbow, 1988;

Erhardt, 1974; Gesell, 1940; Halverson, 1931; Knickerbocker,

1980; Olsen, 1980), motoric aspects of handwriting

(Rosenbloom & Horton, 1971; Wynn-Parry, 1966), and

prerequisite skills needed for handwriting (Newman, 1982;

Stephens & Pratt, 1981). Dynamic features of handwriting

production including the velocity, acceleration, and force

of writing movements have been precisely measured in

psychophysical studies (Kao et al., 1986b; Teulings &

Thomassen, 1983). Neuropsychological and neurophysiological

studies have identified brain areas active during

handwriting activities in impaired clients (Deecke, Heise,

Kornhuber, Lang, & Lang, 1984). Neuropsychological

explanations and models of the processes underlying written

communication have been derived from the study of

handwriting disorders in persons with cortical lesions

(Ellis & Young, 1988; Roeltgen, 1985).

Handwriting is a complex skill that involves language,

motor, and visual perceptual processes (Ellis & Young, 1988;

Roeltgen, 1985). Formal instruction for the development of

this skill usually begins in first grade (DeHaven, 1985;

Petty, Petty, & Becking, 1985). Most children make normal

progress in handwriting and go on to use writing for

communicative purposes. For some children, however, the act

of handwriting itself poses serious problems for academic


Two causes for handwriting problems have been suggested

by Graham and Miller (1980). Some handwriting problems are

due to central nervous system disorders such as cerebral

palsy (Bachmann & Law, 1961) or spina bifida (Hancock &

Alston, 1986) that interfere with handwriting and other fine

motor activities. However, the primary reason posed that

some children fail to develop legible handwriting is

inadequate instruction (Enstrom, 1966; Graham & Miller,


Copying tasks constitute a major form of handwriting

instruction. Children copy from a chalkboard or overhead

transparency and from workbooks and worksheets (Addy &

Wylie, 1973). Although a seemingly simple task, copying

requires the child to focus his or her eyes on the letter

form, visually analyze the components of the letter, recall

the relationship of those components, translate the visual

image into a motor plan of action, and monitor the movements

of the eyes and hand in the reproduction of the letter

(Kirk, 1980).

In an extensive review of the literature on handwriting

processes, instruction, and intervention, it was found that

investigators have not focused upon the effects of the

location of the material to be copied. Copying from a

chalkboard or from a source at the child's desk is a daily

activity throughout a student's schooling. Although copying

is a prevalent form of instruction, investigations of this

instructional task were few in the literature.

Statement of the Problem

The problem investigated in this study was the effect

of stimulus distance on handwriting legibility in children

with handwriting problems. The fundamental question related

to this problem is to what extent the placement of a

handwriting model proximally (near to) or distally (away

from) affects handwriting legibility in children who exhibit

handwriting difficulties. Specifically, this investigation

sought to address the effects, if any, upon 0.5-meter, 3-

meter, or 6-meter distances from the stimulus upon the rate

of correctly formed letters and the rate of correctly spaced

words in samples of second grade children's handwritten


The results of this study can provide school personnel

with information that may impact the use of copying tasks

routinely required on a daily basis in the classroom.

First, the study results provide data on the effect, if any,

on handwriting legibility as a result of the stimulus

distance from the child. Second, the study addressed the

question of stimulus distance upon individual aspects of

handwriting legibility: letter formation and word spacing.

Findings from this study can be useful to regular classroom

teachers, special education teachers, occupational

therapists, and other school personnel for curriculum and

therapy planning. Finally, the results of this study can be

useful to persons from any field who are interested in

handwriting research.

Purpose of the Study

The purpose of this study is to investigate the effects

of three distances of handwriting stimuli on handwriting

legibility of second grade children with handwriting

difficulties. Stimuli were placed either 0.5 m, 3 m, or 6 m

from the child. These three distances are roughly

equivalent to the distance of stimuli when copying from

materials placed at the desk or when copying from the

chalkboard while sitting in the front row of desks or in the

back row of desks in a regular elementary school classroom.

Handwriting legibility was judged for correct letter

formation and for correct word spacing.

Rationale of the Study

In this section the rationale for the study of the

effect of stimulus distance on children's handwriting

legibility is discussed. First, handwriting problems are

reviewed. Second, handwriting instruction is briefly

described. Third, a discussion of copying behavior includes

the instructional use of copying tasks and analysis of the

visual, attentional, and motor subskills needed for copying

and speculations about possible impact of stimulus distance

on handwriting performance.

Handwriting Problems

Most children have the visual acuity, oculomotor

control, eye-hand coordination, motor planning ability, and

attentional capacity to engage in copying and other

instructional tasks and to develop adequate handwriting.

However, for children who may not have had the readiness

activities or who have not benefitted from them as much as

their classmates, handwriting skills may not readily

develop. Similarly, for the child whose central nervous

system maturity is at the lower end of the normal

distribution, highly complex activities such as handwriting

may be difficult. Some children lack maturity in their

ability to attend to classroom tasks and to stay focused

upon relevant material. For children with learning,

emotional, attentional, or motor deficits, handwriting may

be one of many problems encountered in school. For children

with any of these problems, handwriting tasks may challenge

the limits of their abilities.

Problems with handwriting can impede academic progress

and obscure a student's talents and competencies.

Handwriting difficulties are common in children with

learning disabilities (Belka & Williams, 1979) and have been

identified in gifted children (Salend, 1984). An estimated

one child per classroom has severe handwriting problems

(Cavey, 1987). Problems with handwriting and fine motor

skills constitute one of the most common reasons that

children are referred to occupational therapists who work in

the school system (E. Vizvary, personal communication,

October, 1988). Poor handwriting has been cited as the

culprit in spelling errors (Johnson, 1984; Strickling, 1973)

and in the failure of students to meet minimum standards on

ninth-grade aptitude tests of writing proficiency

(Kilpatrick, 1983). Pupils with poor handwriting are likely

to get lower grades although the content of their work is

comparable to that of pupils with good handwriting (Briggs,

1970, 1980; Rondinella, 1963; Markham, 1976; Soloff, 1973).

Children identified by teachers as at-risk for developing

academic problems were found to have significantly lower

scores in efficient handwriting than normal children

(Goodgold, 1983).

Handwriting Instruction

Factors for optimal handwriting instruction are

important for all children, but especially for those

children having handwriting problems or who are at-risk for

developing these problems. Graham (1980) suggested that

inadequate instruction accounts for the majority of

handwriting problems, except for those problems resulting

from a neuromotor deficit. Several things may account for

the inadequacies in handwriting instruction. Teachers may

have been inadequately trained to teach handwriting (Groff,

1962) and may have preferred to teach other subjects

(Greenblatt, 1962). In addition, certification agencies may

not have required the teachers to have handwriting

preparation (King, 1961) and school district curriculums may

not have recommended a formal handwriting program for their

pupils (Addy & Wylie, 1973; King, 1961; Wolfson, 1962).

Handwriting instruction has been conducted typically for the

entire class without individualized instruction (Addy &

Wylie, 1973). The lack of systematic handwriting

evaluations have made it difficult for teachers to identify

specific problem areas in handwriting (Graham, 1982). For

children referred to special education, emphasis upon

mathematics and reading deficits have had priority over

handwriting difficulties. These factors have posed

potential instructional inadequacies for the child who has

had or is at risk for developing handwriting problems.

Analysis of Copying

Every school day, students copy information presented

on the chalkboard or at their desks. In elementary school,

teachers use copying tasks daily to teach handwriting to

first and second grade children. Copying is a complex task

requiring visual, attentional, and motor subskills (Petty et

al., 1985). These subskills needed for handwriting are



Visual subskills. Copying from the chalkboard involves

complex oculomotor and optical processes. The child needs

to have adequate visual acuity so that the material to be

copied is seen clearly. Focusing for near vision (near

triad) and far vision (stereoscopic depth perception) are

involved in the shift of the eyes from the stimulus to the

writing paper. To identify the specific letters or words to

be copied, the eyes must selectively locate the stimulus to

attend to (saccades) and maintain focus upon that stimulus

(fixation). Coordinated movements of the eyes (conjugate

eye movements) and coordinated movements of the eyes and

head (compensatory eye movements) are involved as the eyes

and head move to look from stimulus to writing paper.

To be ready for handwriting instruction, the child

entering first grade should possess, or be able to quickly

develop, a number of readiness skills in visual perception.

Children need to be able to differentiate between right and

left and sizes and shapes, understand left-to-right

progression across a page, and discriminate parts from the

whole and the spatial relationships among those parts

(DeHaven, 1983; Petty et al., 1985; Lindsey & Beck, 1984).

Attentional subskills. During the process of learning

to write, attention is focused upon the act of writing

itself and movements are deliberate and laborious (DeHaven,

1983). For successful copying, the child must visually and

selectively attend to each letter to be copied, refrain from

being distracted as that letter or word is retained in

visual memory and translated to a motor pattern, and,

finally, the child must focus on the motoric reproduction of

the letter on the paper. As motor movements become

established and automatic, attention can be directed to

refinements of the letter forms and the communicative

aspects of handwriting.

Motor subskills. Motor subskills include the ability

to maintain head and body postural control necessary for the

writing position and tool prehension. Eye-hand coordination

must be sufficient for the child to control the movements of

the writing tool. Planning of motor movements is required

to direct the writing tool in letter formation. Visual-

spatial perception must be coordinated with the organization

of movements to produce the desired graphic letter

representation (Gilfoyle & Hays, 1980).

In summary, copying is a highly complex task, demanding

visual acuity and perception, attention to the task, and the

ability to organize sequences of motor movements in response

to visual stimuli. Intact visual, motor, and attentional

abilities are needed for handwriting development.

Copying tasks. Copying tasks are commonly used in

handwriting instruction, yet there is a paucity of

literature to suggest optimum uses of copying tasks.

Copying has been shown to be superior to faded tracing for

learning letter formation (Hirsch & Niedermeyer, 1973) and

for learning to write shorthand symbols (Askov & Gref,

1975). Improved copying as a result of combined verbal

instruction and demonstration (Sovik, 1976) and as a result

of motion models versus still models (Sovik, 1979; Wright &

Wright, cited in Peck & Askov, 1980) have been shown.

Geometric form copying in response to a proximal model was

found to be superior to copying from a remote model in a

study of 7 and 10 year old children (Sovik, 1979). However,

the effects of distance on handwriting performance when

copying from material placed at the desk or on a chalkboard

have not been investigated. For the child who has

handwriting problems or who is at risk for developing these

problems, stimulus distance may affect visual, attentional,

and motor aspects of handwriting.

Distance of stimulus. The distance of the stimulus is

posed by this investigator as a possible factor in copying

performance. As the distance from the stimulus increases,

the visual, attentional, and motor demands are thought to

also increase. For the child at risk for developing

handwriting problems, performance could deteriorate at any

point in the copying process when the stimulus is distant

from the child.

For visual performance, the farther the stimulus is

from the child, the longer the interval between seeing the

stimulus and writing it. The child may have trouble

retaining the visual image of the model as he or she shifts


his or her gaze from the chalkboard and paper. The farther

the stimulus, the greater the demands on stereoscopic depth

perception and the greater the amount of movements and

adjustments of the eyes as the eyes adjust from chalkboard

to paper and back. It may be taxing for the at-risk child

to alternately focus their eyes between stimuli on the desk

and chalkboard.

For attention, the farther the stimulus is from the

child, the greater the amount and variety of stimuli that

the child must screen out in order to attend to the relevant

stimulus. It may be difficult for a child with attentional

problems to screen out the classroom stimuli that compete

with focusing upon a letter on the chalkboard.

For motor performance, with distant stimuli, the child

must retain the visual image, make postural adjustments of

the eyes and head, and organize the motor movement. The

child may become disorganized when composing the movement to

match the visual image or when comparing the letter he or

she has just copied to the model. It is possible that

bringing the stimulus closer to the child could reduce the

visual, attentional, and motor demands of the task and make

it easier for the at-risk child to copy a handwriting model.

