WRITING PASSIVITY: ITS IDENTIFICATION Writing assumes real importance in a secondary school context. Students are expected to communicate clearly their thoughts in writing, and use writing to explore and evaluate their thinking. As students progress through secondary school, writing is required increasingly for academic participation. Written work will form the basis for much of their assessment, with writing ability becoming a prerequisite for academic success.Across the curriculum, students will be required to produce epistemic text (Bryson and Scardamalia, 1991) that has as its objective, inquiry into particular topics with the intention of informing, explaining, analysing and persuading. In order to accommodate the writing of this text, many students reduce what is a knowledge transforming task to a simpler one, by adopting a knowledge telling procedure (Scardamalia and Bereiter, 1982). As a result, students "tell" what they know about the topic, while loosely observing the structural conventions of the genre, when it is a more sophisticated, problem-solving procedure that is required. Difficulties with knowledge transforming are closely related to problems with intentional cognition (Bereiter and Scardamalia, 1987), which for a writer involves the deliberate setting and pursuit of cognitive goals. Metacognitive competence relies on intentional cognition. Failure to employ intentional cognition results in epistemic text that does not match the teacher's expectations.Negative feedback from teachers can compound students' difficulties by affecting their attitude to writing. It is possible for students to develop a reluctance to produce epistemic text, along with a strong dislike for the writing process. There is also evidence to suggest that the incidence of dislike of writing increases steadily across the years of junior secondary schooling (Kearney, Note 1).Students' dislike of writing is often accompanied by negative, affective beliefs concerning themselves as writers. The integral role that affect plays in the writing process has received some attention ( Brand, 1987; McLeod, 1987). Similarly, the connection between inappropriate cognition and negative affect, as determinants of writing failure, has attracted interest (Daly, 1979; Thompson, 1981; Rose, 1985; Cleary, 1991). As a result, several constructs have been proposed and investigated. These include: "writing apprehension", "writing anxiety", and "writer's block".A related, though arguably distinct construct, is that of "writing passivity". The notion of "passivity" was first proposed by Johnston and Winograd (1985), but in connection with reading failure. I linked the concept of passivity with underachievement in writing, drawing on a similar literature base.Writing passivity is situationally-bound in that it concerns school- based, epistemic writing. I believe that it evolves through the complex interplay of affective and cognitive factors, causing a sense of helplessness in the writer. The prototypic features of writing passivity are: . dislike of writing . low self-esteem as a writer . limited expectations of success as a writer . high levels of anxiety associated with writing tasks . avoidance of written work, with frequent non-submission of written assignments . low frustration tolerance with writing tasks . tendency to attribute poor writing performanceto abilitydeficits . attribution of successful writing behaviour to external factors, such as task ease or teacher help . sporadic, limited, or no use of intentional cognition when writing The focus for this paper is the empirical development of a self- report instrument that will identify passive writers, and distinguish them from other subgroups of writers who report a dislike of writing. METHOD PARTICIPANTS Two groups constituted the development sample. The first group acted as an exploratory subsample while the second group assumed a confirmatory role. The exploratory subsample consisted of 363 students (184 female, 179 male). It was drawn from three secondary colleges and was represented by students from Years 8, 9, 10, 11 and 12. Participants in the confirmatory subsample were 161 students (75 female, 86 male), drawn from one school and representing Years 8, 9 and 10. Members of both samples were considered typical of the year group population in each of the schools. Typically, participants ranged across ability levels, cultural origins, and socio-economic backgrounds. MATERIALS A self-report measure consisting of 50 items was used. The items, dealing with attributes of writing passivity, were arranged in a Likert-scale format. Six responses were provided. These ranged from strong agreement to strong disagreement. An even number of response options was selected to preclude equivocation. Half of the items were positive for the construct of writing passivity, while the balance was negative, in order to counter students' oppositional or acquiescent tendencies. In addition, items were assembled randomly in an attempt to avoid response bias due to placement.The term "writing" was defined carefully for students. A deliberate attempt was made to ensure that the term was not confused with handwriting. Students were told that "writing" referred to the factual texts that they