Writing from Sources: The Effect of Explicit …web.macam.ac.il/~offprint/88896.pdfWRITING FROM...

RACHEL SEGEV-MILLER

WRITING FROM SOURCES: THE EFFECT OF EXPLICITINSTRUCTION ON COLLEGE STUDENTS’ PROCESSES AND

PRODUCTS �

ABSTRACT. Writing from sources, or “discourse synthesis”, is a common but cognitivelydemanding reading–writing task requiring students to select, organize, and connect contentfrom source texts as they compose their own new texts. The purpose of the present studywas to investigate the effect of explicit instruction of relevant strategies and assessmentcriteria on the subjects’ discourse synthesis processes and products. The subjects, 24in-service teachers enrolled in the researcher’s 1999–2000 course on “Reading and writingto learn”, were assigned the same discourse synthesis task – a review of the literature –at the beginning and at the end of the course. They were required to document their twoperformances of the task by means of a process log, to assess their pre and post instruc-tion processes and products, and to discuss the differences between these. The findingsobtained from a content analysis of the subjects’ process log summaries and criterial self-assessments of products indicated significant improvement in the subjects’ post instructiondiscourse synthesis processes and products.

KEY WORDS: discourse synthesis (or, writing-from-sources), explicit instruction, inter-textual processing strategies, knowledge-telling and knowledge-transforming, literacy,reading–writing connections, self-assessment, teacher education

1. INTRODUCTION

Writing from sources, or “discourse synthesis”,1 is a common academictask requiring students to select, organize, and connect content from sourcetexts as they compose their own new texts (Spivey, 1997). Discoursesynthesis is, then, a “hybrid task” (Bracewell, Frederiksen & Frederiksen,1982) of reading and writing, or of comprehension and production.

Another reading–writing task that has been extensively investigatedsince the early 1980s is the summary. Most studies of summarizing drawon the theoretical model of Kintsch and Van Dijk (1978), which postu-lates that comprehension involves the production of a mental summary,or “gist”, of the text by means of three major operations: deletion of

� The paper was first presented as an in-progress study at the IAIMTE conferencein Amsterdam, The Netherlands (July 11–13, 2001), and then at the SIG Writing 02conference at Staffordshire University, the UK (July 10–13, 2002).

1 The term more frequently used in US research publications.

L1 – Educational Studies in Language and Literature 4: 5–33, 2004.© 2004 Kluwer Academic Publishers. Printed in the Netherlands.

6 RACHEL SEGEV-MILLER

redundant propositions; substitution of a sequence of propositions by amore general one; and selection of the macroproposition of the text or theconstruction of a macroproposition when one is not explicitly stated.

The discourse synthesis task is, then, similar to the summary, butit is cognitively more demanding. When summarizing a text, studentsfrequently replicate its structure, thus producing a miniature isomorphicversion of the text (Spivey, 1997). When synthesizing, however, studentsare required to create their own macroproposition, or rather “superproposi-tion”, from different – sometimes even contradictory – macropropositionsof several source texts, and to organize these in a previously non-existentconceptual structure. Discourse synthesis, therefore, requires conceptualtransforming and the production of personal and creative perspectives onthe part of students (Schumacher & Gradwohl, 1991). Discourse synthesisis thus an act of literacy in line with the recent educational emphasis onthe development of academic discourse which

places special value on integrating one’s own ideas and knowledge into the written conver-sation with one’s sources (. . .) Such integration [is expected] as a move toward criticalliteracy and toward realizing writing’s epistemic potential to transform knowledge ratherthan to report knowledge (Flower, 1989: 26).

However, previous research indicated that performing discourse synthesistasks proved too demanding for most college students. Indeed, thosestrategies conceived of as crucial to the successful performance of thetask, namely strategies of conceptual transforming, were rarely used by thesubjects in these studies (Hartman, 1995; Kantz, 1989; McGinley, 1992;Segev-Miller, 1997; Stahl, Hynd, Britton, McNish & Bosquet, 1996).In the researcher’s earlier (1997) study, all the subjects reported diffi-culties with the task, although the successful synthesizers used conceptualtransforming and other intertextual processing strategies with significantlyhigher frequency and success than the unsuccessful ones. The subjectsaccounted for these difficulties in terms of lack of relevant knowledge (theyhad not been provided with explicit instruction of the task),2 and lack ofexperience (they had not previously been required to perform discoursesynthesis tasks).

These findings are in line with Geisler’s (1994), Lohman’s (1993),and Perkins’ (1991) criticism leveled against schools, which do notrequire authentic knowledge-transforming tasks, and do not emphasizethe relevant cognitive abilities which may develop by means of explicit

2 Data from interviews with the subjects’ instructors indicated a commonly mistakenassessment of the subjects’ ability to cope with the task as a given (see also Sternglass,1993; Zerger, 1992).

WRITING FROM SOURCES 7

instruction and practice. O’Flahavan, Hartman, and Pearson’s (1988)study, for example, indicated that school teachers’ questions more oftenrequire students to make intratextual rather than intertextual connections –although significant learning is believed to depend on the latter (IRA &NCTE, 1996) – possibly because professional materials (Lenski, 1998)and teacher training programs (Segev-Miller, 1990) contain very fewsuggestions for teachers to promote intertextual processing.

1.1. Review of the Literature

Studies in education (e.g., Guthrie, Van Meter, Hancock, Alao, Anderson& McCann, 1998; Hattie, Biggs & Purdie, 1996; Nist & Simpson, 1990)and writing (Ackerman, 1990; Williams & Colomb, 1993) indicated thatexplicit instruction – of metacognitive strategies in general and of cognitivestrategies relevant to the performance of a specific task in particular –and practice promoted students’ learning and their ability to transfer itto other learning contexts (Butterfield & Nelson, 1991; Butler & Winne,1995). Studies of reading (e.g., Garner, 1987), writing (e.g., Bereiter &Scradamalia, 1987), and summarizing processes (e.g., Brown, Campione& Day, 1982; Segev-Miller, 1989) have also indicated that the major differ-ences between successful and unsuccessful learners were the formers’more frequent and more successful use of metacognitive strategies.