Variables in copying behavior include visual,

attentional, and motor variables within the child and

teacher, setting, and materials variables within handwriting

instruction. It is possible that the distance from the


stimulus has little effect on copying performance and other

factors need to be considered. The results of this study

can offer information to rule out stimulus distance as a

factor to consider for optimal handwriting instruction.

In summary, the rationale for the study has been

described. Handwriting problems in children were briefly

described. Second, handwriting instruction issues were

discussed. Finally, handwriting copying tasks were

discussed including an analysis of the visual, attentional,

and motor aspects of copying, copying tasks, and distance

from the stimulus. Definitions used in this investigation

are listed next.

Definition of Terms

Allograph refers to the various forms for representing

individual letters, i.e., upper case, italic, lower case,


Auditory-language process refers to the analysis of

auditory information for reading, speaking, and composing

information for writing.

Auditory-language-motor process refers to the analysis

of auditory information for language functions expressed

through motor acts such as handwriting.

Brain stem is below the cerebral hemispheres and

subserves motor, sensory, autonomic, and attentional

functions that are reflexive and not voluntary.

Carpal refers to the wrist.


Ciliary muscle attaches to the lens of the eye and when

contracting causes a release of lens tension and rounding of

the lens for near vision.

Cocontraction refers to simultaneous contraction of two

or more muscles.

Cranial nerves are the peripheral nerves of the brain

stem that innervate facial and eye muscles.

Cursive writing refers to flowing script where the

letters are joined together.

Distal means away from the center of the body.

Dvsaraphia refers to an impairment in the ability to

write due to a pathological disorder.

Forward flexion refers to movements in the front-to-

back plane such as bending the head or trunk down to look at

or touch the toes.

Grapheme is the written or printed visible

representation of a letter.

Graphomotor is used synonymously with handwriting and

refers to the motion pertaining to writing.

Head rotation refers to the side-to-side movement of

the head as in responding, "no."

Kinesthetic refers to the sensation of movement,

weight, resistance, and position of the body or parts of the


Lateral rectus muscle refers to the muscle on the

outside of the eyeball that causes the eye to turn to the


Lateral tilt refers to the side-to-side movements of

the head as in the movement of the head to respond "maybe."

Tilting is also seen in the shoulder, pelvis, or trunk.

Lexicon refers to a conceptualized storehouse of

elements of language: phonemic (individual sounds),

graphemic (individual letters), grammar, and syntax.

Manuscript writing is synonymous with printing.

Medial rectus muscle refers to the muscle attached to

the nose-side of the eyeball and causes the eye to turn in

towards the nose.

Metacarpals are the hand bones between the wrist and


Motor programming refers to the psychoneural activity

involved in planning and execution of movements.

Postural adjustments refer to automatic movements to

bring the body parts into alignment for an advantageous

position to orient to a task.

Premotor cortex is the section of the cerebral cortex

that is important for sensory-guided voluntary movements

such as writing.

Primary motor cortex is the section of cerebral cortex

that is responsible for finely graded movements of the

fingers and variations in force required to manipulate

objects such as writing tools.

Proprioception refers to the sensations that arise from

within the body regarding spatial position and muscular

activity or to the sensory receptors that they activate.

Proprioception is contrasted to stimuli arising from the

environment or the internal organs.

Proximal means towards the center of the body.

Pupillary sphincter is the concentric muscle that

causes the pupil to constrict.

Scapulohumeral refers to the shoulderblade (scapula)

and upper arm bone (humerus).

Supplementary motor area is the section of the cerebral

cortex that has a role in programming of movements such as

movements required in handwriting.

Tripod pencil grasp is grasp of the pencil with the

index and middle fingers and the thumb.

Vestibular refers to the sensory system that perceives

movement of the head and position of the head in

relationship to gravity.

Visual language motor process refers to the analysis of

visual information for language functions expressed through

motor acts such as handwriting.

Delimitations of the Study

This study was delimited by geographical region,

subject age and disability, and handwriting variables. The

geographical region was restricted to Gainesville, Florida,

population 84,200 (Gainesville Chamber of Commerce, personal

communication, June 7, 1989) located in Alachua County in

north-central Florida. Subjects for the study were 7 and 8

years of age who had completed at least 1 year of manuscript

writing instruction prior to the study and who had

handwriting difficulties. Excluded from the study were

subjects who had neuromotor problems such as cerebral palsy,

sensory problems such as blindness or deafness, language

disorders such as aphasia, or emotional disorders such as

autism. Subjects attended Terwilliger Elementary School and

were in regular second grade classes and some also received

services in the varying exceptionalities resource room or

the program for gifted students (Alachua School District,

1989) or the Chapter I classroom (School Board of Alachua

County, 1989). All subjects were male and were right handed

for writing. Subjects represented middle and low

socioeconomic backgrounds and included minority and

nonminority races. The independent variables consisted of

three distances of a handwriting copying model from the

subject. The sentences to be copied consisted of nonwords.

The legibility of handwriting was assessed for letter

formation and word spacing and not for letter slant or

handwriting speed. Stimuli to be copied were placed on an

upright surface directly in front of the child. Copying

from a workbook, from stimuli placed on the desk or to one


side of the subject was not addressed in this study. Other

aspects of copying that were not considered included

content, composition, writing to dictation, and spontaneous

writing. Visual and attentional aspects of handwriting were

addressed, particularly as they related to the motor aspects

of handwriting. Visual perceptual processes and language

issues related to handwriting were not part of this


Limitations of the Study

This study included second grade children identified by

their teachers as having difficulty with handwriting or

second grade children who were in the regular classroom and

received services in a varying exceptionalities resource

room or in a Chapter I program and therefore the findings

should not be considered for other handicapped or for

nonhandicapped individuals. In addition, the results cannot

be applied to other areas of handwriting. The conditions

were administered for children in a regular second grade

classroom and therefore may not be appropriate for other

grades without replication of the study.


In summary, the rationale for this investigation stems

from several observations. Copying from the chalkboard or

from material at the desk is an everyday occurrence in

handwriting instruction. Copying from the chalkboard is a

complex visual, attentional, and motor task. Some children

develop or are at risk for developing handwriting problems

and this may be due primarily to handwriting instructional

practices. Optimizing copying tasks could enhance

handwriting instruction, especially for those children at

risk for developing handwriting problems. There is a

paucity of information about optimal factors for copying

tasks. The distance of the stimulus from the child is posed

as one factor that could affect handwriting performance.

In Chapter II literature relevant to this study is

reviewed, and in Chapter III the methods for conducting the

study are delineated. Chapter IV contains the results of

the study and Chapter V includes the implications and




In Chapter II, a summary and an analysis of the

professional literature related to neuropsychological

processes for handwriting and to handwriting assessment and

instruction are presented. The chapter is divided into four

major sections. The first section includes selection

criteria for the literature that was reviewed. Section two

includes neuropsychological and neurological literature

related to motor, visual, and attentional processes for

handwriting. In sections two and three educational

literature regarding handwriting evaluation and handwriting

instruction is reviewed.

Selection of Relevant Literature

An initial step in the review of the literature was

that of determining the criteria for the inclusion of the

references. All relevant studies completed in the last 10

years (1979-1989) were examined. In addition, any notable

researched cited in the literature earlier than the 1979-

1989 time period was also considered. Professional

literature concerning neuropsychological processes for

handwriting and handwriting evaluation and instruction was

examined by using the following criteria for inclusion in

the review:

1. The subjects and the settings in which the

experimentation took place had to be thoroughly


2. The treatment conditions and experimental

procedures were detailed enough to permit


3. The experimental design and data analysis

procedures were presented without significant

losses of information.

4. The interpretations of the experimenter had to be

consistent with the results displayed.

In order to conduct a comprehensive review of the

literature relating to handwriting, the following sources of

literature review were consulted: Dissertation Abstracts

International, Educational Resources Information

Clearinghouse (ERIC), Psychological Abstracts, Current Index

to Journals in Education (CIJE), and Index Medicus.

Literature on the neuromotor aspects of handwriting was

reviewed in journals including Annals of New York Academy of

Sciences, Annual Review of Neuroscience, Brain, Brain

Research, Experimental Brain Research, Journal of

Neurophvsiologv, and Human Neurobiology. Psychophysical

analyses of handwriting were reviewed in journals including

Acta Psychologica, Biological Cybernetics, Cybernetics,

Ergonomics, and Visible Language. Handwriting grip and

posture were reviewed in the American Journal of

Occupational Therapy, Developmental Medicine and Child

Neurology, Educational Review, and Archives of Physical

Medicine and Rehabilitation.

Literature on handwriting assessment was reviewed in

the American Educational Research Journal, Educational

Research, Educational Review, Journal of Educational

Research, and the Journal of School Psychology. Handwriting

assessment manuals were examined for validity and

reliability test development data.

Handwriting instruction was reviewed in educational

journals. Since single subject research design was used in

the study, the literature review of handwriting intervention

included studies in which single subject design was used.

Journal sources for these studies included Educational

Psychology, Education and Treatment of Children, Elementary

English, Journal of Applied Behavior Analysis, Journal of

Learning Disabilities, Learning Disability Quarterly,

Progress in Behavior Modification, and Pointer.

References initially selected were located through the

library at the University of Florida, through the

interlibrary loan system, or through other professionals in

the field. Descriptors used in this literature search

included handwriting, penmanship, writing, dysgraphia,

agraphia, written expression, and combinations of these

terms with evaluation, instruction, intervention, learning

disabilities, rehabilitation, research, and processes.

The references that were selected were critically

reviewed and those that described empirical investigations

were chosen based on the investigator's judgment that the

references presented a clear description of subject

selection, methodology, and results. Professional

literature other than empirical investigations were also

included if, in the author's judgment, the information that

was included provided a valuable contribution to the

knowledge base about or an understanding of the motor

aspects of handwriting and handwriting evaluation and


Handwriting as a Neuropsychological Process

The refined use of distal musculature for tool use and

speaking is distinctly human. One form of tool use,

handwriting, is a visible form of language characterized by

the ability to use brushes, crayons, sticks, pencils, pens,

markers, and other writing tools to form graphic

representations of thoughts and concepts. Such fine visual-

motor skill is not explicable in terms of the

musculoskeletal system since humans and old world monkeys

have very similar hands (Rothwell, 1987). In addition, the

organism must be oriented to and attentive during

handwriting tasks. The visual and language abilities to

understand and use graphic symbols to represent concepts and

ideas are expressed through handwriting. Neural control

systems account for the refined movements in handwriting and

for associated attentional, visual, and language processes.

In this section, handwriting is reviewed in five parts.

In part one, a neuropsychological model for the handwriting

process is described. In part two, motor aspects of

handwriting will include three subsections: (a) cortical

substrates of graphomotor skills identified from literature

on human and monkey studies of motor cortices, (b)

psychophysical analysis of handwriting features, and (c) the

ergonomics of grip and posture. In part three, visual

aspects of handwriting will be analyzed. In part four,

attentional mechanisms important for graphomotor functions

will be reviewed. Finally, in part five, an observational

analysis of handwriting posture in normal first grade

children will be summarized. While the literature reviewed

does not usually address handwriting in children, it is

relevant for considering neuropsychological processes

involved in handwriting. In addition, lesion studies

related to handwriting have been conducted with adults and

this provides background information for considering

neuropsychological processes required in children's


Neuropsvchological Model of Handwriting

In neuropsychological models, various "systems" or

"components" are posed that underlie language functions

(Ellis & Young, 1988; Margolin, 1984; Roeltgen, 1985). Such

models have been used to explain the ability or inability to

speak, spell, or write regular, irregular, and nonwords in

persons suffering from aphasia, alexia, or agraphia. A

neuropsychological model (Figure 1) for graphomotor ability

(handwriting) is proposed by the investigator and includes

visual, auditory, language, and motor systems. The model is

adapted from Roeltgen's (1985, p. 80) model for writing and

oral spelling and Ellis and Young's (1988, pp. 175, 181)

functional model for spelling. With the explanation of the

model, reference will be made to copying as a graphomotor


In this model, the visual system receives and analyzes

individual elements of the visual stimulus to be copied.

