Childrens use of drawing as a pre-writing strategy : (Research Note)

  • Published on

  • View

  • Download


  • Journal of Research in Reading, ISSN 0141-0423Volume 21, Issue 1, 1998, pp 6974

    Children's use of drawing as apre-writing strategy(Research Note)

    Edith NorrisUniversity of Texas at El Paso

    Kouider Mokhtari and Carla ReichardOklahoma State University


    In this study the writing products of 60 third-grade students, who drew beforewriting a story on a self-selected topic, was compared with the writing products of59 third-grade students who wrote without drawing. The students in the group whichdrew before writing tended to produce more words, more sentences and more ideaunits, and their overall writing performance was higher than the students who wrotewithout drawing. These findings were consistent for boys as well as girls. Impli-cations for writing research and instruction are discussed.


    Lutilisation de dessins par les enfants comme strategie de pre-ecriture

    On a compare dans cette etude les productions ecrites de 60 ele`ves de troisie`me anneequi avaient fait un dessin avant decrire une histoire sur un sujet libre avec lesproductions ecrites de 59 ele`ves de troisie`me annee qui avaient ecrit directementsans faire de dessin. Les ele`ves du groupe qui avaient dessine avant decrire ont eutendance a` produire plus de mots, plus de phrases et plus didees, et leur productionecrite en general a ete meilleure que celle des etudiants qui ont ecrit sans dessiner. Cesconclusions valent pour les garcons et pour les filles. On discute des implicationspour la recherche et pour lenseignement de lecriture.


    The relationship between drawing and writing has been discussed in relation tochildrens literacy development (Bissex, 1980; Calkins, 1986; Harste, Woodward and

    # United Kingdom Reading Association 1998. Published by Blackwell Publishers, 108 Cowley Road,Oxford OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA

  • Burke, 1984). Several researchers (e.g. Atwell, 1990; Graves, 1983; and Wilson andWilson, 1979) have written about the unique kinship of drawing and writing duringthe planning phase of childrens writing. The use of drawing and other art activities as a pre-writing strategy has recently been recommended by Tompkins andHoskisson (1991), especially with children who otherwise have problems expressingthemselves in written form. Olsons (1992) research and work with young studentssuggested numerous benefits of such an integration, and she has discovered thatchildrens vocabulary improved as much as their drawing skills when the twoprocesses were integrated. Olson also noted that the characters whom children havefirst brought to life in drawings are easier to develop in stories.There has been limited formal study of the role of drawing in the writing process of

    children in primary grades one through three. Two of the studies which investigatedsuch a relationship are unpublished and include Zalusky (1982), who analysed therelationship between drawing and writing in first grade children, and Skupa (1985),who did a similar study with second graders. The findings in both studies stress theimportance of drawing as a way of facilitating idea generation for writing. Thepresent study seeks to explore the influence of drawing on third grade studentswriting performance.



    The participants were 119 third-grade students from three elementary schools in asmall lower-middle class community in the midwestern US. School records indicatedthat the children shared similar linguistic, cultural, and socioeconomic backgrounds.None of the children was identified as having any specific learning problems orhandicapping conditions. All of the children had completed their first and secondgrade years in the same school district with comparable records of socioeconomicstatus, student achievement, and teacher competency. The six third-grade classroomsavailable in the school district were randomly assigned, three to each of the twotreatment conditions. As a result, sixty students were allocated to the ExperimentalGroup, which drew prior to writing stories, and fifty-nine students to the ControlGroup, which wrote without drawing.

    Data Collection

    The data collection consisted of an initial assessment of writing ability, drawing andwriting samples. The Test of Written Language-2 (TOWL-2, Form A) by Hammilland Larson (1988) was administered to check for possible pre-existing differencesin writing ability between the two groups. T-test analyses showed no significantdifferences (t=0.366, df=117), suggesting that the two groups began the study withsimilar levels of writing ability. However, TOWL-2 composite scores were used as acovariate with the aim of controlling for pre-existing differences in writing ability,and increasing statistical power (Keppel, 1991).Each student was asked to write three different stories during three sessions

    approximately one week apart. Several choices of popular story topics were given aswell as the option of choosing their own topic. Children in the control group were

    # United Kingdom Reading Association 1998


  • given thirty minutes to write their stories, following the presentation of suggestedtopics and instructions. Those in the experimental group were given thirty minutes todraw a picture about their chosen topic and write a story about that topic.