Current definitions of metacognition emphasize both the learners’declarative and procedural knowledge of cognitive processes (Flavell,1977/1985) and their ability to assess and plan these processes (Baker &Brown, 1984). These include, respectively, the learners’ self-assessmentand self-regulation of cognition (Paris & Winograd, 1990; Paris, 2001).Current definitions of metacognition have been expanded to include othercharacteristics of successful learners such as motivation. Research indi-cated that motivation as well as self-efficacy , i.e., confidence in theirabilities to perform a task (Guthrie & Wigfield, 1997, 2000; Landis, 2002;Raedts, 2002; Zimmerman & Bandura, 1994), resulted in the subjectsfocusing on their performance (Miller-Cleary, 1991), taking personalresponsibility for their learning, and investing cognitive effort and timein their learning (Charney, Newman & Palmquist, 1996; Pressley, VanEtten, Yokoy, Freeben & Van Meter, 1998). Motivation, then, had an effecton the subjects’ cognitive engagement, or use of appropriate cognitiveand metacognitive strategies. Ertmer, Newby, and MacDougall (1996),for example, found that differences in motivation resulted in differentgoals and the use of different strategies, especially when coping withdemanding tasks: Those who adopted product goals used “surface-level


strategies”; those who adopted process goals, on the other hand, engagedin more significant processing of the information, were more persistentand reflective, and used effective cognitive and especially metacognitvestrategies.

Few discourse synthesis studies investigated the effect of explicitinstruction of the task on either process or product quality.3 However,instruction in these studies consisted solely of the strategy of mapping4 –graphically representing the conceptual structure of the text (e.g., problem-solution). Murray (1993) found that explicit instruction of mapping thesource texts had a significant effect on the processing of these texts, butnot on the discourse synthesis process or product. Schlumberger (1992)found that such instruction had a significant effect on the process – especi-ally with regard to the invention of a macroproposition – but not on theproduct. Risemberg (1993), on the other hand, found that explicit instruc-tion of mapping had a significant effect both on the process – especiallywith regard to rhetorical planning – and on the quality of the product.These equivocal findings may be accounted for in terms of differences ininstruction, such as mapping the source texts (Murray and Schlumberger),which requires the use of textual processing strategies, or mapping boththe source texts and the product text (Risemberg),5 which requires the useof intertextual processing strategies as well.

For explicit instruction to be effective, it should consist of more thanone strategy (i.e., mapping); it should consist of metacognitive strategiesand cognitive strategies relevant to the performance of the task (see:p. 3), in particular “research-validated” strategies (Nist & Simpson, 2000),that is, strategies which significantly characterized successful synthesizersin previous discourse synthesis research: (1) Metacognitive strategies:Assessing, planning, and revising (McGinley, 1992; Segev-Miller, 1997);and task representation (Kantz, 1989; Nelson, 1988; Stahl et al., 1996;

3 Product quality in discourse synthesis research (e.g., Spivey, 1990, 1997; Spivey &King, 1989) has usually been defined in terms of the selection, organization, and connec-tion of relevant information from the source texts. Process quality has been defined in termsof the use of the strategies relevant to the performance of the task.

4 Variously referred to in these studies as topical net (Murray, 1993), diagram(Schlumberger, 1992), or graphic organizer (Risemberg, 1993). For a review on theresearch of graphic or visual representations of texts see: Pearson and Fielding (1996).

5 Sarig (1991) found that the mapping strategy uniquely characterized her successfulsynthesizers – two literacy experts. The researcher’s earlier (1997) study indicated that themapping strategy characterized only some of her successful synthesizers; and that theseused the strategy to map only their discourse synthesis product text.


Sternglass, 1988); (2) Intertextual processing strategies: Conceptual,rhetorical, and linguistic transforming (Segev-Miller, 1997).6

Assessing strategies are especially relevant in the case of in-serviceteachers (as were the subjects of the present study), who are expectednot only to assess their students but also to help them assess themselvesin light of educational reforms encouraging students to gradually takemore responsibility for their own learning (e.g., Calfee, 1997; Pearson &Fielding, 1996). Although Falchikov and Goldfinch (2002: 288) arguedthat “the primary benefit of involving students in assessment resides inthe improvement in learning which can result”, hardly any of the studiesreviewed in their meta-analysis took account of the learning benefits ofself-assessment. Only two discourse synthesis studies7 investigated theissue of self-assessment: Ackerman (1989) and Sullivan and Hall (1997)found a relatively low rate of agreement (37% and 44%, respectively)between college students’ and their raters’ assessments, and students’ over-estimation of their performance. However, like the studies in the abovemeta-analysis, they focused on the assessment of products, and neitherinvolved explicit instruction of the assessment criteria.

1.2. Purpose of the Present Study

The purpose of the present study, then, was to elaborate on previousdiscourse synthesis exploratory studies which focused either on thesubjects’ ability to assess their products without any explicit instruction(Ackerman, 1989; Sullivan & Hall, 1997), or on the effect of the explicitinstruction of a single strategy (i.e., mapping) on the subjects’ productsand processes (Murray, 1993; Risemberg, 1993; Schlumberger, 1992), byproviding a more comprehensive framework of instruction including bothstrategies and assessment criteria relevant to the performance of the task.Underlying this purpose was also a concern for improving the subjects’learning and quality of instruction. The present study, therefore, set toanswer the following research questions:

(1) What were the differences between the subjects’ self-assessments oftheir pre and post instruction discourse synthesis processes?

(2) What were the differences between the subjects’ self-assessmentscores of their pre and post instruction discourse synthesis products?

6 These are actually categories of strategies, consisting each of several strategies. Theywill, however, be referred to in the present paper as strategies (for a list of these, seeAppendix A; for a complete list of discourse synthesis strategies see: Segev-Miller, 1997,in press).

7 Not included in the meta-analysis.


2. METHODOLOGY

2.1. Subjects

The subjects were 24 elementary education in-service teachers enrolled inthe researcher’s 1999–2000 course on “Reading and writing to learn”. Theclass met once a week for 2 academic hours (90 minutes). The course wasoffered within the framework of a two-year B.Ed. program for experiencedteachers who graduated from college before it was entitled in 1985 by theCouncil of Higher Education to grant academic degrees. Data pertainingto the subjects’ (1) background (major, teaching experience, etc.), and (2)self-assessment of reading and writing proficiency, were collected over thefirst two weeks of the course by means of two questionnaires.