Spelling of the word can occur through two processes: the

graphemic lexicon or the phonological lexicon and phoneme-

grapheme conversion. The graphemic lexicon stores visual

word images and spellings of words used in writing. This

system uses a "whole word retrieval process" (Roeltgen,

1985, p. 81). When a word is to be written, this lexicon

releases a series of graphemes in a prescribed sequence and

this sequence of graphemes can be expressed in writing,

typing, or oral spelling (Ellis & Young, 1988). This system



0 1-

I L-
-i4 E O
O- 5 0

a. o
S0 o o

F- -'-
o pmom m

xO C.) OO 20
0 0R H
I\ 0

\4 O

cw O

m z
R1 0.- I


E cc c


is useful for spelling irregular words that cannot be

spelled by matching the sound directly to the letter such as

"comb," "knife," and "phone." Through the learning process,

this system also stores regular words, such as "animal"

where the sound and the letters in the word have one-to-one

correspondence. Persons with lesions in the dominant

parietal lobe may show deficits in the graphemic lexicon and

have difficulty spelling irregular words but can spell

regular words due to the preserved abilities of the

phonological system and phoneme-grapheme conversion.

In the phoneme-grapheme conversion system, words are

broken down into individual phonemes (sounds) and converted

to letters, the visual representations of the sounds. Thus,

words that are heard can be converted into graphemes for

writing. In the process of learning to spell words,

children use the phoneme-grapheme conversion system to

assemble words from individual sounds, or phonemes. As the

spelling vocabulary increases, spellings are increasingly

stored in the graphemic lexicon, bypassing the phoneme-

grapheme conversion system (Ellis & Young, 1988). The

phoneme-grapheme system remains useful for spelling

pronounceable nonwords such as "gebit" or for devising the

spelling for unfamiliar regular words. With a lesion to the

dominant parietal lobe, persons may experience deficits in

the phonological system and phoneme-grapheme conversion.

They have difficulty spelling nonwords but are able to spell

familiar regular and irregular words due to the storehouse

of learned words in the graphemic lexicon (Roeltgen, 1985).

For copying, spellings of the words in the stimulus to

be copied can be retrieved from the graphemic lexicon or can

be devised through the phonological lexicon and phoneme-

grapheme conversion. The semantic system provides meaning

to the words being copied (Ellis & Young, 1988).

Output of the written word occurs through the grapheme

system (Ellis & Young, 1988). The grapheme system is

thought to subserve motor movements for writing, whether it

be handwriting, typing, assembling plastic letters, or

operating a Morse code instrument. A grapheme is a

conceptual form of a letter expressed in writing or speaking

(Ellis, 1979, 1982). Presumably, visuokinetic motor engrams

for graphemes are stored, retrieved, and combined to

formulate writing in the grapheme system. Patients with

apraxic agraphia can spell orally but exhibit illegible

writing response to dictated material. Thus, they know how

to select letters in spelling a word but are unable to write

a word due to damage to their visuokinetic motor engrams.

Their writing improves with copying due to intact visuomotor

abilities. Neurological lesions in patients with graphic

apraxia involve the dominant parietal lobe (Roeltgen, 1985).

The graphemic output programming system (Roeltgen,

1985) corresponds to Ellis' (1979, 1982; Ellis & Young,

1988) allograph level where a particular grapheme (written

letter) is represented by several allographs: upper case,

lower case, manuscript, cursive, and so on. For example,

allographs for the grapheme "a" include "a," "A," "a," and

"A." The selection of the desired letter shape occurs at

this level.

Programming for graphemic output is influenced by

nonverbal, visual-spatial orientation abilities (Roeltgen,

1985). Each letter is written as a series of circular,

horizontal, vertical, and diagonal strokes. Individual

strokes are thought to comprise the fundamental unit of

letter formation and each letter is programmed as a sequence

and spatial relationship of one or more strokes (Margolin &

Wing, 1983). Patients with spatial agraphia who make errors

in single strokes are unable to write on a straight

horizontal line, leave blank spaces between graphemes, have

problems copying, and tend to repeat the same strokes.

Abilities to spell orally and to pronounce words that are

spelled indicate intact language areas. The nondominant

parietal lobe has been identified as the lesion site in

patients with spatial agraphia (Roeltgen, 1985).

Finally, "motor programming" (Roeltgen, 1985) or the

"graphic motor patterns" (Ellis & Young, 1988) involve the

selection of muscles, timing of movements, and neuromuscular

coordination for graphic production. The sequence of the

strokes required to create letter forms is formulated here.

Such programming is thought to occur in area 4, the primary

motor cortex (Roeltgen, 1985).

Parallel to the visual-language-motor process just

described for seeing and then writing a word is the

auditory-language-motor process for hearing and then writing

a word. In the auditory-language-motor process, the

auditory system analyzes the acoustic elements of the

sounds. The phonological lexicon stores auditory word

engrams and organizes the elements for speaking via the

utterance of phonemes to say a word. The semantic system

lends meaning to the sounds. These auditory processes can

influence graphemic processes through the phonological

lexicon and phoneme-grapheme conversion acting on the

grapheme system. Through this auditory-language-motor

process, dictated words can be spelled and written.

In summary, handwriting is a complex task, requiring

combined motoric, visual-spatial, and linguistic efforts. A

model of handwriting is useful as a framework for

considering the role of various systems in graphomotor

performance, especially in clinical populations (Margolin,

1984). As identified in the model, visual, auditory,

language, and motor processes culminate in visible

expression through the motor act of handwriting.

Motor Aspects of Graphomotor Skills

Motor aspects of graphomotor functions are discussed in

four areas. First, cortical substrates for graphomotor

skills are reviewed including primary motor cortex,

supplementary motor cortex, and premotor cortex. Second,

psychophysical studies of features of handwriting are

summarized. Third, ergonomic factors including grip and

posture are described. Finally, the relationship of these

discussions to graphomotor skills is summarized.

Cortical substrates

Findings from animal and human studies of motor cortex

functions provide further clarification of the contribution

of motor cortices to handwriting and provide a background

for study of motor substrates of handwriting in humans.

These studies involve direct measurement of cellular

activity in the cortex. It is not possible to link findings

from cellular research to specific classroom writing

activities. However, review of cellular-based research

illuminates the complexity of the human skill of

handwriting, as considered from a neurocellular level.

Six motor areas have been identified in humans

including the primary motor area (MI or area 4), MII at the

bottom end of area 4, the supplementary motor area (SMA),

the frontal eye fields (area 8), the premotor area (PM or

area 6), and Broca's area (frontal operculum) (Rothwell,

1987). Three areas are considered in terms of graphomotor

skills: MI, SMA, and PM.

Primary motor cortex (MI). Five functions of the

primary motor cortex are briefly discussed: fine finger

movements, force of muscle contraction, sensory control of

movement, execution of movement, and purposeful reaching.

In addition, the primary motor cortex is preferentially

involved in control of finely graded movements rather than

gross limb movements (Kuypers, 1982).

Primary motor cortex neurons synapse directly on motor

neurons in the spinal cord and such direct connections are

believed to subserve highly fractionated movements of the

fingers (Kuypers, 1982; Muir & Lemon, 1983). Direct spinal

motor neuron connections are numerous in monkeys, chimps,

and humans and are not seem in marsupials and carnivores

with limited fine motor control. Primary motor cortex

neurons to the spinal cord develop early during postnatal

life and this development coincides with increasing fine

motor control of the hand (Kuypers, 1982). Primary motor

neurons also determine fine gradations of the force of

muscle contraction needed for a given load (Evarts, 1981).

In a human with a discrete lesion of the hand region of MI,

Freund (1987) observed decreased dexterity and force in the

hand and fingers.

Terminations of MI in the spinal cord subserve sensory

control (Wiesendanger, 1981). Tight coupling of sensory

input and motor output between the brain and spinal cord was

found in intracortical microstimulation studies of the

individual cells in MI. Primary motor cortex cells that,

when stimulated, resulted in movement of a joint also

received input from the muscles and joint receptors of that

same joint (Rosen & Asanuma, 1972).

The primary motor cortex has been described as the

"executive locus for voluntary movements" (Roland, 1985, p.

155). In metabolic mapping studies in humans, MI was

consistently active during voluntary movements and was not

active when voluntary movements were not being performed

(Roland & Friberg, 1985).

In addition to fine finger movements and voluntary

motor control, MI is involved in planning and executing

purposeful reaching. Georgopoulos (1986, 1987) recorded

from single cells in MI and found that 80% of the MI cells

that were active in reaching were active when reaching

occurred in a "preferred" direction.

Supplementary motor area (SMA). The supplementary

motor area is located on the medial surface of area 6,

premotor cortex. The SMA exerts its influence bilaterally

and is thought to have a role in the programming of

movements (Roland, Skinhoj, Larsen, & Endo, 1977; Rothwell,

1987). Five functions include initiation of movements,

planning for movements, visually guided movements, visual

motor learning, and anticipatory postural preparation for


The role of SMA in the initiation of voluntary movement

has been identified in human studies of the

bereitschaftspotential (BP) readiness potential averaged

from electroencephalograms and compared to electro-myograms

(Deecke, 1987). Widespread cortical activity is seen prior

to movement and is initially bilateral and then becomes

unilateral. This bilateral-to-unilateral shift is thought

to reflect SMA, then MI activity (Deecke, Weinberg, &

Brickett, 1982; Roland, 1987). For example, BPs in the SMA

were recorded 1.2 seconds prior to the onset of EMG activity

in bilateral flexion of the fingers; BPs in MI were seen 0.7

seconds prior to the onset of EMG activity (Kristeva &

Deecke, 1980).

The role of the SMA in planning movements has also been

shown in metabolic mapping studies of humans (Roland, 1987).

During the performance of complex actions such as finger-to-

thumb sequential movements, the SMA was metabolically active

(Roland, 1985).

The SMA has also been found to be active in the early

preparation phases for visually guided activities (Deecke,

Heise, Kornhuber, Lang, & Lang, 1984). Human subjects drew

with a pen the shape of a visually presented stimulus.

Subject's eyes were fixated centrally and the stimulus was

presented in the left visual field. The subject wrote with

the right hand. Before the subject pressed the pen to the

paper to initiate the stimulus, BPs were seen in occipital

and frontal leads. The SMA BPs stopped 90 milliseconds

prior to the pressing down of the pen to initiate the task.

In contrast, the occipital BPs stopped 200 milliseconds

after the pen was pressed and moved into the correct

direction. The authors interpreted this as the SMA having a

preparatory, motivational role in initiating the task. The

role of the occipital area was to attend to relevant visual

stimuli needed to carry out the tracking task.

The possible role of SMA in visual motor learning was

shown in a task requiring the subject to visually fixate on

a moving target and to track it by writing with a pen (Lang,

Lang, Kornhuber, Deecke, & Kornhuber, 1983). In addition to

the regular tracking task, a learning task involved inverted

mirror tracking. When activity of the SMA and frontolateral

convexity during the learning task was compared to activity

of these cortical areas during the regular tracking task,

"good" learners showed significant increases in cortical

activity and "poor" learners showed the same or less

cortical activity on the two tasks. Thus, the frontomedial

SMA region and the frontolateral convexity appear to be

important for visual motor learning.

The role of SMA in anticipatory postural preparation

for movements was demonstrated by comparisons of SMA

cortical cell activity and EMG recordings in a monkey

performing a choice-reaction task (Wiesendanger et al.,

1987). Increased SMA neuron activity was observed between

the "ready" and "go" signals and this activity corresponded

with EMG activity. Thus, in anticipation of the signal to

move, the SMA presumably had a role either directly or via

MI in tensing of muscles to prepare for movement.

Humans with SMA lesions involving the cingulate cortex

had reduced speech, facial expressiveness and spontaneous

motor activity (Freund, 1987). Patients had difficulty with

bimanual coordination and this persisted after the recovery

period whereas unilateral limb movements recovered. Thus,

lesions of the SMA resulted in problems with bilateral

movement coordination and initiation of movements.