were required to write at school. Paragraphs that informed, explained, analysed or persuaded were provided to exemplify the meaning of the term. PROCEDURE Items were subjected to two systems of critical review. Firstly, teachers involved in postgraduate literacy programs reviewed the items for their relevance to the domain of interest, with the intention of maximising item appropriateness. In addition, they reviewed each of the items, along with instructions, for clarity and conciseness. The second review group consisted of students in upper primary and secondary schools. Their responses to the pilot instrument alerted the researcher to ambiguities and unnecessary wordiness.On the completion of pilot investigations, a 50-item instrument was administered to the exploratory subsample of students who were required to indicate the degree to which each statement applied to them. Reverse scoring operated in that items positive for writing passivity that were strongly agreed with, and those that were negative for writing passivity and strongly disagreed with, attracted a minimum score of one point at the time of coding. In contrast, items negative for writing passivity that were strongly agreed with, and those that were negative for writing passivity and strongly disagreed with, attracted the maximum score of six points.Two sets of analyses were used to make a determination at the exploratory phase. Firstly, data obtained from the exploratory subsample were submitted to exploratory factor analysis in order to determine whether meaningful factors would evolve and what, if any, reduction might be made to the item pool. Secondly, the resulting factors and reduced item pool were subjected to cluster analysis. This second form of analysis was used to confirm the solution resulting from factor analytic techniques. At the confirmatory phase, the resulting 15-item instrument was administered to the confirmatory subsample with the purpose of establishing the instrument's reliability with an independent sample. RESULTS AND DISCUSSION Factor Extraction The FACTOR procedure in the Statistical Package for Social Sciences (SPSS Inc, 1988) was used for factor extraction. As a first approximation, data from the exploratory subsample were submitted to a Principal Components analysis with varimax rotation to determine the number of components or factors underlying the item set. Examination of the scree plot revealed three factors that explained 47.2% of the variance. Each factor provided meaningful item groups with strong, unambiguous loadings. Factor Rotation Subsequent analysis used oblique rotation as this form of rotation provided a better representation of conceptual reality. I anticipated that there would be correlation amongst the three factors if the writing passivity profile was to be apparent. Use of orthogonal rotation would have artificially eliminated the predicted correlations. The OBLIMIN method for oblique rotation in the Statistical Package for Social Sciences (SPSS Inc, 1988) was used. Factor loadings of those items with loadings greater than .40 on the three factors are reported in Tables 1, 2 and 3. With the exception of Q50 in Factor 1 and Q9 in Factor 3, there were no crossloadings exceeding .40 on any item. (Insert Tables 1, 2 and 3 here) Factor Labelling The items for Factor 1 referred to the writer's self-system, and made reference to students' self-esteem, attributional beliefs, histories, expectations and fears as writers. This factor was labelled "perception of self as a writer".The items constituting Factor 2 related to cognitive writing behaviour that was intentional and persistent, rather than spontaneous and incidental. Reference was made to behaviours such as goal- setting, planning, monitoring and regulating. This second factor was labelled "intentional cognition".Items representing Factor 3 dealt with the writer's attitude to writing. In general, the items did not require the students to make judgements of themselves as writers. This third factor was labelled "attitude to writing".As predicted, correlation amongst the factors was confirmed. The relationships were as follows: the first factor, perception of self as a writer, correlated at .351 with the third factor, attitude to writing, and at .447 with the second factor, intentional cognition. The correlation between the second and third factors was lower, at .263. Reduction of Item Pool Six items were selected to represent each of the three factors, in an 18-item measure. These items are indicated in Tables 1, 2 and 3. Initially, items chosen were with moderate to high loadings that, in total, allowed a balance in terms of item polarity. This resulted in equal numbers of items that were positive and negative for the construct of writing passivity.The internal consistencies of each six-item subscale were reflected in the Cronbach alpha reliabilities which were as follows: perception of self as a writer, .89; attitude to writing, .91; and intentional cognition, .83. The same data obtained from the exploratory subsample were subjected to exploratory cluster analysis (the ICLUST procedure, see Revelle, 1978) with the 18 selected items providing a focus. A matrix of correlations amongst the 18 items was formed, with those items that