2.2. Task

The subjects were assigned the same discourse synthesis task – a reviewof the literature – at the beginning and at the end of the academic year.They were required to process a collection of articles8 on the topic of therelationships between reading and writing (see Appendix B), and to usethese to write the review. The task in the present study differed from tasksin previous discourse synthesis studies in three major respects:

(1) Number and nature of source texts: In contrast with previous studieswhere the subjects were usually assigned 2–3 short informative (some-times contrived) texts, the source texts in this study were numerous,longer, and conceptually and rhetorically more complex authenticarticles.

(2) Authenticity of task: Brown, Collins, and Duguid (1989) argued thatsince all knowledge and learning are situated in context, explicitinstruction of writing must take place in an authentic learning context.McGinley and Tierney (1989) maintained that writing (with readingor other learning tasks) will not “invite” learning unless the learner isengaged with the task and topic. In the present study, the task wasexpected to be relevant and to engage the subjects: First, the topicof the collection of the source texts was also the title of the course.Second, the immediate purpose of the course and the task was toprepare the subjects for similar tasks in other courses. Finally, thesubjects were allowed, in fact encouraged, to use the review for theresearch requirements in their second year of studies.

(3) Time on task: Although intensive writing courses had a positiveeffect on students’ attitudes to writing (Hilgers, Bayer, Stitt-Bergh &

8 Translated into the subjects’ L1 – Hebrew.


Taniguchi, 1995), no similar effect on the quality of their processes andproducts was indicated (Jenson, 1992). Ackerman (1993: 359) arguedthat “time and practice are major variables in learning”. Similarly,Perfetti, Britt, and Georgi (1995: 161) argued that

Learning, including school-based learning, is often an event stretched out over timeand instructional events, in contrast to the brief times available for learning exper-iments (. . .) What is missing in short experimental studies of learning (. . .) is theopportunity for increased learning (. . .) Obviously, this is what ordinary learning isreally about: an extended opportunity to acquire information from multiple sources.9

Indeed, time on task was found to have a significant effect on the qualityof subjects’ written processes and products (e.g., Rijlaarsdam & Van denBergh, 1996; Segev-Miller, 1997). In contrast with previous studies, then,where explicit instruction of the discourse synthesis task took place over abrief period (usually 1–2 sessions), explicit instruction in the present studytook place over the second semester (15 sessions).10 The subjects wereassigned three weeks for the performance of each task.

2.3. Instruments

2.3.1. Process LogThe subjects were required to document their two performances of thetask by means of a process log. They were provided with instructionswith regard to conducting the process log: To document on a daily basisthoughts and actions related to the performance of the task, to be candidand elaborate, not to erase anything from the log, etc. The subjects werealso required to enclose all their by-products (marked photocopies andsummaries of the source texts, notes, and drafts) to the proccess log.

Very few discourse synthesis studies used the process log as a researchinstrument. The subjects in these studies were required to retrospectivelyreport in their process logs the strategies they employed in the process(Nelson, 1988; Segev-Miller, 1996, 1997; Sternglass, 1988), or the timethey devoted to performing the task (Greene, 1993).11 Greene and Higgins(1994: 118) pointed out that

Because retrospective accounts allow researchers to get a glimpse into writers’ strategiesand decisions after the fact, they have the advantage of allowing writers to explain andreflect on their decisions without interfering directly with their attention to the task, freeinga writer from the ‘cognitive load’ (Afflerbach & Johnston, 1984) that the concurrentverbalization of a think-aloud would require.

9 Italics not in original text.10 In the first semester explicit instruction focused on processing and summarizing

strategies of single texts in preparation for the discourse synthesis task.11 For a discussion of the reliability of the process log, see Segev-Miller (2000, 2004).


2.3.2. Assessment CriteriaThe subjects were also required to assess their products using a modi-fied version of an assessment model developed and validated elsewhere(Segev-Miller, 1990, 1991), and to verbally account for the scores on eachcriterion.

Falchikov and Boud (1989) suggested that explicitness of criteria andfamiliarity with criteria probably enhance student-teacher agreement, andthat students should be able to discuss assessment criteria with theirteachers and agree on them, to have a greater sense of “ownership” ofthese. Falchikov and Goldfinch (2002: 315) also suggested that “assess-ment using many individual dimensions seems more difficult than assess-ment using (. . .) few dimensions”. The subjects in the present study wereconsidered “partners” in the process and as such were in a position tonegotiate the number of the criteria (only 8 of the original list of criteriapresented in class) and their weights (equal points allotted, a total of 8points). These criteria pertain partly to the requirements of the discoursesynthesis task – e.g., criteria (1) and (2), which require conceptual trans-forming; and partly to the requirements of all academic tasks – e.g., criteria(3) and (5). The assessment criteria were:

(1) topic(2) macroproposition(3) elaboration(4) appropriate rhetorical structure(5) explicit cohesion(6) linguistic transformation(7) interpretation(8) citations

2.3.3. Process Log SummaryFinally, the subjects were required to compare (1) their two processes orperformances of the task documented in their process logs; and (2) theirproducts, i.e., both their criterial scores and verbal assessments of theirproducts; and to discuss the differences between their pre and post instruc-tion processes and products in their process log summaries. The subjectswere required to be as comprehensive as possible in their discussion.

2.4. Procedures

2.4.1. Presentation of the RequirementsThe subjects were informed of the purpose of the study and of thesignificance of the process log. They were provided with oral and written


instructions with regard to conducting the process log (see p. 11). Thepresentation of the discourse synthesis task – a review of the literature –included an explanation of the function of the review and a sample reviewfrom an article on the course reading list.

2.4.2. Explicit InstructionStudies in writing (e.g., Brown, Collins & Duguid, 1989: 39) indicated thatteachers who adopted the approach of explicit instruction, or “cognitiveapprenticeship”

promote learning, first by making explicit their tacit knowledge or by modeling theirstrategies for students in authentic activity. Then, teachers and colleagues support students’attempts at doing the task. And finally they empower the students to continue independ-ently.