Premotor cortex. The premotor cortex comprises area 6

that lies rostral to MI and on the lateral surface of the

frontal lobe. It is continuous with the supplementary motor

area that is on the medial surface. In humans, the premotor

area is about six times larger than MI and many premotor

neurons project to the primary motor area (Rothwell, 1987).

Three functions will be described: sensory guided

movements, purposive responding, and proximal muscle


The premotor area is thought to be important for

sensory-guided voluntary movements, that is, movements

occurring through interaction with the environment versus

movements that are centrally programmed such as crawling or

walking. The premotor cortex gets abundant visual input and

is thought to be especially involved in visually-guided

movements (Rothwell, 1987). In studies with humans, Roland

and colleagues (Roland, 1987; Roland, Larsen, Lassen, &

Skinhoj, 1980a, 1980b) found that PM was metabolically

active during sensory-guided voluntary movements. Neurons

in the premotor cortex also responded well to tactile

stimuli (Rizzolatti, 1987).

Weinrich, Wise, and Mauritz (1984) identified "set-

related neurons" that were active between the onset of a

stimulus and the signal to move. Cells were directionally

specific and were active if the stimulus was moved in a

certain direction in space.

The role of PM in purposive movements was suggested by

single cell recordings that showed that PM cells were active

during distal arm movements. Cells were more active when

the arm movement was directed toward some specific aim and

were not as active when the arm was simply moved using the

same musculature. The author proposed that the premotor

area contains a "vocabulary of motor acts such as reaching,

grasping, and holding and that this vocabulary can be

addressed by visual and somatosensory stimuli" (Rizzolatti,

1987, p. 180).

The premotor cortex appears to influence proximal as

well as distal movements. Lesions of the premotor cortex in

humans result in poor bilateral coordination for movements

such as pedalling and "windmill" arm motions requiring

coordination of proximal musculature (Freund & Hummelsheim,

1985). Distal movements of the two limbs together such as

shoe tying and clothes buttoning were unimpaired. However,


in patients with lesions that were laterally placed in area

6 just anterior to the hand area of the primary motor

cortex, disturbances in distal musculature were seen

(Freund, 1987).

Relationship to graphomotor skills. From the foregoing

discussion it is reasonable to assume that graphomotor

performance involves MI, SMA, and PM. The primary motor

area is involved in the fractionation of finger movements

and the variations in force required to manipulate a writing

tool. As the executive of motor commands, MI is the final

link in cortical chains of commands that involve

supplementary and premotor areas. All three areas, MI, SMA,

and PM are involved in sensory guided movements.

Handwriting requires the guiding of movements via

visualization of the movement needed to form the grapheme,

via visual feedback from the written page and via tactile,

proprioceptive and kinesthetic sensations arising from and

requisite for the movements as they are produced. All three

areas are involved in the purposive nature of handwriting.

One writes with the desire and intent to express one's

thoughts. Unlike other refined skills, handwriting

engenders language in addition to motor and visual systems.

Finally, the anticipatory postural preparation for movement

provides the writer with body orientation to the task. The

writer automatically assumes a writing posture and the

stability of the trunk, neck, and shoulder provide adequate

support for refined distal movement. In short, MI, SMA, and

PM motor cortices are involved in graphomotor performance.

Possible cortical substrates of handwriting have been

discussed briefly. Next, the study of psychophysical

features of handwriting is reviewed.

PsvchoDhvsical features

Through computerized analysis of handwriting

performance, several features of handwriting have been

observed (Hollerbach, 1981; Hulstijn & van Galen, 1983; Kao

et al., 1986a; Teulings, Thomassen, Keuss, van Galen, &

Grootveld, 1983; van Galen & Teulings, 1983). A handwriting

signal was generated through the use of an electronic pen on

ordinary paper that covers an xy (horizontal-vertical)

digitizer. Pen tip position registered on the xy

coordinates from a fixed point and was sampled across time

with a frequency of 200 per second and an accuracy of 0.2 mm

(Teulings & Thomassen, 1983). Velocity, acceleration, and

form features of handwriting (Teulings & Thomassen, 1983;

van Galen & Teulings, 1983), and reaction time and movement

time were monitored (Hulstije & van Galen, 1983).

Velocity was measured horizontally, vertically, or in

combination (absolute velocity) to give an indication of

speed of responding. Normally, in smooth writing, the

absolute speed within a word was seldom close to zero for

more than 30 milliseconds. The time needed to draw a

certain pattern seemed relatively independent of the total

length of the pattern that was proportional to its size.

For example, a 16-fold increase in writing size resulted in

an increase in time by 1.6 (Teulings & Thomassen, 1983).

Acceleration was measured in terms of change in x and y

velocity of the pen per unit of time. The acceleration

indicated that a force was applied to the pen. The net

force equaled the sum of the muscle force and friction

forces of the pen tip and hand. Acceleration was considered

as the net force pattern produced by the muscle since force

exceeded friction. The acceleration pattern looked

repetitive although writing contained different letters.

This was because various letters were produced by only tiny

variations in duration and amplitude of individual xy phases

(Teulings & Thomassen, 1983).

Models of handwriting were generated through research

on feature analysis of handwriting. In one model, a three-

stage process was proposed including (a) retrieval of an

abstract motor program from long-term memory; (b) parameter

setting of size, accuracy, and speed; and (c) translation of

the program and parameters into a neuromuscular response

(van Galen & Teulings, 1983). Neurological substrates for

these processes were not proposed.

Ergonomic features of grip and posture

Therapists, physicians, educators, and others concerned

with early childhood development and with the development of

hand skills have studied the motoric aspects of handwriting.

Halverson (1931) and Gesell (1940) used cinema analysis to

study the developmental progression of object manipulation

in normal children.

Focusing upon finger grasp, Wynn-Parry (1966) labeled

the "dynamic tripod" grasp of writing tools that included

the thumb, index, and middle fingers functioning together to

perform smooth writing movements. This tripod posture was

further delineated by Rosenbloom and Horton (1971) into two

stages. First, the static tripod posture was used but the

fingers remained relatively silent while the wrist moved the

hand to write. In the second stage, the dynamic tripod

developed in which the fingers moved from the knuckles to

produce coordinated pencil movements. Subsequently, Erhardt

(1974) identified a developmental progression of grasp for

handwriting: (a) fisted grasp at age 1; (b) grasp with all

the fingers at age 2 to 3 years; (c) stiff-fingered, static

tripod grasp at 3 1/2 to 4 years; and (d) mature, dynamic,

tripod posture at 4 1/2 to 6 years.

In a study of normal adults (Kamakura, Matsuo, Ishii,

Mitsuboshi, & Miura, 1980), subjects grasped inked objects

and the hand was then photographed to show the areas of the

hand involved in static prehension of 98 objects. Subjects

used the tripod grip to grasp a pencil using

[a] the radial aspect approximately at the level
of the distal interphalangeal joint of the middle
finger; [b] the pulp of the index finger; and [c]
the pulp of the thumb and when a stick is
longer, the radial aspect of the

metacarpophalangeal joint of the index finger
makes contact with it. (p. 441)

Trombly and Cole (1979) used electromyography to

determine the involvement of four hand muscles in 16

activities. They found that the extensor digitorum, dorsal

interosseus 1, abductor pollicis brevis, and flexor

digitorum profundus were about equally involved in the

upstroke and downstroke of a pencil but that there were

individual variabilities in joint motion among the 15


Ziviani (1983) used photographic analyses to study the

qualitative changes in dynamic tripod grip between ages 7

and 14. She identified the following variables: degree of

flexion of the index finger, degree of forearm

pronation/supination, fingers used, and opposition. In a

subsequent study, Ziviani and Elkins (1986) found that the

type of grip did not influence the speed and legibility of

writing. However, they noted that the dynamic application

(i.e., the use) of the grip may be more pertinent to writing

performance than the type of static grip.

Therapists who have worked with handicapped children

have developed evaluations of handwriting that focus upon

the motor process of handwriting versus the analysis of the

written handwriting product (Goodgold, 1983; Ziviani &

Elkins, 1984). The Handwriting Movement Rating Scale was

developed to assess the quality of handwriting movements in

prekindergarten and kindergarten children (Goodgold, 1983).

This assessment included ratings of grip, posture, upper

extremity muscle coordination, and sequence of movements


Upper extremity strength and dexterity have been

determined to be needed for handwriting. Individuals with

muscle weakness have received therapy for improving strength

and coordination as a basis for facilitating function in

various upper extremity skills (Stephens & Pratt, 1989).

However, research on the effects of such therapy

specifically on handwriting has been limited to one study.

In Hong Kong, Kao (1973) examined the effects of hand and

finger exercises on tracing in normal adults and concluded

that these exercises could enhance precise visual motor

control for tracing.

Motorically, the novice handwriter should present with

adequate muscle tone and coordination of the trunk,

shoulder, and hip girdles and upper and lower extremities to

assume and maintain writing postures of the body and hand

(Stephens & Pratt, 1989). Adequate upper extremity function

depends upon trunk and scapulohumeral stability. In order

for the hand to function, the trunk must maintain an upright

position so that the arms are not used to support the body.

The scapula must be stable upon the upper thorax with the

humerus well articulated into the shoulder joint. Hand

grasp must be mature for adequate grip of the pencil and

forearm supination must be well developed to allow fluid

movement of the hand across the page (Ziviani, 1983). In

addition, experienced eye-hand coordination is required for

the child to manipulate the pencil and reproduce geometric

shapes that are the precursors to letter forms (Petty et

al., 1985; DeHaven, 1983).

In order to establish the motor plan of movements

needed for future encounters with a given letter, the child

must be able to integrate the visual image of the letter

with muscle kinesthetic and proprioceptive sensations

(Ayres, 1985). As the hand moves to write a letter, and as

the eyes move to focus on relevant features of the letter,

kinesthetic and proprioceptive sensations are produced in

the hand muscles and eye muscles. These sensations are

characteristics of the eye and hand movements needed to form

the letter and are matched to the visual image of the

letter. Through this process, the nervous system

establishes motor patterns for each letter.

In this section, aspects of grip and posture were

described. In the next section, visual aspects of

handwriting copying tasks are discussed.

Visual Aspects of Graphomotor Skills

Visual aspects of graphomotor skills include acuity,

eye movements, and depth perception. Each of these visual

aspects will be discussed.

Visually, the child needs to be able to be free of

refractive errors so that the information on the chalkboard

is seen clearly. Kindergarten screening for visual deficits

is assumed to have identified those children needing

refractive examination. During a copying task, the eyes

must alternately adjust for near point focus as required to

view the paper and for far focus as required to view the

chalkboard. Once the chalkboard or paper is viewed, the

eyes move to seek out the relevant stimulus, focus on that

stimulus, and move in coordination with the head. These

oculomotor functions are accomplished through five types of

eye movements: the "near triad," conjugate eye movements,

saccades, fixation, and compensatory movements.

For near point vision, three eye movements, the "near

triad" occur: convergence, lens rounding, and pupillary

constriction (Kuffler, Nicholls, & Martin, 1984). The two

eyes converge to align the eyes with a near object and this

results from cocontraction of the medial rectus muscles of

the eyes. Rounding of the lens increases its optical power.

Contraction of the ciliary muscles results in relaxation of

tension, and thus rounding, of the lens. Constriction of

the pupil improves vision through limiting the entrance of

stray light. Pupillary constriction results from

contraction of the pupillary sphincter. For far vision, the

reverse of the near triad occurs.

In order to locate the relevant stimulus on the paper

or chalkboard, four eye movements are involved: conjugate,

fixation, saccades, and compensatory (Kuffler, Nicholls, &

Martin, 1984). Conjugate eye movements provide movements of

the eyes in the same direction. For example, to look to one

side, the medial rectus of one eye contracts synchronously

with the lateral rectus of the other eye. This movement

involves the occipital lobe, frontal lobe, paramedian

pontine reticular formulation center for lateral gaze in the

brainstem pons and cranial nerves III and IV.

Saccadic eye movements refer to voluntary movements of

the eyes in which the eyes flick rapidly from one point of

fixation to another as when reading. These movements

involve the frontal lobe and the brain stem.