were most highly correlated being clustered together. This produced a three-cluster solution with a very high degree of fit to the data. (Revelle, 1978, VSS goodness-of- fit criterion = .958; mean residual correlation = .042). The clusters themselves were moderately correlated (median r = .46) and matched with the three identified factors. As three items (Q43 in Cluster 1; Q8 in Cluster 2; and Q3 in Cluster 3) were not defined as members of any cluster, the exploratory cluster analysis was rerun excluding these items. Consequently, a three- cluster solution with an even higher degree of fit to the data was obtained (Revelle, 1978, VSS goodness-of-fit criterion = .975; mean residual correlation = .029). Thus, the 15-item was retained. This instrument is presented in Appendix A. Confirmatory Analysis It was important to replicate the scale's reliability using an independent sample to confirm that the results, so far, were not a one-time, chance occurrence. The final 15-item instrument was administered to a confirmatory sample consisting of 161 secondary students representing years 8, 9 and 10. This procedure involved specifying a priori, three sets of items as cluster scales. Confirmatory analysis allowed examination of both the hierarchical structure of each of the predefined clusters and the quality of the overall solution ( Revelle, 1978).The three- cluster solution was accepted with an extremely high degree of fit (VSS goodness-of-fit criterion = .974, mean residual correlation = 0.035). Statistics and loadings are provided in Tables 4, 5 and 6. These include: Cronbach's estimate of alpha, which provides an estimate of internal consistency; beta, which gives gives an estimate of the first factor saturation or the worst split-half reliability to be obtained; gamma, being an estimate of the total variance explained by the cluster; and r bar, being the average correlation of items within a cluster.(Insert Tables 4,5 and 6 here) Profiling Writing Passivity The Writing Passivity Measure is a 15-item instrument comprised of three subscales, with each subscale consisting of five items. A maximum score of 30 and a minimum score of five can be obtained on each subscale. Lower scores are indicative of a more problematic profile.Of particular interest is the subgroup of writing passivity. As shown in Figure 1, passive writers score poorly on each subscale. (Insert Figure 1 here)With the writing passivity profile there is an interaction between maladaptive, affective beliefs and inappropriate cognition to produce for the writer a sense of helplessness. The profile of the passive writer indicates metacognitive dysfunction.The Writing Passivity Measure distinguishes amongst secondary school writers in terms of their related attitudes, self-perceptions, and cognitive behaviours as writers. In addition to passive writers, other subgroups of writers who dislike writing can be identified. These profiles are provided in Appendix B.My future research will aim at describing the profiles for different writing groups. It may be that, in addition to being different in the profile, writers in such other groups will have different types of problems. If this is the case, there will be a need for interventions that are quite different from those proposed for writers who are passive. CONCLUDING COMMENTS The Writing Passivity Measure can help teachers to better understand the writing needs of students in secondary schools. Junior secondary school students, in particular, have been neglected in terms of writing research. Investigations of writing processes and writing instruction have tended to focus on university-level students, while studies of writers' contexts have looked primarily at preschool and primary-aged children (Durst, 1991).Writing in the early years of secondary schooling must be addressed. Often in the minds of parents, teachers, and students, these years are not considered as important as the senior years. However, they are formative years when students are being challenged to meet expectations of writing competency across a range of genres addressing a variety of topics. In addition, emphasis on the production of epistemic text will be constantly increasing.Junior secondary schooling is a time of transition for most student writers. On their arrival in secondary school, students will be required to interact with several teachers, with whom individual contact-time is limited. Different teachers will also assume different levels of ownership for the teaching of writing within their classrooms. The idea that writing was to be taught in primary school, and is now the domain of the English classroom, will be strongly inferred by many students. Furthermore, there will be a commonly-shared expectation amongst teachers that students, having entered secondary school, should be competent users of written language.In order for teachers to respond to students' transitional writing needs, they must first appreciate difficulties that are associated with the demands of producing epistemic text. Then, teachers must accept that metacognition is fundamental to the process of writing. Their clear understanding of metacognition is essential to the appropriate scaffolding of writing development.Understanding the construct