Instruction in the present study consisted accordingly of the followingmethods: Presentation and explanation of strategies (one at a time andaccumulatively); demonstration and practice (from guided to independent)of strategy use (mostly by means of think-aloud) while processing sourcetexts (from collections of 2–3 short texts to collections of numerous andlonger texts) and while writing a discourse synthesis text; weekly assign-ments (from completion of products begun in class to more independentones); and evaluation of products using the assessment criteria.

2.5. Data Analysis

In order to answer research question (1) with regard to the differencesbetween the subjects’ self-assessments of their pre and post instructionprocesses, a content analysis12 of the subjects’ process log summaries wascarried out for the purpose of identifying major assessment categories. Theprocess log summaries were systematically compared with the subjects’process logs, as well as with all their by-products and verbal assessments oftheir products, for the purpose of establishing the reliability of the processlog summaries. In order to answer research question (2) with regard to thesubjects’ self-assessments of their pre and post instruction products, t-testsfor the average differences between these assessments were carried out.The researcher also assessed the post instruction products13 for the purpose

12 Following Strauss and Corbin’s (1990) model, according to which “data are brokendown into discrete parts, closely examined, compared for similarities and difference”(p. 62). These units are further conceptualized (i.e., named) and grouped in categoriesand subcategories which emerge from the data.

13 The researcher’s scores on the subjects’ pre instruction products were so low a jointdecision was made to leave them out.


of calculating the agreement rate between hers and the subjects’ assess-ments. All 24 post instruction products were also assessed by anotherexpert researcher. Inter-rater agreement was almost 90%.

3. FINDINGS AND DISCUSSION

In this section the findings pertaining to each of the research questions arepresented and discussed.

3.1. Research Question (1) Was: What Were the Differences between theSubjects’ Self-Assessments of Their Pre and Post InstructionProcesses?

A content analysis of the subjects’ process-log summaries in which thesubjects discussed the differences between their pre and post instructionprocesses yielded three major categories of assessment: (1) knowledge;(2) motivation; and (3) self-reassessment.

3.1.1. KnowledgeThe subjects reported that by the time they set out to perform Task Btowards the end of the course, they had acquired both the declarativeand procedural knowledge relevant to the performance of the task. Thisknowledge related to their evolving task representation, the strategies theynow engaged in, and the assessment criteria they applied to evaluate andrevise Product B.

3.1.1.1. Task Representation. Task representation has been defined as

an interpretive process which translates the rhetorical situation – as a writer reads it – intothe act of composing. As such it is the major bridge which links the public context ofwriting with the private process of an individual writer (. . .). The task as students representit to themselves is, by definition, the one they perform, but that representation is subject tomany influences and may evolve in surprising ways during writing (Flower, 1987: 1–2).

A task representation of knowledge-telling characterized most of thesubjects in Process A. This finding is in line with studies of writingin schools (see pp. 6–7) and in academic institutions (e.g., Bereiter& Scardamalia, 1987; Bridgeman & Carlson, 1984), which indicated acommon representation of knowledge-telling rather than the more complexand cognitively demanding representation of knowledge-transforming.However, the subjects’ initial task representation had, by the time theywere assigned Task B, evolved into one of knowledge-transforming:


I’ve come a long way (. . .) Today I’m aware of the requirements ofthe task [Elinor].14

The second time I was more aware of the need to make connections[Goldie].

3.1.1.2. Strategies. The subjects referred to the strategies they acquired inthe course as “tools”:

At the beginning of the year I didn’t have the tools and the ways toprocess the texts: I read them. I marked important ideas, but I couldn’tsynthesize them [Esther].

The strategies that the subjects acquired were both the transformingstrategies relevant to the discourse synthesis task and the metacognitivestrategies of assessing and revising:

I felt that I was making progress from step to step: selecting, organiz-ing, connecting, inventing a topic, drafting, monitoring, assessingmyself all the time [Oprah].

The subjects found the strategy of mapping both the source texts and theproduct text, which required both textual and intertextual processing of thesource texts (cf. Risemberg, 1993), especially helpful:

The second time I knew what I was going to do I mapped each ofthe texts, compared the maps to establish connections among them, Icould actually “see” these connections with my eyes and put them inone map [Naomi].

Mapping also helped me find out the connections between the ideasin the texts I failed to see the first time [Ada].

Mapping helped the subjects to either deliberately “establish connections”or “to find out the connections between the ideas in the texts” as a resultof discovery. Discovery has usually been associated in the literaturewith the conception of writing as a means to promote students’ knowl-edge-transforming and learning (Galbraith, 1992, 1996, 1999). It hasoften been observed to be “sudden”, “unexpected”, or “surprising” [seeMcGinley (1992), Segev-Miller (1997), and Sternglass (1988), whosesuccessful synthesizers’ processes were determined to a large extent bysuch discoveries rather than by deliberate planning].

In the present study, however, the subjects attributed their discoveries tothe strategies they acquired, which helped them to “actually see” what they

14 Italics in this quote and in the following quotes not in original process logs. Thesubjects’ names have been replaced by pseudonyms.


“failed to see the first time”, a metaphor often used by the subjects to referto their discoveries. The subjects also used the phrase “(be) aware” often torefer to their evolving metacognitive strategies. This was in contrast withtheir first performance of the task, which they carried out “mechanically”and “automatically”. In addition to these, they often used the strategy ofself-questioning, another metacognitive strategy, which was not explicitlytaught, to cope with difficulties in their second performance of the task:

The first time I did it very mechanically (. . .) The second time I keptasking myself questions [Mira].

This finding is in line with recent studies (Harpaz & Lefstein, 2000; King,1995; Spires & Donely, 1998) which indicated that student-generatedquestions improved textual processing and comprehension.

3.1.1.3. Assessment Criteria. The subjects found the product assessmentcriteria as helpful as the process strategies and reported an interactionbetween these. For example, as a result of applying the assessment criteria,they used different revision strategies:

I feel I went through a significant process because the second time Idevoted more attention and thinking with the help of the criteria tothe way I wrote or rewrote [Hilary].

This time I had criteria I could use (. . .) unlike the first time whenI worked on my “gut feeling” (. . .). The second time I imagined adialog with a future reader of my text. At one point, for example,when I realized I hadn’t defined some basic concepts vital to under-standing the main idea, I revised my text accordingly [Naomi].