Fixation on a visual target is accomplished and

maintained through foveation or moving the eyes to locate

the target on the foveas, smooth pursuit of the target as it

moves so as to maintain fixation and near triad to maintain

fixation as the target nears. Fixation involves the

occipital lobe and the brain stem.

Compensatory movements of the eyes occur as the head

moves. So that visual fixation is maintained, the eyes move

in an equal and opposite direction to the head.

Compensatory movements involve the vestibular system and

cranial nerves to the eyes muscles.

These five types of eye movements are involved in

copying from the chalkboard. For close work, the near triad

allows the child to focus clearly on the paper. Conjugate

movements occur as the child looks up to the chalkboard and

across the line of words written there. Saccadic movements

occur as the child focuses on a specific letter and then

another until the desired letter is located. Once the

letter is located, the eyes fixate on the letter while the

visual information is being processed. Finally,

compensatory eye movements have been observed as the child

begins to turn the head to the paper but the eyes linger on

the chalkboard a moment longer after head rotation is


Stereoscopic depth perception (Barlow & Mollon, 1982;

Crawford, 1986) is involved in the copying task as the child

focuses from desk to chalkboard. When the child looks

directly at a letter on a page, the letter is projected onto

the fovea of each retina, the foveas being the points of

sharpest vision. These points are in corresponding

locations on the two retina, thus the two images formed by

the two eyes are perceived as fused into one image. As the

child looks to the chalkboard, disparate points are

stimulated on each retina and some objects in front of the

child will appear closer than others. For example, if the

child is looking straight ahead with the eyes focused on a

letter on the chalkboard, the letter will stimulate

corresponding points on the two retina. However, the

teacher standing in front of the chalkboard, the children in

the desks in front of the child, the bulletin board to the

side of the chalkboard will all stimulate different retinal

points. Depth perception occurs through the resolution of

stimulation of the disparate retinal points and this

resolution is brought about by computational aspects of

visual processing in the visual cortex.

To see a word and then write it, the brain must attend

to the word to be written and not attend to other stimuli in

the environment. Memory for what is to be copied must be

retained during lexical, semantic, and graphemic processes.

Attention to the motor movements are required for

graphomotor responses. Neurological mechanisms underlying

arousal, attention, and intention will be discussed next.

Attentional Mechanisms for Graphomotor Skills

The role of the brain stem arousal system, the frontal

lobes and the inferior parietal lobule will be described

briefly. In the brain stem, the midbrain reticular

formation (MRF) receives excitation by visual, auditory, and

somatosensory sensory systems. The MRF, in turn, has an

excitatory input to cortical areas where sensory information

is interpreted and processed. The MRF provides arousal of

the cortex through the inhibition of the nucleus

reticularis, an inhibitor of the flow of sensory input from

the thalamus to the cortex (Heilman, Watson, & Valenstein,


Attentional mechanisms mediated by the inferior

parietal lobule have been investigated in recent years in

single cell recordings in awake monkeys (Heilman, Watson, &

Valenstein, 1985). Cells in this lobule are most active

when stimuli are presented that are important to the animal

and not active during presentations of insignificant

stimuli. Several types of cells have been identified and

were named for the function during which they were very

active. Projection cells were active when the limb reached

for or the hand manipulated a significant stimulus such as

food. Visual fixation neurons were active when the distance

of the stimulus was within arm's reach, the stimulus was

biologically significant or when the eyes were directed

towards a certain part of the visual field. Visual tracking

neurons were active during visual pursuit of a moving object

in a given direction within arm's reach. Similarly, saccade

cells were active during saccadic eye movements to visually

explore biologically significant stimuli. The inferior

parietal lobule is thought to be important for visual

attention to stimuli that are important to the organism.

Aside from the motor and language functions of the

frontal lobes, these lobes have additional roles in

attention, intention, and arousal (Damasio, 1985). Frontal

lobe functions have been determined through study of persons

with lesions affecting the frontal lobes and from animal


The frontal lobes help to focus attention on the task

at hand. The frontal eye fields located rostral to the

frontal motor areas have been cited as the locus of

attentional mechanisms. Individuals with lesions to these

areas have difficulty with selective attention and lose the

ability to orient to stimuli in contralateral space

(Damasio, Damasio, & Chui, 1980; Heilman & Valenstein,

1972). Some persons are unable to orient to stimuli that

are relevant and, instead, attend to irrelevant stimuli.

Organized movements of the eyes to scan a visual stimulus

and increasingly fixate on important aspects of the stimulus

are impaired and the eyes move in random fashion (Luria,


The dorsolateral frontal cortex is thought to be

important for intention, a plan of action based upon

information inherent in a situation. This function of the

frontal lobe enables one to notice cues in the environment

or within one's self as to what needs to be done next.

One's motives or intentions are tempered by social and

personal constraints. Persons with lesions in the

dorsolateral frontal cortex are unable to develop strategies

and plans and lack the ability to benefit from feedback.

Problems with disorganization, decision-making, taking

responsibility, and observing social rules are seen

(Damasio, 1985).

When a child copies information from a chalkboard in

school, the eyes must engage the relevant stimuli on the

chalkboard, the sights and sounds of neighboring students

must be ignored, the child must develop a plan to complete

the task and be able to persevere until finished. These

behaviors require the attentional mechanisms described


In summary, writing movement is an end product that

reflects linguistic, visual, motor, and attentional

programming functions of the brain. Graphic motor output is

possible when linguistic components are converted to motor

output and nonverbal, visual spatial orientation contributes

to the programming for graphemic motor output in the context

of attention to the given task. Discussed next are

observations of these graphomotor processes in normal first

grade children learning to write.

Qualitative Investigation of Copying Behavior

In a qualitative investigation of 20 first grade

children in a normal first grade classroom, the following

motor, visual, and attentional subtasks were observed to be

involved in copying from the chalkboard (Walker, 1988). The

typical head posture for handwriting included the head bent

in slight forward flexion, tilted slightly towards the right

to bring the nose slightly in the direction of the right


shoulder. This posture enabled the child to bring the eyes

to the paper to see what was being written. Shoulders were

generally level and not tilted. Upper arms were held in

elbow and shoulder flexion to orient to the task. The

nonwriting hand stabilized the paper. The trunk was held

erect with slight forward flexion of the trunk to bring the

body against the table edge. The pelvis was level and

lateral tilting was not observed. The thighs and lower legs

were held in relaxed position with both feet on the floor.

Some children crossed their legs at the ankles but none of

them crossed their legs at the knees. The mature three-

finger tripod pencil grasp (Ziviani, 1983) was predominantly


The children used their eyes to monitor and guide the

hand, but the eyes were directed to the letter being

produced on the paper and not on the hand itself (Walker,

1988). The children were able to integrate the kinesthetic

and proprioceptive sensations being produced by the movement

with the visual image of what was being produced on the

paper and to use this integration in eye-hand coordination.

Most children copied only one letter at a time. To look for

the next letter to be copied, the child rotated his or her

to midline and held the head steady while looking at the

chalkboard to locate and focus upon the next letter to be

written. Having found the letter, they made postural

adjustments to bring the head, eyes, and hands again into


the writing position. Children concentrated on the task and

their demeanor was one of absorption with the task.

In summary, copying from the chalkboard is a highly

complex task, demanding visual acuity and perception,

attention to the task, and ability to organize sequences of

motor movements in response to visual stimuli. Intact

visual, motor, and attentional abilities are needed for

handwriting development.

Handwriting Evaluation

Several assessments of handwriting products exist in

which a sample of the writer's handwriting is rated by an

examiner (Barbe, Lucas, Hackney, Brain, & Wasylyk, 1984;

Bezzi, 1962; Freeman, 1959; Hammill & Larson, 1983; Phelps,

Stempel, & Speck, 1985; Stott, Moyes, & Henderson, 1985;

Ziviani & Elkins, 1984). However, the psychometric

properties of handwriting measures have been questioned for

flaws in validity, reliability, and usability (Graham,

1986a, 1986b).


The validity of handwriting assessment is affected by

"ambiguity of the criterion variables" (Graham, 1986a, p.

64) in which definitions of terms such as "legibility,"

"well-proportioned," or "properly formed" are imprecise. An

individual's handwriting varies from day to day and

"physiological, experiential, motivation, and psychological

factors" (Graham, 1986a, p. 65) have been determined to


influence handwriting performance. Thus, it may be invalid

to assume that performance on a single handwriting sample is

representative of the individual's capacity. The nature of

the assignment can also influence handwriting performance

since persons perform differently under different

performance expectations. The person who scores the

handwriting sample can introduce variability by imprecise

application of a set of criteria, by fatigue or by knowledge

of the intended consequences to the subject of the scoring

results. Although handwriting scales have adequate face

validity, the population of writing samples are not

comprehensive for the range of performance for a group of

children. Lack of concurrent validity study is due to the

paucity of handwriting measures and to the standardization

limitations of the available instruments.

Of the handwriting instruments reviewed, the Evaluation

of Handwriting Performance (EHP) (Ziviani & Elkins, 1984)

included content, criterion-related, and construct validity

investigations. For content validity, a specification table

was used to judge that EHP subtests were representative of

handwriting components in research reported in the

literature, including formation, spacing, alignment, size,

and speed. Each test component included items for letters

and symbols. For criterion validity, subjects' scores on

letter and symbol test items were thought to be measures of

the same handwriting criteria and these scores were

correlated. Some agreement was found on measures of the

same criteria: .72 for formation, .52 for spacing, .67 for

alignment, and .76 for size. Construct validity was

investigated through factor analysis. Of the total

variance, 70% was accounted for by four factors: formation,

size, spacing, and alignment. Other tests reviewed did not

include validity investigations.


For the tests reviewed, reliability data were scarce,

yet, when available, were generally acceptable. For the EHP

(Ziviani & Elkins, 1984) interrater reliability correlations

for formation, spacing, alignment, and size were,

respectively, .76, .91, .88, .97, and .95. Test-retest

reliability correlations were .79 for formation, .51 for

spacing, .48 for alignment, .61 for size, and .93 for speed.

Test-retest correlations for the grade 3 subjects were lower

than retest correlations than the grade 6 subjects and the

investigators interpreted this as being due to a greater

variability and range of performance of younger children as

compared to older children.

Test-retest reliability coefficients were reported at

.81 to .90 on the Test of Written Language (TOWL) (Hammill &

larson, 1983). Interrater reliability of the TOWL was

established by comparing 15 teacher's ratings to the average

rating. If the individual teacher's rating was within one

point of the average rating, the teacher was assumed to be

in agreement. The percentage of agreement was .76.

Reliability of .38 to .87 has been reported for the

Freeman Scales (Feldt, 1962). Zaner-Bloser (Barbe et al.,

1984) reliability is reported at .74 to .98. The average

scorer agreement in the ratings of two judges on the

Children's Handwriting Evaluation Scale (CHES) (Phelps et

al., 1985) ranged from .88 to .95 and reliability ratings

for a single rater ranged from .64 to .82. Interrater

reliability for the Diagnosis and Remediation of Handwriting

Problems (DRHP) (Stott et al., 1988) was assessed by

comparing the ratings of two assistants to the primary

author of the test and reported at .56 and .66.


Although the results of existing handwriting

assessments provide general information about handwriting

performance, they provide limited diagnostic information as

to the specific handwriting problems. A handwriting sample

is rated in toto on a 0-10 scale (Hammill & Larsen, 1983;

Phelps et al., 1985) or according to criteria lists (Ziviani

& Elkins, 1984; Stott et al., 1986). Criteria for letter

formation is limited either to manuscript writing (Ziviani &

Elkins, 1984), to cursive writing (Hammill & Larsen, 1983)

or contains overlapping categories (Stott et al., 1986).

Norm-referenced information is lacking except for one test

developed in the United States where handwriting is scored


as part of an overall language evaluation (Hammill & Larsen,

1983) and one test developed in Australia (Ziviani & Elkins,

1984) which has limited use for non-Australian children.