of writing passivity can cause a clearer concept of metacognition. Particularly, it will enable teachers to challenge reductionist views of metacognition. The term "metacognition" was introduced by Flavell in the seventies. During the last twenty years the term has been popularized and commonly defined with a focus on two activities: knowledge of cognition and regulation of cognitive processes (Brown, 1984; Rowe, 1988; Chan, 1991). This focus ignores the contribution of affect, and I believe that, as a direct consequence, this type of definition has constrained the understanding of metacognition as a viable working model for teachers. In actual fact, Flavell himself described metacognition as follows:the concept could reasonably be broadened to include anything psychological....if one has cognition about one's own or someone else's emotions or motives it could be considered metacognitive (Flavell, 1987, p21)Teachers' perceptions of metacognition need to include both affective and motivational characteristics of thinking (Paris and Winograd, 1990: Borkowski, 1992). There is a very powerful relationship between affect and cognition; one that cannot be denied. This relationship must feature prominently in teachers' models of student writing development, as these models inform classroom instruction and determine the ways teachers interact with student writers. Table 1 Items Representing Factor 1 (Perception of Self as a Writer) with Factor Loadings for Data Obtained from the Exploratory Subsample~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~#* Q45 My written work is weak. (.86)^^#* Q2 I am not good at writing. (.85) Q44 I am good at writing. (.84)#* Q4 I have always had problems with writing. (.81) Q13 I am a good writer. (.78) Q34 I have always received good results for my written work. (.76)#* Q5 I only write well if someone helps me. (.74) # Q43 I expect good results for my written work. (.74) #* Q14 My ideas are poorly expressed when I write. (.72) Q25 Writing is not difficult for me. (.70) * Q28 When I hand in a piece of writing I know I'll do badly. (.68) Q22 I do well on challenging writing tasks. (.66) * Q37 My teachers don't like my written work. (.63) * Q26 I don't write as well as other people. (.62) * Q7 I never know how to improve a piece of writing. (.60) * Q49 No matter what I do, I will never get better at writing. (.58) Q36 I can clearly express my ideas in writing. (.57) Q50 In my final draft my ideas are well-organised. (.53) * Q48 I find it hard to judge whether my writing is good or bad. (.47) Q33 When I put in effort, I write well. (.46) Q23 My writing ability will improve each year. (.41)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ * Reverse scoring used for these items.^^ The numbers in parentheses are factor loadings. # Items selected to represent the subscale in an 18-item instrument. Table 2 Items Representing Factor 2 (Intentional Cognition) with Factor Loadings for Data Obtained from the Exploratory Subsample~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~# Q32 I plan carefully before writing an essay. (.82)^^ Q1 I usually plan before I write. (.73)# Q20 In my final draft I check that my ideas are always organised. (.69)# Q46 I will keep drafting until I'm satisfied with my writing. (.64)#* Q30 I don't ever think of ways to improve my writing. (.64)# Q8 When I spot problems in my writing I try to fix them. (.63) Q40 When writing I think a lot about what I'm doing. (.59)# Q41 I always check that what I write makes sense. (.57) Q10 If I work hard I will become a better writer. (.53) * Q18 I don't bother to spend a lot of time on a piece of writing. (.50) * Q19 I just start writing and hope for the best. (.50) Q21 I always submit drafts if my teacher requests them. (.46) * Q39 I like to write my ideas as I think of them, rather than plan them. (.41)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ * Reverse scoring used for these items.^^ The numbers in parentheses are factor loadings.# Items selected to represent the subscale in an 18-item instrument. Table 3 Items Representing Factor 3 (Attitude to Writing) with Factor Loadings for Data Obtained from the Exploratory Subsample ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#* Q27 Writing is boring. (.76)^^# Q12 Writing is fun. (.75)# Q35 I enjoy writing. (.69)#* Q15 I don't like writing. (.66)# Q11 I enjoy subjects that involve a lot of writing. (.61) Q38 I expect to do better in subjects that don't require a lot of writing. (.51) Q9 I am interested in learning how to improve as a writer. (.46)#* Q3 I avoid writing. (.43) Q24 I like to write my ideas. (.41)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ * Reverse scoring used for these items^^ The numbers in parentheses are factor loadings# Items selected to represent the subscale in an 18-item instrument Table 4 Confirmatory Cluster Analysis Statistics and Loadings For Cluster 1 (Perception of Self As A Writer) For the Confirmatory Subsample Using a 15-Item Instrument~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~ Alpha=.84 Beta=.78 Gamma =.88 R Bar= .52 Item Loading Q45 My written work is weak. 0.83 Q4 I have always had problems with writing. 0.79 Q2 I am not good at writing. 