Significant revision has been defined as conceptual and rhetorical ratherthan as linguistic, i.e., revision at the word or sentence level (Fitzgerald,1987; Flower, Hayes, Carey, Schriver & Stratman, 1986). Unlike previousstudies of writing processes (e.g., Yagelsky, 1995) and of discoursesynthesis processes (e.g., Nelson, 1990), which indicated that studentsrarely engaged in significant revision strategies, the present study indicatedthat most of the subjects “rewrote” and “revised” their products signifi-cantly. These findings may be accounted for in terms of the effect of theexplicit instruction of assessment and revision strategies and of the assess-ment criteria. These findings may also be accounted for in terms of thecontext of the task: the long period over which the performance of the taskextended, and the motivation of the subjects (see section 3.1.2.) to investcognitive effort in a task they considered to be significant and relevant totheir learning. Falchikov and Goldfinch (2002) argued that hardly any of


the studies they reviewed took account of the learning benefits of self-assessment. The subjects in the present study, however, attributed a majorpart of their “significant learning process(es)” to their ability to assesstheir processes and products, and eventually to assess, or rather reassess,themselves (see section 3.1.3.).

3.1.2. MotivationHaving acquired the knowledge relevant to the performance of the task didnot make the task less demanding. For some of the subjects it was just theother way round, but they were confident in their ability to perform thetask successfully and motivated:

The second time wasn’t easy and maybe it was more difficult, but Iam confident in my abilities [Elinor].

I was more motivated to perform the second task, I was eager to applywhat we learnt in the course [Oprah].

Some reported having actually enjoyed performing the task:

I enjoyed this time because I knew what I was doing and how to do it[Goldie].

All the subjects reported that the topic of the collection of the source textsthey had to process was relevant to their learning:

I also liked the idea that the contents of the texts were relevant to whatwe were asked to do [Oprah].

It was only fitting that the contents of the texts on reading & writing-from-reading we read for the task related to the task [Jane].

The subjects’ motivation was realized in their management of the task andin their ability to transfer their learning from the course to other contexts.

3.1.2.1. Task Management.15 The subjects reported that in Process A theydid not manage the task successfully:

When I approached the task the first time I didn’t plan it in advance[Shirley].

The first time was spontaneous and emotional writing, somewhatliterary, guided by my associations or brainstorming like an amebaspreading in all directions, sometimes digressing from the main point(. . .) it may be more interesting but it is less efficient [Sarah].

15 More often referred to as self-regulation (see p. 7).


In Process B, however, the subjects invested more time and cognitive effortin assessing, planning, and revising multiple drafts:

I invested more time and work this time (. . .) putting in time andthinking, struggling with myself [Nelly].

I worked harder the second time and it took me more time: more thantwo weeks cf. with one day the first time (!); 9 drafts on paper (plusmore drafts on the computer) cf. with one draft the first time [Louise].

These findings are in line with the findings of studies (e.g., Guthrie &Wigfield, 1997, 2000), which indicated that confidence in their abilities,as well as motivation, had an effect on the subjects’ cognitive engagement.

3.1.2.2. Transfer. The subjects in the present study transferred theirlearning from the course to other contexts:

(1) Other courses, especially with regard to performing similar tasks:

I rewrote the review of a seminar paper I had to submit in anothercourse [Oprah].

When writing a seminar paper later this year for another course Iwas more aware of my responsibility for the communication withmy audience, my reader (. . .) when I read a text now I know whatI’m going to do to get the most out of it [Jane].

(2) Their teaching:

I teach 7th and 8th Special Education grades: I’m already planninghow to rewrite a course in RC for them based on what I learnt inthis course [Oprah].

In my teaching reading to Special Ed kids I see how what I learntserves me as great practical tools to promote my students’ readingand writing [Jane].

(3) Their personal lives with regard to acting as more critical readers andwriters:

The course contributed to my personal and professional life – I evenread the newspaper differently now [Mira].

I am also more critical of my own writing and others’ today – Iknow that not every text that is published is coherent [Caroline].

These findings are in line with the findings of studies (e.g., Butterfield& Nelson, 1994) which indicated the significant effect of metacognitivestrategies on subjects’ learning and on their ability to transfer it to other


learning contexts. The subjects in the present study also expressed theirbelief in their ability to transfer their learning to future learning contexts:

Now I have more tools at hand that will help me with any academictask [Tammy].

What I learnt will serve me in the future [Caroline].

3.1.3. Self-ReassessmentFor many of the subjects the course was not only a process of learning,but also a process of self-discovery, or revelation (Ford & Ford, 1992),especially with regard to their reassessment of themselves as learners:

I thought that on account of the positive feedback I used to get in thepast on my papers that I knew how to, but I realized that my writingwas pretty bad and that I improved tremendously in the course of thisyear [Oprah].

I always thought I was very good at writing, I was sure that I automati-cally did what needs to be done, but in this course I found out thingsabout myself I didn’t know before [Hilary].

The subjects assessed themselves at the beginning of the course as“good” to “very good” readers and writers based on their previous experi-ences both at school and at college, as data obtained from the secondquestionnaire indeed indicated. They discovered, however, that their self-assessments were inaccurate. The subjects’ evolving ability to assess, orrather re-assess, themselves more accurately is in line with current defini-tions of metacognition and with the findings of studies (e.g., Hattie et al.,1996), which indicated that students’ metacognitive strategies graduallyimproved over time with explicit instruction.

3.2. Research Question (2) Was: What Were the Differences between theSubjects’ Self-Assessment scores of Their Pre and Post InstructionProducts?

The average score for Product A for all the subjects together was 4.72(SD 0.91) of a total of 8 points (see p. 12), that is, 58.87%, or below thepassing grade of 60% at college; the average score for Product B was 7.13(SD 0.50), that is, 89.12%.

Figure 1 represents the subjects’ individual self-assessments of theirpre and post instruction discourse synthesis products (Product A and B,respectively). A t-test for the average differences between these scoresindicated that these differences were significant (t = 12.11; DF = 23; P= 0.00 (9.09137E-12)). The subjects, then, assessed their Product B to


Figure 1. Subjects’ self-assessments of pre- and post-instruction products.

be of higher quality than Product A. Judging by their assessments, then,explicit instruction of the performance of the discourse synthesis task had asignificant effect on their final products. These findings are in line with thesubjects’ process log summaries, particularly their comments attributingtheir learning to the instruction they received in the course:

I feel that I was able to cope with the task well on account of thecourse which was organized according to the theories – it offered thestudents scaffolds and opportunities to slowly try out on their ownand become independent [Oprah].