    Four dependent variables were selected as measures of the students writingperformance: number of words; number of sentences; number of idea units; andoverall story grade. An idea unit was defined as a focus of consciousness that islinguistically expressed in written form, the completion of which is often, but notalways, signalled by a period (UK: full stop) or other end mark (Chafe andDanielewicz, 1987; Gere and Abbott, 1985). The number of idea units for each storywas determined by a group of three judges, all of whom have graduate degrees andexperience as elementary school teachers. Any written selections for which differingnumbers of idea units were obtained were discussed by the judges until unanimousagreement was reached.The overall quality of the childrens writing was evaluated using a modified

    composition scale developed by Hughey, Wormuth, Hartfiel and Jacobs (1983). Thisscale weights content: 50%; organization: 30%; and mechanics: 20%. The scaledirects the judges attention to specific features of the writing product andsuggests relative point values for each feature. Each childs story was rated threetimes by the judges. The scores were averaged and the resulting data were usedfor purposes of analysis. The interrater reliability obtained from the three ratingswas 0.88.


    The data obtained were analysed using repeated measures ANCOVAs, with group(experimental and control) and gender as independent variables, story as a repeatedmeasures variable (each child wrote three stories), and TOWL-2 composite scoresas a covariate. The main variable of interest was group: did children who drewpictures before writing produce better stories as measured by number of words,sentences, ideas units and overall writing grades? Gender was added to check forany possible interactions. The story variable was used to obtain more data withoutgreatly increasing error variability. The desired outcome was that there would be nosignificant differences in performance on the three stories. The results are presentedin Tables 1 and 2.The results revealed significant differences between the experimental and the

    control groups with respect to each of the four dependent variables. As shown inTable 1, students who drew before writing wrote significantly longer and betterstories, on average, than those in the control group who wrote without drawing.These subjects wrote more words (M=113.81 vs 71.20), more sentences (M=10.02vs 7.05), produced more idea units (M=11.39 vs 7.65), and earned higher storygrades (M=69.34 vs 45.37) than did their counterparts in the control group.However, there were no significant interactions between gender and group, nor werethere any significant gender differences, for any of the dependent variables (see Table2). The story variable was not significant for any of the four dependent variables,

    # United Kingdom Reading Association 1998


  • indicating that children did not change their performance across the three differentstories they wrote, making all of the story data consistent and valid.


    Two important findings resulted from this study. First, significant differences werefound between the two groups on all the measures used. Students who drew before

    Table 1. Dependent variable means and standard deviations by group.

    Experimental Group Control Group

    M SD M SD

    Words 113.81 70.94 71.20 18.17

    Sentences 10.02 5.80 7.05 2.67

    Idea Units 11.39 6.39 7.65 2.95

    Overall Grade 69.34 16.77 45.37 18.17

    Table 2. F-values for group, gender, story, and gender by group interaction.

    Variables df F p


    Words 1 19.26 .0001*

    Sentences 1 12.98 .0005*

    Idea units 1 17.98 .0001*

    Overall Grade 1 60.04 .0001*


    Words 1 .96 .3302

    Sentences 1 3.88 .0513

    Idea Units 1 4.72 .0319

    Overall Grade 1 1.05 .3085

    Gender By Group

    Words 1 .06 .8116

    Sentences 1 .51 .4787

    Idea Units 1 .60 .4404

    Overall Grade 1 .10 .7495


    Words 2 .51 .5828

    Sentences 2 .17 .8418

    Idea Units 2 .27 .7669

    Overall Grade 2 .67 .5142

    * Significant at alpha= .0125.E Values adjusted by Huynh-Feldt epsilon correction.