Existing handwriting scales provide cursory evaluation of

handwriting performance but are insufficient for planning

handwriting intervention.

In using existing instruments to assess handwriting of

children with learning disabilities, the investigator

assessed the usability of three instruments: the TOWL

(Hammill & Larsen, 1983), the DRHP (Stott et al., 1985), and

the Evaluation of Handwriting Performance (EHP) (Ziviani &

Elkins, 1984). An advantage of the TOWL is that it was

standardized on children 7 to 18 years of age in the United

States. In addition, normative data are provided for

children in the United States. The TOWL (Hammill & Larsen,

1983) and the DHRP handwriting samples were limited to

spontaneous writing in response to pictures and, thus,

handwriting for copying or in response to dictation could

not be measured. The DHRP was developed in England as a

comprehensive, criterion-referenced assessment of

handwriting that could be used with children and adults.

For assessing letter formation, the DHRP was useful because

Part I of the instrument contained 14 categories of "faults"

that could occur with handwriting and representative

problems were depicted. However, the categories were not

mutually exclusive and it was difficult to separate letter


formation "faults" in Part I on the instrument from problems

in Part II of the instrument. For example, one fault in

Part I was that the "letter has an unusual feature that

interferes with reading" (Stott et al., 1986, p. 16), yet in

Part II there was a rating for "random letter distortion"

that was scored 0 to 3. As noted in Daniels' (1988) review,

the clinical usefulness of the DHRP is limited by the lack

of guidelines for score interpretation.

Handwriting samples on the EHP included drawing

geometric forms, writing words to dictation, copying a

sentence, and writing a phrase repeatedly to assess writing

speed. Scoring methodologies for word alignment, spacing,

and size were readily usable in the EHP. However, the

evaluation of letter formation was limited to manuscript

writing of children in grades 1 through 3. The EHP was

standardized on Australian children. Furthermore, samples

were written on unlined paper, which may not be appropriate

for use in the United States where children traditionally

are not expected to write on unlined paper. However,

scoring methods for letter formation and word spacing,

alignment, and size were well-described.

In this section, handwriting evaluation has been

discussed. Problems with validity, reliability, and

usability were reviewed. Next, methodologies for

handwriting instruction will be described.

Handwriting Instruction

In this section, general information on handwriting

instruction will be reviewed. Second, single subject

research studies of handwriting instruction will be


Background Information

Background information on handwriting from sundry

sources will be considered next. The following aspects of

handwriting instruction will be briefly described:

developmental aspects; modeling; manuscript versus cursive;

Palmer, Zaner-Bloser, and D'Nealian methods; the letter as

the basic unit of handwriting; handwriting legibility;

handwriting rate; and computer versus handwriting modes of


Developmental aspects

Manuscript handwriting instruction begins in first

grade with cursive introduced in the second half of the

second grade (Mercer, 1987). Formation of letters are

emphasized in the first years, and after grade 3 composition

and writing as a form of communication are stressed. By the

end of the first three years of school, most children have

developed handwriting that is legible and have begun to

exhibit individual styles of handwriting.

During the first year in school, the child learns to

write all 26 alphabet letters, lower and upper case, and the

9 numerals. In grade 2, the child refines the letter


formations and begins cursive writing in the second semester

of that year. By the end of the third year in school, the

child has mastered 26 upper and lower case manuscript and

cursive letters for a total of 104 letter forms and uses

handwriting for creative writing and in practical classroom

situations (DeHaven, 1983).

By school age, the child is able to adopt the typical

position for writing with the pencil held with a tripod

grasp, 1 inch above the tip with the eraser end pointed

towards the right shoulder for right-handed persons and the

left shoulder for left-handed persons. The paper is

positioned straight for manuscript writing and tilted with

the lower left corner (or for left handers, the lower right

corner) towards the center of the body (Petty et al., 1985).

Studies of the developmental mastery of letter

formation in normal children have inconsistent results in

terms of the order in which letters are mastered. In a case

study of a 2 1/2 year old girl, McCarthy (1977) observed

that circular letters (o, c, q) were acquired first,

straight line letters (1, t, i) next, letters with loops (b,

p, r) next, and letters with acute angles and diagonal lines

(m, n, z) last). Coleman (1970) found that the first

letters 5 year olds learn to copy were 1, r, f, t, and x,

and the last letters they learned to copy were w, d, z, y,

and q. Lewis and Lewis (1964) ranked the letters 1, v, c,

x, and h as the easiest letters and q, g, p, y, and j as the

most difficult letters for first graders to write.

Instructional and measurement variables may have been

significant factors in the results of these studies.


Modeling of the letters can affect outcomes of

handwriting instruction. For a child to be able to write a

letter, the child first needs a visible model of the letter.

Next, the child can produce the letter from a model

"written" in the air. Third, verbal descriptions of the

letter can guide the child in writing. Finally, the child

can produce the letter by hearing just its name (McCarthy,

1977). Motion models of how to move to write the letters

were found to be more effective than still letters in

children's acquisition of letter forms (Wright & Wright,

cited in Peck, Askov, & Fairchild, 1980; Sovik, 1979). In a

study of copying abilities of 21 7-10 year old children,

Sovik (1979) found that children performed better copying

from a model placed 2 feet away than one placed 12 feet


Manuscript versus cursive

Educators have mixed opinions as to whether manuscript

or cursive writing is preferred for early instruction.

Proponents of manuscript (Barbe, Milone, & Wasylyk, 1983;

Graham & Miller, 1980) questioned the need to teach two

writing forms, manuscript and cursive, when manuscript can

be just as fast and legible as cursive. Movements for

manuscript writing are thought to be simpler and easier to

learn than cursive. Barbe and colleagues (1983) analyzed

the writing of the word "was" and found 12 strokes, 11

joining, 3 letter joining, and 6 retraces for cursive

writing as compared to 7 strokes, 4 stroke joining, and no

letter joining and retracings for manuscript writing. They

concluded that cursive was four times as difficult as

manuscript writing. Similarly, Jackson (1971) found that

manuscript writing was significantly faster than cursive for

intermediate grade children. Contrary to this, Suen (1983)

found cursive to be faster than manuscript or block

printing. Manuscript writing is needed throughout life for

its legibility as required for completing application and

other forms. The independence of letters in manuscript is

also believed to promote spelling (Hagin, 1983).

Manuscript letters more closely resemble typeset

letters than do cursive letters. Since reading and writing

are part of the same language process, it is believed that

manuscript writing is preferable for its resemblance to the

letters the child encounters when reading (Koenke, 1986).

However, Duvall (1984) found that kindergarten children

could do equally well in matching or reading manuscript,

cursive, and italic handwritten letters compared to typeset

letters. She concluded that the rationale for using

manuscript because it resembles text was not justified.

Apparently, the children in Duvall's (1984) study had

established concepts of the invariant aspects of letters

that allowed them to identify the letters in various

contexts. This ability is what Ellis (1979, 1982) referred

to as graphemes, the abstract specifications of a particular

letter that can be expressed in a variety of forms.

Cursive has been suggested as preferable for learning

disabled children for several reasons (Serio, 1970). The

writing movement is more rhythmic, flowing, and efficient.

Second, spacing is easier since letters in a word are not

segmented. Third, there is less opportunity for

directionality confusion and letter reversals of "b" and

"d," "p" and "q," "m" and "w." In addition, Hagin (1983)

suggested that, in cursive writing, words are dealt with as

units rather than segmented letters and this benefits the

learning disabled child.

Palmer. Zaner-Bloser, and D'Nealian methods

To date, there is no one agreed-upon best method of

handwriting instruction. However, the most frequently

mentioned commercial methods include the Palmer, Zaner-

Bloser, and D'Nealian (Wood, Webster, Gullickson, & Walker,

1987) with italic handwriting as an alternative (DeHaven,


More than a century old, the Palmer method is the

oldest method in use and includes vertical manuscript

letters with a change over to slanted, flowing cursive in

grade 3. The Zaner-Bloser program was begun in 1913 and,

like the Palmer method, has markedly different manuscript

and cursive letter forms. The D'Nealian method (Thurber,

1983, 1984) was devised in 1968 and differs from the two

established methods in that the manuscript letters are

simplified cursive letter forms that ease the transition

from manuscript to cursive writing. Italic handwriting,

which dates to the 16th century, has been revived in recent

years (Dubay & Getty, 1980) because the letter forms require

few lifts of the pencil and the child learns only one set of

lower and uppercase letters.

Little research has been done on any of the methods,

yet these methods are widely used. Wood and colleagues

(1987) found no significant differences in legibility when

the Palmer, Zaner-Bloser, and D'Nealian methods were

compared for teaching 3,376 fourth and sixth grade children

in South Dakota. These authors reported that they received

3,376 individual handwriting samples from children in 89

elementary schools, yet a subject demographics table

depicted a total of 347 subjects. Since each subject was

only evaluated on one sample, it was not clear whether the

analysis of variance findings were based upon data from 347

subjects or 3,376 samples. If the authors reduced the

original 3,376 samples for data analyses they did not

indicate how this was done. Nonetheless, the finding that

three prominently-used methodologies did not differentially

affect handwriting legibility was relevant for considering

handwriting instruction.

In a longitudinal study of 86 first and second graders

in four classrooms, Farris (1982) using D'Nealian or Zaner-

Bloser instruction found no significant differences in

manuscript writing at the end of first grade. Students

instructed with the D'Nealian method had fewer letter

reversals and the author suggested that the continuous

stroke aspect of the D'Nealian method prevented the

development of perceptual motor difficulties found in more

fragmented letter formations. For the second grade cursive

writing, the Zaner-Bloser method resulted in significantly

higher legibility ratings. However, for both grades, there

was a significant correlation between reading ability and

legibility of handwriting produced with D'Nealian


Duvall (1985) investigated the difficulty of four

handwriting instructional systems: italic, Zaner-Bloser

manuscript, D'Nealian manuscript, and Palmer cursive. This

study provided useful criteria for calculating a letter

difficulty score. However, comparisons between manuscript

and cursive styles and promotion of the author's italic

style as the preferred method are limitations of this study.

Letter as the basic unit in handwriting

In psychophysical analyses of handwriting features,

Teulings, Thomassen, and van Galen (1983) found that the

whole letter was the basic unit of writing in adult's

cursive writing. Thus, emphasis on individual letter forms,

regardless of which form is taught may be the most

appropriate instructional approach. Using methodology

similar to Teulings et al. (1983), Wing, Nimmo-Smith, and

Elridge (1983) found that cursive letter forms were more

consistently produced in the middle or end of the word than

in the beginning. For example, when writing the words

"advantage" and "amalgam" there was greater variability in

subjects' formation of the letter "a" at the beginning of

the word when compared from word to word than there was for

the letter "a" compared within a word. Although an initial

"a" varied from word to word, once the initial "a" for a

word was written, the tendency was to write the letter "a"

in similar fashion throughout the word. Apparently, a motor

plan for writing the letter is generated at the beginning of

a word and when that letter is used subsequently in the

word, the same motor plan is used. Thus, when letters are

evaluated for letter form, the letter order in the word may

be considered.

Handwriting legibility

Five factors are commonly referred to when discussing

factors affecting handwriting legibility: letter form,

uniformity of slant, uniformity of alignment of letters,

line quality, and letter and word spacing. Use of these

factors to assess handwriting legibility date to Freeman's

(1915) scale for assessing school children's handwriting

quality. Early studies of errors in letter formation and

errors in spacing, alignment, slant, and line quality will

be reviewed briefly.

Letters formation errors were studied in terms of

letters that were most frequently illegible. In a study by

Pressy and Pressy (1927) 16 graduate students analyzed 3,000

illegibilities in 650 papers written by elementary school

children and high school and college students. Nearly one-

half of the illegibilities were due to malformations of the

letters r, n, e, a, d, and o. In Newland's (1932) study, 24

raters analyzed in excess of one million letters written by

2,381 persons including elementary school children, high

school students, and adults. Four letters, a, e, r, and t,

accounted for nearly half of the illegibilities in the three

groups tested. Handwriting samples were from papers written

by elementary, high school, and college students and from

letters to the editor written by adults. Lewis and Lewis

(1965) found that the letters q, g, p, y, and j were most

frequently malformed in their analysis of 52 manuscript

letters in the upper and lower case alphabet written by 354

first grade children. In the cursive writing of 1,000 sixth

grade students, Horton (1970) found that 28% of all the

letters were illegible. The letter "r" accounted for 12% of

the illegibilities; the letters r, h, i, k, p, and z

accounted for 30% of the illegibilities; and the letters a,

b, c, i, 1, m, n, u, v, and x accounted for 12%.