The second time as a result of the course and learning from one weekto the next, I knew how to approach the texts, what to look for first,I evaluated them in terms of the criteria. This helped me process thetexts and locate the writers’ arguments [Tammy].

The subjects also described their learning in the course:

There has been a development in my ability to process texts [Esther].

I feel that the whole year was building up towards the final task. It hasbeen a long and very significant process; from mapping one text tomapping multiple texts, multiple drafts (. . .) crystallizing the criteriaof evaluation which reflect the process I went through [Nelly].

The subjects conceived of their learning as a process that takes time. Theyalso noticed that instruction put into practice the learning theories theyread about in the course: It offered “guidelines”, “scaffolds and opportuni-ties”, “examples” and “practice” – all in all, the constituents of cognitive


Figure 2. Average differences of subjects’ criterial self-assessments.

apprenticeship, which explicit instruction in the course had been planned toinclude for the purpose of helping the subjects “to try out on their own andbecome independent”. These findings are in line with Ackerman’s (1993)and Perfetti et al.’s (1995) arguments, that time and practice are necessaryconditions for learning; and with the findings of studies (e.g., Rijlaarsdam& Van den Bergh, 1996; Segev-Miller, 1997), which indicated that time ontask had a significant effect on the quality of the subjects’ writing processesand products.

Figure 2 represents the average differences between the subjects’criterial self-assessments of their pre and post instruction products.According to these assessments, the subjects’ major instruction gainsrelated to the following criteria: Inventing a macroproposition (an averagedifference of 0.48 points), organizing the information previously selectedin an appropriate rhetorical structure, and linguistically transforming theinformation (an average difference of 0.32 points in each case). Altogether,these differences accounted for 46.47%, or almost half of the total differ-ence between the average scores on Products A and B. That is, the subjectsattributed a major part of the improvement in their final products to theintertextual processing strategies they acquired: Conceptual, rhetorical,and linguistic transforming, respectively.

3.2.1. Conceptual TransformingThe majority of the subjects assessed Product A to be unsuccessful withregard to the crucial requirement of the task to invent a macroproposition:

(1) 20% of the subjects with regard to the absence of a macroproposition:


I didn’t write a macroproposition for the first product [Mira].

There was no macroproposition for all the texts and therefore nofocus [Ada].

(2) 75% of the subjects with regard to an unsuccessfully formulated one:

Macroproposition 1 (. . .) does not represent all the texts I read noris it general enough [Elinor].

One of the subjects circumvented the difficulty by selecting the macropro-position of one source text and imposing it as a “superproposition” on theother source texts:

Macroproposition 1 is from one text and does not include all the othertexts; that is, it is not one central sentence [Goldie].

This strategy is similar to the one used by some of the subjects in theresearcher’s earlier (1997) study, and referred to by them as a “by default”strategy. Several subjects circumvented the difficulty by not relating to allthe source texts:

Then I looked for connections but found it difficult, esp. with one ofthe texts, so I left it out which turned out to be a mistake [Tammy].

Most subjects assessed their macropropositions in Product B to be animprovement also in terms of syntax or accessibility:

Macroproposition 1 looks more like a title [a noun phrase]. Macro-proposition 2 is syntactically a complete sentence: subject, predicate[Lily].

Macroproposition 1 is too short and presented abruptly (. . .) Macro-proposition 2 is preceded by old information, introduced by BUT andincludes the organizing elements of my text [Sarah].

3.2.2. Rhetorical TransformingIn Product A the subjects often replicated the original rhetorical structuresof the source texts:

What I found useful was the rhetorical structure of the original textswhich I used in my text [Michelle].

They also used other “by default” strategies such as “focusing on theauthors rather than on the ideas”, and consequently producing a text whichlooked like a “shopping list”, a metaphor many of the subjects used todescribe Product A:


The first product is pretty bad, each paragraph relating to a differentidea; all in all it looks like a shopping list – a collection of differentthings that different people said; the second product has a much betterrhetorical structure [Oprah].

Product 1 text is not well organized – leaps from one idea to anotherwith no overall structure; product 2 is very explicit [Shirley].

The subjects, then, assessed Product A to be unsuccessful with regard tothis criterion, with “no overall structure” and “no real synthesis”. Listing– a knowledge-telling strategy requiring no conceptual transforming – hasbeen found to be the most frequent structure subjects used in previousdiscourse synthesis studies. McCarthy Young and Leinhardt (1998), forexample, found that their subjects used listing in 85% of their products.

Improvement consisted in some cases of presenting the structure ofProduct A more explicitly in Product B:

The rhetorical structure of the first product is unclear (. . .), it isn’tpresented from the beginning and makes comprehension of the textdifficult; but in the second product the explicit rhetorical structure ofproblem-solution already appears in the advance organizer [Louise].

More often, improvement consisted of replacing the structure in ProductA with a more appropriate one in Product B. This time the subjects used avariety of rhetorical structures, such as problem-solution, cause and effect,comparison, and contrast. The subjects also reported using a rhetoricalstructure of their own rather than replicating those of the source texts:

There’s rhetorical restructuring, in that the rhetorical structure isdifferent from that of the source texts [Jane].

I see great improvement in my ability to reorganize the text. I mademy own changes in the original structures [Gillian].

3.2.3. Linguistic TransformingThe subjects reported that writing Product A was a matter of “simplyreading and marking”, “mechanically copying and quoting”. These find-ings are in line with other discourse synthesis studies which also foundthat subjects engaged in extensive copying (Horowitz, Kennedy & Cole,1997; Kennedy, 1985; Mani, Fyfe, Lewis & Mitchell, 1996; Nelson, 1988,1990; Raphael & Boyd, 1991). Greene (1993) found that his subjects oftenquoted as a substitute, rather than as support, for their own texts. Campbell(1990) argued that copying, paraphrasing, and summarizing reflecteddifferent “depths” of textual processing. Indeed, linguistic transformingproved very difficult to the subjects:


The most difficult part, however, was linguistic transformation: Icould not get away from the language of the texts, so I talked to myselfand tried to say what I read without looking at the texts [Sharon].