    # United Kingdom Reading Association 1998


  • writing tended to produce more words, sentences, and idea units, and their overallwriting performance was higher. Such results indicate that the act of drawing priorto writing appeared to be beneficial to writing performance among third-gradechildren in the experimental group. Anecdotal evidence collected by the first authorduring the course of the study supports this point: the students who were allowed todraw first seemed to be much more enthusiastic about the visits from this researcherthan did the children who simply wrote stories without drawing. Groans were oftenheard in the experimental classrooms each time they were told the time to draw hadcome to an end, and it was time to stop writing.Second, the results were consistent for both boys and girls regardless of group

    membership. This was a welcome discovery, since in most elementary schools, boyswriting has usually been found to lag behind that of girls (Silberman, 1989). In fact,one of the teachers of some of the students in the experimental group expressedsurprise when she was told that all the boys in her class had participated willingly inthe writing portion of the study.The data indicate that drawing became an effective planning strategy for the

    students who appeared to rely on their drawings as a reference point to promptthem toward what should come next in their writing. These findings are encour-aging, especially to those elementary school teachers who are concerned about theirstudents writing skills. Integrating drawing and writing may be used as a way ofmotivating students to write and enjoy doing it. However, while it may be pre-sumptuous to state that drawing always should take place before writing occurs,perhaps it would be reasonable to suggest that drawing before writing could becomea valuable extension of the overall writing curriculum in third grade classrooms.


    ATWELL, N. (Ed.) (1990) Coming to know: Writing to learn in the intermediate grades. Portsmouth, NH:


    BISSEX, G. (1980) A child learns to write and read. Cambridge, MA: Harvard University Press.

    CALKINS, L. (1986) The art of teaching writing. Portsmouth, NH: Heinemann.

    CHAFE, W. and DANIELEWICZ, J. (1987) Properties of spoken and written language. In R. Horowitz and S.

    Samuels (Eds.) Comprehending oral and written language. New York: Academic Press.

    GERE, A. and ABBOTT, R. (1985) Talking about writing: the language of writing groups. Research in the

    Teaching of English, 19, 362385.

    GRAVES, D. (1983) We wont let them write. Language Arts, 55, 63540.

    HAMMILL, D. and LARSEN, S. (1988) Test of written language-2. Austin: Pro-Ed.

    HARSTE, J., WOODWARD, V. and BURKE, C. (1984) Language stories and literacy lessons. Portsmouth, NH:


    HUGHEY, J., WORMUTH, D., HARTFIEL, V. and JACOBS, H. (1983) Teaching ESL composition: Principles and

    techniques. Cambridge, England: Newbury House.

    KEPPEL, G. (1991) Design and Analysis. Englewood Cliffs, NJ: Prentice-Hall.

    OLSON, J. (1992) Envisioning writing: toward an integration of drawing and writing. Portsmouth, NH:


    SILBERMAN, A. (1989) Growing up writing: teaching our children to write, think, and learn. New York:

    Random House.

    SKUPA, J. (1985) An analysis of the relationship between drawing and idea production in writing for second

    grade children across three aims of discourse. Unpublished Ph.D. Dissertation, University of Texas at


    TOMPKINS, G. and HOSKISSON, K. (1991) Language arts: Content and teaching strategies. New York:


    # United Kingdom Reading Association 1998


  • WILSON, B. and WILSON, M. (1979) Childrens story drawings: Reinventing worlds. School Arts, 78, 611.

    ZALUSKY, V. (1982) Relationships between selected dimensions of writing and drawing in first grade childrens

    compositions. Unpublished Ph. D. Dissertation. Ohio State University.

    Address for correspondence: DR EDITH A. NORRIS, Department of Teacher Edu-cation, University of Texas at El Paso, 500 W. University Avenue, 601 EducationBuilding, El Paso, Texas 79968, USA.

    # United Kingdom Reading Association 1998