Common letter formations errors in manuscript and

cursive writing considered together included n like u; r

like i, e closed, d like cl, c like a, a like o, a like ci

(Newland, 1932; Pressy & Pressy, 1927). Common manuscript

errors included: reversals, omissions, additions,

malalignments, mishapeness, rotations, retracings,

inversion, incorrect relationship of letter parts, and sizes

that are too large or too small (Lewis & Lewis, 1965).

Compared to common practice in educational research

today, these early studies have several methodological flaws

(Horton, 1970; Newland, 1932; Pressy & Pressy, 1927).

Subjects' academic ability and age and cursive versus

manuscript writing were not considered. Data analysis was

limited to percentages rather than statistical significance.

Methodology for selecting the handwriting samples to be

analyzed and training of the raters were limited.

In a comprehensive study of factors that affected

handwriting legibility, Quant (1946) considered legibility

to be synonymous with readability. Handwriting legibility

was measured according to the rate and accuracy by which

material could be read at a given distance. Subjects' eye

movements were filmed with a 35 mm motion picture camera and

the number, location, and duration of each fixation of the

eyes during reading was recorded. Legibility was measured


for the average number of words read per regressive movement

of the eyes to reread something. The following conclusions

were made. Cursive writing was less legible than manuscript

writing. Letter formation was the most important factor in

readability. Compact spacing of letters (0.7 sixteenths of

an inch) and words (3.1 sixteenths of an inch) was more

readable as measured in amount of words read per eye

fixation, than was widely spaced letters and words. Effects

of overcrowding of letters or words was not studied. There

was no difference in readability of lines placed 5/8" or

1/4" apart. Unevenly aligned writing did not significantly

affect legibility. Irregularity of letter slant resulted in

a decrease in the average number of words read each time the

eyes had to look back a second time at a word. Results

regarding the effects of the heaviness of the writing line

were inconclusive.

The influence of handwriting variables, sex and age, on

handwriting legibility was investigated in 588 children in

grades 4, 5, and 6 (Andersen, 1969). The findings were that

females are superior to males and that females write with a

more perpendicular slant than do males. With increased age

of subjects, the legibility, uniformity of size and

uniformity of slant improve and writing size decreases.

Improved legibility was related to larger writing and

uniformity of slant. Larger sized writing was less uniform

in size. Writing with a pronounced slant had a more uniform



In summary, the types of letter formation errors (Lewis

& Lewis, 1965; Newland, 1932; Pressy & press, 1927) and the

five factors affecting legibility (Freeman, 1915; Quant,

1946) are frequently cited in the instructional and

measurement literature on handwriting. The lists of letter

formation errors and the five legibility factors have been

used to devise handwriting checklists for classroom use

(Cohen & Plaskon, 1980; DeHaven, 1983; Mercer, 1987; Petty

et al., 1985) and for handwriting measurement (Anderson,

1969; Hammill & Larsen, 1983; Stott et al., 1985).

Handwriting rate

Rate of motor responding for handwriting has been

calculated in letters per minute. Research on handwriting

rate is minimal although rate is easily measured (Graham,

1986). Rates from the Zaner-Bloser Scales are frequently

cited (Hammill & Larsen, 1983) and these rates are

comparable to rates published in other evaluation systems

(Phelps et al., 1985; Ziviani & Elkins, 1984). Rates in

letters per minute for writing words by children grades one

through seven are as follows: grade 1 25, grade 2 -30,

grade 3 38, grade 4 45, grade 5 60, grade 6 67, and

grade 7 74.

Computers and handwriting

Computers have been used as an alternative to

handwriting for students with handwriting difficulties.

Word processing allows students with handwriting problems to


write, edit, and produce legible work that is self-enhancing

(Bing, 1988). Computer-assisted instruction to improve

handwriting motions was identified in only one study (Lally,

1982). The superiority of computers for speed and

composition has not been supported. MacArthur and Graham

(1987) found no differences in length, quality, and grammar

in handwritten and computer-produced stories. Students

revised at the end with handwriting and as they went along

with computers. In contrast, Schanck (1986) found no

significant differences in fourth grade students' editing

and revising on handwritten versus computer stories. In

terms of speed, handwriting was faster than use of a

computer in adolescents with mild mental handicaps (Vacc,

1987) and in students with learning disabilities (MacArthur

& Graham, 1987).

Single Subiect Research

Applied behavioral analyses and single subject research

in handwriting have been limited in number and have focused

on the effects of feedback, self-correction, and

reinforcement on rate and accuracy (Kerr & Lambert, 1982).

Twelve studies were reviewed and are summarized here by

describing the subjects, designs, measurements, dependent

variables, independent variables, and results. Applied

behavioral analysis was used in two studies that were

nonexperimental field tests of handwriting evaluation

instruments (Helwig, Johns, Norman, & Cooper, 1976; Jones,

Trap, & Cooper, 1977).

Subjects included regular elementary school children

(Jones et al., 1977; Trap, Milner-Davis, Joseph, & Cooper,

1978) children with handwriting problems (Helwig, 1976),

learning disabilities (Blandford & Lloyd, 1987; Graham,

1983; Kosiewicz, Hallahan, & Lloyd, 1981; Smith & Lovitt,

1973), mental retardation (Clark, Boyd, & Macrae, 1975;

Stowitschek, Ghezzi, & Safely, 1987), low socioeconomic

background (Fauke, Burnett, Powers, & Sulzer-Azaroff, 1973),

and gifted abilities (Salend, 1984). The design most

frequently used was the multiple baseline design (Blanford &

Lloyd, 1987; Clark et al., 1975; Graham, 1983; Helwig, 1976;

Salend, 1984; Stowitschek et al., 1987; Trap et al., 1978)

with reversal designs used in three studies (Fauke et al.,

1973; Kosiewicz et al., 1981; Smith & Lovitt, 1973).

Multiple baseline design was preferred since reversal of the

motoric academic skill being trained was undesirable (Kerr &

Lambert, 1982).

The lack of valid and reliable measures of handwriting

legibility was apparent in the use of researcher-developed

measurement techniques in most studies. Transparent

overlays were used to judge 1, 2, and 3 mm leeway of the

approximation of the child's letter form to model letter

forms (Fauke et al., 1973; Helwig, 1976; Helwig et al.,

1976; Jones et al., 1977; Trap et al., 1978). Stowitschek

and colleagues (1987) developed a correction template with

white letters against a black background and used

semitransparent worksheets that were placed over the

templates for evaluation. Criteria for judging letters were

developed for assigning points to letter form. For example,

Blanford and Lloyd (1987) assigned one point for each

criterion when words were 1/8 but not more than 1/4 inch

apart and when the letter was not more than 1/16 inch above

or below the line; correctly formed; the correct height; and

one-half space tall, if lower case. This rating system was

tested on four non-learning disabled students whom the

classroom teacher identified as having average handwriting.

Dependent variables consisted of rate of percentage

correct letters. Children received training in 2 letters

(Graham, 1983; Smith & Lovitt, 1973); 4 letters (Salend,

1984); 8 letters (Stowitschek et al., 1987); 10 letters

(Helwig, 1976; Jones et al., 1977; Trap et al., 1978); or

the 11 letters in their name (Fauke et al., 1973). Daily

journal writing (Blanford & Lloyd, 1987), paragraph copying

(Kosiewicz et al., 1981), and completing biographical

information forms (Clark et al., 1975) were dependent

variables in three studies.

Self-teaching and self-assessment comprised the

independent variables in most studies. Independent

variables included self-instruction (Blandford & Lloyd,

1987; Graham, 1983; Koseiwicz et al., 1981), self-evaluation

(Blandford & Lloyd, 1987; Kosiewicz et al., 1981;

Stowitschek et al., 1987), self-verbalization of errors

(Helwig, 1976), and self-recording (Jones et al., 1977).

Careful modeling of letter formation (Salend, 1984; Trap et

al., 1978), the use of modeling and yarn letters on cards

(Fauke et al., 1973), comparisons of b and d reversed

letters (Smith & Lovitt, 1973), and emphasis on practice

(Clark et al., 1975) were used in handwriting instruction.

All the studies reported improvements in handwriting.

Thus, self-assessment, self-instruction, careful modeling,

and practice were effective techniques for improving

handwriting. Children were reliable in assessing their own

work (Jones et al., 1977). An average of 48 practice

minutes were needed to reach criteria on a set of eight

letters in children with mental retardation (Stowitschek et

al., 1987).

Improvements showed some generalization to other

settings (Blanford & Lloyd, 1987; Stowitschek et al., 1987)

and to the ability to write letters for which intervention

had not been provided (Helwig et al., 1976; Jones et al.,

1977; Trap et al., 1978). However, Graham (1983) found that

the improvements in the ability to write the two letters for

which intervention had been provided did not carry over to

four other letters which were merely practiced but for which

intervention was not provided. He concluded that this

handwriting training was not cost-effective. He suggested


that the complexity of the treatment procedures and too few

training sessions may have contributed to the results. The

children in this study verbalized the movements as they

formed the letters and Graham (1983) posed that this

technique could interfere with, rather than enhance, the

writing motion.


The literature on handwriting has been reviewed in

terms of handwriting as a neuromotor process. A

neuropsychological model of graphomotor skills was proposed

and motor cortices, psychophysical features of handwriting

and grip and posture were discussed. Visual aspects of

handwriting were described and attentional mechanisms for

writing were reviewed. Limitations of measurement tools for

assessing handwriting were described. Finally, handwriting

instruction was discussed with emphasis upon single subject

intervention in handwriting problems.

In this literature review, an instructional issue that

has not received attention in the literature was identified:

distance of the stimulus copying tasks. In summary, the

literature reviewed in this chapter on motor, visual,

attention, instructional, and measurement issues related to

handwriting has provided relevant and useful information for

considering stimulus distance for copying by school

children. In the next chapter, the methodology for the

study is presented.



The purpose of this study was to investigate the

effects of three distances of the stimulus from the subject

on handwriting legibility, as measured for letter formation

and word spacing, of children performing a copying task.

The three stimulus distances as measured from the stimulus

to edge of the table next to the child were 0.5 meters from

the stimulus placed on the desk, 3 meters from the stimulus

placed on the chalkboard, and 6 meters from the stimulus

placed on the chalkboard.

These three distances are commonly used for copying

instruction in the regular classroom setting. They have

been used when the child copies material placed on the desk

or when the child copies material written on the chalkboard

while seated in either the front row of desks or in the back

row of desks. Letter formation and word spacing are

features commonly addressed in assessment of handwriting


Methods and procedures of the study are presented in

this chapter. The chapter includes the following eight

sections: questions, subjects, setting, personnel,

dependent variable, independent variable, design, and

treatment of the data.


To address the purpose of the study, two research

questions were posed. The questions are stated below.

1. What are the effects on the level, variability, and

trend of letter formation scores with copying stimuli placed

at 0.5 m, 3 m, and 6 m distances from the subject?

2. What are the effects on the level, variability, and

trend of word spacing scores with copying stimuli placed at

0.5 m, 3 m, and 6 m distances from the subject?


Subjects included 13 elementary school children who

attended a public school and had been nominated for the

study because of poor handwriting performance in school.