I know now that I have to sum up the texts, etc., rather than copy partsof texts (. . .) And I realized that to summarize a text you really needto understand it: select and reorganize. This has been very significantlearning for me [Caroline].

Significant linguistic transforming, then, like rhetorical transforming, wasclosely related to conceptual transforming and reflected “deep” textualprocessing. Significant linguistic transforming was also related to thesubjects’ “authorship” (see: Greene, 1995; Spivey, 1997) of their textualproducts:

I was surprised to find out that most of my product 1 was quotingfrom the texts, how text-based I was (. . .) the second time it was verydifficult to write in my own words but most of the text is mine this time[Goldie].

It was very important to me to write with my own words this time, topresent the authors’ ideas in my own way [Jane].

Overall, the subjects’ average assessment scores for Product B were higherthan the researcher’s: 7.13 (SD 0.50), or 89.12%, and 5.54 (SD 1.07), or69.25%, respectively. The agreement rate between the subjects’ and theresearcher’s scores was 41.66%. These findings seem to be in line withthe findings of Ackerman (1989) and Sullivan and Hall (1997), whichalso indicated a low rate of agreement between college students’ and theirraters’ assessments, and students’ overestimation of their performance.However, the criteria used in the two studies were fewer (five differentcriteria in each vs. eight in the present study), were hardly directly relevantto the discourse synthesis task (e.g., Sullivan and Hall’s “selectivity” wasan exception), and related to fewer dimensions (e.g., Ackerman’s mostlyto rhetorical planning). Falchikov and Goldfinch (2002) argued that assess-ment is more difficult when more dimensions are used, as was the case inthe present study. The low agreement rate in the present study, then, may beaccounted for in terms of the more demanding assessment criteria. Indeed,the subjects expressed the need for more practice of the assessment criteria:

It still does not come easy to me (. . .) I wish we had one more yearto expand our knowledge (. . .) especially with regard to the criteria[Esther].


4. CONCLUSION

The findings of the present study indicated that, according to the subjects’self-assessments, explicit instruction of the performance of the discoursesynthesis task had a significant effect both on their processes and ontheir products. First, the subjects attributed the improvement in their postinstruction processes to the effect of explicit instruction in terms of threemajor categories: (1) knowledge of task representation as knowledge-transforming, of metacognitive and cognitive strategies relevant to theperformance of the task, and of assessment criteria; (2) motivation andself-efficacy, which resulted in the subjects’ ability to manage the task andto transfer their learning from the course to other contexts; and (3) self-reassessment as learners. Second, the subjects assessed their post instruc-tion products to be of higher quality than their pre instruction products,especially with regard to three criteria: Inventing a macroproposition,organizing the information previously selected in an appropriate rhetoricalstructure, and linguistically transforming the information. These criteriacorresponded to the intertextual processing strategies emphasized in thecourse.

Although they made significant progress, the subjects found discoursesynthesis to be a highly demanding task, and concluded that the course was

only the beginning of a personal learning process [Jane].

I still feel that I have so much more to learn [Louise].

Sternglass’ (1993) suggestion that “supportive” guidance throughout theircollege years of study be offered to students, should, therefore, be adopted.Two further suggestions are implied by the findings of the present study:First, that explicit instruction be as authentic and relevant as possible. Thefailures of present introductory writing courses (e.g., Chenoweth, Hayes,Gripp, Littleton, Steinberg & Van Every, 1999) may be accounted for interms of the lack of “situated” or authentic instruction. It has already beenargued (Greene, 2001: 562) that “Whether disciplinary courses in writingprovide a coherent – read authentic – alternative to the space generalwriting skills courses currently occupy is an argument yet to be made”.Certainly, disciplinary courses which provide no writing instruction (e.g.,McCarthy Young & Leinhardt, 1998) or very little writing instruction (e.g.,Greene, 2001), with an emphasis on the product rather than on the process,have not been very helpful.

Second, that students would be required to conduct a process log inwhich they relate to their learning (Segev-Miller, 1996) and to their self-assessment of their task performance (Evans, 2001; O’Neill, 1998), and be


provided with feedback (i.e., assessment) on a regular basis. The processlog would also enable the lecturer to gain insight into the dynamic develop-ment of her students’ strategy use over time, provide her with informationwith regard to the students’ individual difficulties, and allow her to adapther instruction accordingly.

Underlying the present study was also the hope that the subjects – allin-service elementary education teachers – would adopt an intertextualapproach in their teaching, which, as mentioned earlier (see pp. 6–7), isnot prevalent at schools. Indeed, some of the subjects reported immediatetransfer of their learning to their teaching (see pp. 18–19). Future researchmay investigate the extent to which the subjects of the present study haveadopted an intertextual approach in their teaching. Since all significantlearning involves synthesis (see p. 7), it would be appropriate to start theinstruction of discourse synthesis tasks at an early age (Harvey, McAuliffe,Benson, Cameron, Kempton, Lusche, Miller, Schroeder & Weaver, 1996;Jongsma, 1991; Lipson, Valencia, Wixon & Peters, 1993; Moss, Leone& Dipillo, 1997; Oyler & Barry, 1996).16 This instruction would be inline with the curricular changes recently being introduced into elementaryschools with an integrative perspective on learning (e.g., Barab & Landa,1997).17

APPENDIX A

Intertextual Processing, or Transfroming, Strategies

2. Transforming2.1. Conceptual transforming2.1.1. detecting the conceptual connection among source texts2.1.2. formulating the conceptual connection2.1.2.1. creating (“inventing”) a superproposition2.1.2.2. using an existing proposition2.1.2.2.1. detecting the macroproposition in one text and imposing it as a super-proposition on other texts2.1.2.2.2. selecting a proposition that is not a macropoposition in one text andimposing it as a super-proposition on other texts2.1.2.2.3. selecting a proposition that is not a macropoposition but appears acrossall texts and imposing it as a superproposition on all texts

16 Other advocates of explicit instruction of synthesis at school with an emphasis onreading are Afflerbach and Vansledright (2001); Hartman and Hartman (1995); and Hynd(1999).