0.69 Q5 I only write well if someone helps me. 0.67 Q14 My ideas are poorly expressed when I write. 0.63~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ T able 5 Confirmatory Cluster Analysis Statistics and Loadings For Cluster 2 (Intentional Cognition) For the Confirmatory Subsample Using a 15-Item Instrument ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Alpha=.79 Beta=.61 Gamma =.82 R Bar= .43Item LoadingQ32 I plan carefully before writing an essay. 0.75Q46 I will keep drafting until I am happy with my writing. 0.72Q20 In my final draft I check that my ideas are always organised. 0.69Q41 I always check that what I write makes sense. 0.68Q30 I don't ever think of ways to improve my writing. 0.48~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Table 6 Confirmatory Cluster Analysis Statistics and Loadings For Cluster 3 (Attitude to Writing) For the Confirmatory Subsample Using a 15-Item Instrument ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Alpha=.92 Beta=.85 Gamma =.93 R Bar= .70Item LoadingQ15 I don't like writing. 0.88Q12 Writing is fun. 0.87Q35 I enjoy writing. 0.85Q11 I enjoy subjects that involve a lot of writing. 0.85Q27 Writing is boring. 0.75~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Reference Notes 1. Kearney, J. F. Writing Passivity In A Secondary School Context. Paper presented at the Meanjin Reading Conference, Brisbane, March, 1993. References Bereiter, C., & Scardamalia, M. (1987). The Psychology of Written Composition. Hillsdale, N. J: Erlbaum. Borkowski, J.G. (1992). Metacognitive Theory: A Framework for Teaching Literacy, Writing, and Math Skills. Journal of Learning Disabilities, 25(4), 253-257. Brand, A.G. (1987). Hot Cognition: Emotions And Writing Behaviour. Journal of Advanced Composition, Volume VI, 5-15. Brown, A. (1987). Metacognition, Executive Control, Self- Regulation and Other More Mysterious Mechanisms. In F. E. Weinhert (Ed.) Metacognition, motivation and understanding. N.J: Hillsdale, 65-117. Bryson & Scardamalia, M. (1991). Teaching Writing To Students At Risk for Academic Failure. ERIC Document. ED 338725. Chan, L. (1991) Metacognition and Remedial Education. Australian Journal of Remedial Education, 23(1), 4-10. Cleary, L. M. Affect and Cognition in the Writing Processes of Eleventh Graders. Written Communication, 8(4), 473-507. Daly, J. (1979) Writing Apprehension in the Classroom: Teacher Role Expectancies of the Apprehensive Writer. Research in the Teaching of English. 13, 37-44. Durst, R. K. (1990) The Mongoose and the Rat in Composition Research: Insights from RTE Annotated Bibliography. College Composition and Communication. 41, 393-408. Flavell, J. H. (1987) Speculations About The Nature and Development of Metacognition. In F.E. Weinert (Ed.) Metacognition, motivation and understanding. Hillsdale. Johnston, P. H. and Winograd, P.N. (1985) Passive Failure in Reading. Journal of Reading Behavior. 17(4), 279-301. McLeod, S. H. (1987). Some Thoughts about Feelings: The Affective Domain and the Writing Process. College Composition and Communication, 38(4), 426-435. Paris, S. G. and Winograd, P. (1990). How Metacognition Can Promote Academic Learning and Instruction. In B.F. Jones and L. Idol (Eds.) Dimensions of Thinking and Cognitive Instruction. Hillsdale, N.J: Erlbaum. Revelle, W. (1978). ICLUST: A cluster analytic approach to exploratory and confirmatory scale construction. Behavior Research Methods & Instrumentation, 10, 739-742. Rose, M. Complexity, rigor, evolving method, and the puzzle of writer's block: Thoughts on composing-process research. In M.Rose (Ed.), When a Writer Can't Write. New York: Guilford. Rowe, H. (1988) Metacognitive skills: promises and problems. Australian Journal of Reading, 11(4), 227-237. Scardamalia, M and Bereiter, C. (1982) Assimilative Processes in Composition Planning. Educational Psychologist, 17(3), 165-171. SPSS Inc. (1988). SPSS-x User's Guide, 3rd edition. Chicago: SPSS Inc. Thompson, M. (1981) Writing Anxiety and Freshman Composition. ERIC Document. ED183235. APPENDIX A Writing passivity measure WRITING SURVEY You are being asked to respond to a series of statements about writing. There are no right or wrong answers to these statements. Would you please indicate the degree to which each statement applies to you. Do this by marking the response boxes. For example, if you agree slightly with the statement below you would indicate as follows: ****IMPORTANT***** The term "writing" refers to the writing of factual text and would include paragraphs or essays that inform, explain, analyse or persuade. It is the type of writing you do in school. It DOES NOT refer to handwriting.If you have no questions, please proceed to the next page. 1. Writing is boring. 2. I am not good at writing. 3. I will keep drafting until I'm satisfied with my writing. 4. I have always had problems with writing. 5. I only write well if someone helps me. 6. My written work is weak. 7. I always check that what I write makes sense. 8. I enjoy writing. 9. I don't ever think about ways to improve my writing. 10. I don't like writing. 11. I enjoy subjects that involve a lot of writing. 12. Writing is fun. 13. I plan carefully before writing an essay. 14. My ideas are poorly expressed when I write. 15. In my final draft I check that my ideas are always