Subject demographic data are presented in Table 1 and

include sex, age, race, intelligence quotient (IQ) when

available, first grade vision screening results, services

being received, handwriting grades, and reading and

mathematics achievement test scores. Vision was normal for

all subjects except for one subject with 20/100 vision who

was referred for an eye exam and one subject whose vision

was corrected with glasses. Of the 13 subjects, 3 received

services in a resource room for varying exceptionalities; 4

received extra help for academic problems in Chapter I

m, 1 -L" -o

ur m m %D N i n1 ..

mn PD- -%- Ch
D 4 .0 ( 0 0 IM 0% %0

a 1
4J 1



w -

* A




















P-- CO CO'


* -- %C w

w w zm w w w



CM 1



0 3m =3 3 3 3

r U4 e- N 0o -0
!,m .'

'.~ .-4 r
I I I Ir

PS 4%.

r m- 0 0 -



bo 1



u a

0 >

S l 0
0 4 4
C 0 0
C m 0
co o .o B

Ie o w oa
0 4U 14 l.1 o (
oJ4 0 0 Z

4 OM 144 C B

e o64 0 $4 t
r a e owni a
e U mo4 aa

.4 0C U H 0-40iU"
J.0 a. 4o-4oa -

< eu eCu a u4
> e 10 14 6 u

o Cc oA .ao" u0
CC E-0 .i CW 0 a -4

e0g i oa 4 oe L .
u>* a, 1.' 60 0 14 a
P% 0sc asB 13

Sa,00 +:0 0
cM to a 40 1a 12
0 a .4 B 4 3c
14aM.=0 Jo-C 6
-4 0 14 CC 4 1, 41 -.s0 a, 4J1
U4 C4M 0 S 60B aJ 0 04
6 0 = wE-#+Wwr4-AV
w- i02 -- Mou~a

OD 00 0% 0% c0
+ +

t t

ffl U2


o n o o
D U, 0 0
c 4-r -4 c
o o o o
C-NN 4 C4


program; and 6 had been identified by the classroom teacher

as having difficulty with handwriting, but they were not

receiving additional services. The 3 subjects in the

resource room had been identified as having an emotionally

handicapping condition. This condition is defined as

"resulting in persistent and consistent maladaptive

behavior, which exists to a marked degree, which interferes

with the student's learning process and which may include

but is not limited to any of the following characteristics"

(Alachua School District, 1989, p. 57). Characteristics

include inadequate academic progress, difficulty in

interpersonal relationships, inappropriate behavior,

pervasive unhappiness or physical symptoms related to school

or personal difficulties (Alachua School District, 1989).

These children had received one semester of services in the

resource room during the previous school year. They were

identified by the varying exceptionalities teacher as having

handwriting problems.

The 4 subjects in the Chapter I program were identified

for this service because of their low math and reading

scores on the California Achievement Tests, a behavioral

checklist, current academic performance, and teacher

recommendation (School Board of Alachua County, 1989). The

Chapter I program is a federally-funded program that

provides a self-contained basic skills classroom for

children in grade 1 and a basic skills development

laboratory for children grades 2-5. Children receive 240

minutes of instruction per week that includes direct

instruction in reading and math and computer-assisted

instruction. Subjects in the Chapter I program were

identified for this study through teacher referral.

The remaining 6 subjects were identified by the four

second grade teachers as the children in the class having

the most difficulty with handwriting. Other than one

subject who was in a program for the gifted, these children

had never received, nor had been referred for, nor were

currently receiving any special services for academic or

other problems. The criteria for selecting students for the

program for the gifted include: intelligence quotient of

130 or above, significantly above average academic

achievement, teacher-identified characteristics and needs

that are not being met in the regular classroom (Alachua

School District, 1989).

Teacher referral was verified by administering a sample

nonword sentence to 22 children in one classroom that had

five children referred for the study. The stimulus sentence

was placed at the 6-meter distance. The children wrote

their names on the back of the paper and the writers were

not identified until all scoring was completed. Using

visual inspection, the investigator sorted the papers into

the good, fair and poor stack. There were eight samples in

the "poor" stack, accounting for 36% of the sample. The

investigator and two raters used the Error Analysis of

Children's Handwriting (EACH) to independently score the

letter formation and word spacing of the eight papers with

the poorest handwriting. In this group of eight children,

the four children who had been referred by the teacher had

the lowest scores with 92% agreement among the three raters.

The fifth child referred by the teacher was absent the day

the sample was taken. Because of illness, this child was

subsequently replaced in the study by a child who was not

originally referred to the study but who had the same score

as one of the four lowest scoring children. The teacher

strongly endorsed the inclusion of this child in the study

and noted that she thought that she should have referred him

when she referred the other five children. Thus, all four

of the children originally referred by the teacher had the

lowest scores on a handwriting sample scored independently

by the investigator and two raters.

Subjects were selected who were 7 or 8 years of age,

male, right handed for writing, had parental consent and

were not taking any medication. Subjects had not been

identified in the educational system as having motor,

sensory, or language disorders such as cerebral palsy,

blindness, deafness, or aphasia. One of the children wore

glasses to correct a refractive error, one had been referred

for an eye exam due to the vision screening score of 20/100.

The remaining subjects had passed the first grade visual

screening. All the subjects had at least one year of

instruction in the D'Nealian (Thurber, 1987) approach to

handwriting that is the handwriting method stipulated by

county school board policy (S. Hollenger & M. Buchanan,

personal communication, April 21, 1989).

Permission to conduct the study was obtained through

the University of Florida Institutional Review Board and the

Alachua County School Board. A parental permission form was

signed by the parents of the participating students.


The activities of the study were conducted at

Terwilliger Elementary School in the school library and on

the school cafeteria stage. These settings versus the

regular classroom setting were chosen in order to avoid the

distractions posed in a regular classroom and to prevent

interruption of regular classroom procedures. Sessions were

scheduled during the times when these settings were quiet.

Both settings were well-lighted and were devoid of wall

decorations. In the library, the room had three walls and

the open end of the room was open to the main library.

Portable screens were used to prevent the subject from

viewing activity in the library. In the cafeteria, curtains

surrounded the testing area on all sides. Easels supporting

the copying model were situated in front of the subject and

could be moved towards or away from the subject to provide

the desired stimulus distance. A child-sized table and

chair were properly fitted to the child's height, as

specified by Benbow (1988). The posture used for the study

included the child positioned squarely on the chair. With

the child's arm hanging straight down, the height of the

writing surface was two inches above the bend in the elbow.


Personnel involved in the study included four research

assistants, three raters and the investigator. The research

assistants and one of the raters were undergraduate students

who were enrolled in a course or had taken a course in

applied behavior analysis and had experience with timing

children's behavior. Assistants were trained to position

the stimulus in relationship to the subject and to collect

the handwriting samples. The purpose of the study was not

discussed with the assistants and the results of daily

handwriting sampling were not disclosed. Two raters were

doctoral students in special education with experience in

tests and measurements and who had completed required

doctoral courses in statistics. The investigator scored the

letter formation and one of the raters scored the word

spacing. The other two raters participated in the

establishment of interrater agreement prior to the outset of

the scoring process.

The raters were trained by the investigator to use the

EACH. On independent ratings of handwriting samples from 8

subjects in a regular classroom, the investigator and two

raters achieved 80% or better agreement on 88% of the

subjects for letter formation and for word spacing ratings

(see Table 2). More specifically, for letter formation,

interrater agreements were 90% or more on 18 of the 24

interrater agreements. Median interrater agreements for the

three raters were 84%, 93%, and 96%, and the agreements

ranged from 71% to 99%. For word spacing, the interrater

agreements were 90% or more on 13 of the 24 interrater

agreements. Median interrater agreements for the three

raters were 83%, 86%, and 88%, and agreements ranged from

50% to 100%.

Interrater agreements between the investigator and the

rater who scored the word spacing were observed for 10

subjects prior to the outset of the scoring of samples from

the subjects (see Table 2). Interrater agreements were 100%

on eight samples, and 80% and 75% on the remaining two

samples. Dependent Variables

A criterion-based handwriting checklist, Error Analysis

of Children's Handwriting (EACH), was devised for this study

and used to assess the subject's daily handwriting samples.

This checklist will be described in this section and is

shown in Appendix A.

Daily handwriting samples were assessed in two areas:

word spacing and letter form. Rate of correct responding

was used because rate is a sensitive measure of behavioral

changes and it provides a standard measure to compare

Table 2

Percentage of Agreement of Rater Pairs Prior to and During
the Study

Letter Formation Word Spacing

Rater Pairs

Sample 1&2 1&3 2&3 1&2 1&3 2&3 1&4

Prior to Study

1 96 92 88 88 100 88 100

2 96 90 94 75 50 67 80

3 95 84 80 100 100 100 100

4 71 94 67 88 100 88 100

5 99 96 97 83 100 83 100

6 99 90 90 90 100 100 100

7 99 92 93 86 86 100 100

8 94 90 96 100 100 100 100

9 100

10 75

During Study

1 95 100

2 83 100

3 98 100

behavior over several days (Tawney & Gast, 1984).

Handwriting legibility was assessed in two rates: number of

correctly spaced words per minute (WPM) and number of

correctly formed letters per minute (LPM).

Word spacing was assessed by measuring the space

between the words. If the space after the word was not less

than 6 mm or more than 12 mm, the word was scored 1 as not

having a spacing error. In the absence of any normative

data on spacing, a spacing criteria was determined by

examining D'Nealian workbooks (Thurber, 1987) and measuring

the spacing between words in models used for children's

copying practice. The rate of correctly spaced words was

calculated by dividing the number of correctly spaced words

(1 point each) in the sample by the 2 minutes for the


Letter form (Stott et al., 1985) was assessed by

examining each letter for five possible errors: reversal,

incomplete letter, last stroke of letter in the wrong place,

poor line quality, and incorrect alignment or size. Each

letter was scored 0 to 5 points. For each of the five

errors, 1 point was assigned if the letter were deemed to be

correct and free of that error and a zero was recorded for

each error that was found. Criteria for each of the five

errors are described below.

1. The letter was reversed, inverted, or rotated (such

as b/d, p/q, g/q, s/z, u/n, w/m).

2. The letter was incomplete (such as t or f not

crossed, i or j not dotted, e without an eye, q without a

tail); or the letter was incompletely closed with a gap

measuring more than 1 millimeter; or ascenders (b, d, f, h,

k, 1, t) or descenders (g, j, p, q) were too short.

3. The ending stroke of the letter was incorrect or

positioned in the wrong place such as an additional stroke

added to the letter m.

4. Line quality was poor in that the letter had bent,

shaky, irregular, broken, or jagged lines, crumpled curves

or collapsed loops.

5. Letter alignment and letter size were treated

together as one error type for measurement purposes in this

study. Petty et al. (1985, p. 301) stated: "Alignment and

size are closely integrated and should be studied together."

Letter alignment refers to the position of the letter in

relationship to the writing lines (Hammill & Larsen, 1983)

and is usually assessed by identifying the letters that are

above or below these lines (Petty et al., 1985). Size

refers to how well the letter is positioned between the

writing line (Hammill & Larsen, 1983) and is usually

assessed by determining how well the letter conforms to the

space between the lines (Petty et al., 1985). For example,

a lower case "a" should fit in the lower one-half of the

space between the lower and middle lines of the paper.

Since size and alignment are so closely related (Petty et

al., 1985) and are measured in the same way, that is, by

measuring the position of the letter or its parts in

relationship to the line, size and alignment were considered


For the measurement purposes of assigning a point value

to error criteria, the separate scoring of size and

alignment was not used in this study because to do so would

result in a double penalty for the same error. If size and

alignment were considered separate errors, yet size and

alignment were judged in relationship to the writing lines,

it would be difficult to determine a size versus an

alignment error. For example, all of the following errors

can be measured in relationship to the writing lines: an

"a" that is so large that it goes beyond the guidelines on

the writing paper ( I a ); an "a" that is so small that it

does not come to the guidelines ( a. ); or an "a" that

appears to be the correct size but is malaligned, that is,

placed below the baseline ( g ). For the "a" that is

placed below the baseline, an alignment error would be

determined by measuring the distance below the baseline.

Since distance below the baseline was used to determine an

alignment error, the rater would have no objective measure

of size and would have to make a judgment that the letter

would have been the correct size if it had been placed on

the line. Were the rater to measure the distance below the

baseline to determine a size error, the letter would by