17 Similar changes have been suggested by the Israeli Ministry of Education (see Kaspi,1992).


2.1.3. categorizing2.1.3.1. creating categories previously non-existent in source texts2.1.3.2. collapsing existing categories2.2. Rhetorical transforming2.2.1. summarizing one text2.2.2. using one text as a frame to incorporate other texts2.2.3. listing texts by author2.2.4. decomposing and recomposing texts2.2.5. synthesizing2.3. Linguistic transforming2.3.1. speech acts2.3.2. lexical repetition

APPENDIX B

Collection of Source Texts for the Discourse Synthesis Task

Folman, S. (1989). The benefits of interrelating reading and writing in the EFL classroom.Interface: Journal of Applied Linguistics, 3, 107–115.

Kucer, S.L. (1985). The making of meaning: Reading and writing as parallel processes.Written Communication, 2, 317–336.

Nelson, N. (1998). Reading and writing contextualized. In N. Nelson & R.C. Calfee (Eds.),The reading–writing connection. The 97th Yearbook of the National Society for the Studyof Education, Part II (pp. 266–285). Chicago, Illinois: The University of Chicago Press.

Tierney, R.J. & Pearson, P.D. (1983). Toward a composing model of reading. LanguageArts, 60, 568–578.

Widdowson, H.G. (1984). Reading and communication. In J.C. Alderson & A.H. Urquhart(Eds.), Reading in a foreign language (pp. 213–226). New York: Longman.

REFERENCES

Ackerman, J.M. (1989). Students’ self-analyses and judges’ perceptions: Where do theyagree? (Technical Report No. 23). The Center for the Study of Writing, Pittsburgh,Pennsylvania/Berkeley, California.

Ackerman, J.M. (1990). Translating context into action. In L. Flower, V. Stein, J.Ackerman, M. Kantz, K. McCormick & W. Peck (Eds.), Reading to write: Exploringa cognitive and social process (pp. 173–193). New York: Oxford University Press.

Ackerman, J.M. (1993). The promise of writing to learn. Written Communication, 10, 334–370.

Afflerbach, P. & Vansledright, B. (2001). Hath! Doth! What? Middle graders readinginnovative history text. Journal of Adolescent and Adult Literacy, 44(8), 696–707.

Baker, L. & Brown, A.L. (1984). Metacognitive skills in reading. In P.D. Pearson (Ed.), Ahandbook of reading research (pp. 353–394). New York: Longman.

Barab, S.A. & Landa, A. (1997). Designing effective interdisciplinary anchors. Educa-tional Leadership, 54(6), 52–55.


Bereiter, C. & Scradamalia, M. (1987). The psychology of written composition. Hillsdale,New Jersey: Erlbaum.

Bracewell, R.J., Frederiksen, C.H. & Frederiksen, J.E. (1982). Cognitive processes incomposing and comprehending. Educational Psychologist, 17, 146–174.

Bridgeman, B. & Carlson, S.B. (1984). A survey of academic writing tasks. WrittenCommunication, 1, 247–280.

Brown, A.L., Campione, J.C. & Day, J.D. (1982). Learning to learn: On training studentsto learn from texts. Educational Researcher, 10, 14–21.

Brown, J.S., Collins, A. & Duguid, P. (1989). Situated cognition and the culture of learning.Educational Researcher, 18(1), 32–41.

Butler, D.L. & Winne, P.H. (1995). Feedback and self-regulated learning: A theoreticalsynthesis. Review of Educational Research, 65(3), 245–282.

Butterfield, E.C. & Nelson, G.D. (1991). Promoting positive transfer of different kinds.Cognition and Instruction, 8(1), 69–102.

Calfee, R. (1997). Assessing development and learning over time. In J. Flood, S. BriceHeath & D. Lapp (Eds.), Handbook of research on teaching literacy through thecommunicative and visual arts (pp. 144–166). New York: Simon & Schuster Macmillan.

Campbell, C. (1990). Writing with others’ words: Using background reading text inacademic composition. In B. Kroll (Ed.), Second language writing: Research insightsfor the classroom (pp. 211–230). Cambridge: Cambridge University Press.

Charney, D.H., Newman, J.H. & Palmquist, M. (1996). “I’m just no good at writing”:Epistemological style and attitude toward writing. Written Communication, 12(3), 298–329.

Chenoweth, N.N., Hayes, J.R., Gripp, P., Littleton, E.B., Steinberg, E.R. & Van Every, D.A.(1999). Are our courses working? Measuring student learning. Written Communication,16(1), 29–50.

Evans, K.A. (2001). Rethinking self-assessment as a tool for response. Teaching Englishin the Two-Year College, 28(3), 293–301.

Falchikov, N. & Boud, D. (1989). Student self-assessment in higher education: A meta-analysis. Review of Educational Research, 59(4), 395–430.

Falchikov, N. & Goldfinch, J. (2002). Student peer assessment in higher education: A meta-analysis comparing peer and teacher marks. Review of Educational Research, 70(3),287–322.

Fitzgerald, J. (1987). Research on revision in writing. Review of Educational Research,57(4), 481–506.

Flavell, J.H. (1977/1985). Cognitive development, 2nd edn. Englewood Cliffs, New Jersey:Prentice-Hall.

Flower, L. (1987). The role of task representation in reading-to-write (Reading-to-WriteReport No. 2). The Center for the Study of Writing, Pittsburgh, Pennsylvania/Berkeley,California.

Flower, L. (1989). Negotiating academic discourse (Reading-to-Write Report No. 29). TheCenter for the Study of Writing, Pittsburgh, Pennsylvania/Berkeley, California.

Flower, L., Hayes, J.R., Carey, L., Schriver K. & Stratman, J. (1986). Detection, diagnosis,and the strategies of revision. College Communication and Composition, 37, 16–55.

Ford, M. & Ford, J. (1992). Writing as revelation. New York: Harper Collins.Fulwiler, T. & Young, A. (Eds.) (1990). Programs and methods for writing across the

curriculum. Upper Montclair, New Jersey: Boynton/Cook.Galbraith, D. (1992). Conditions for discovery through writing. Instructional Science, 21,

45–72.


Galbraith, D. (1996). Self-monitoring, discovery through writing and individual differ-ences in drafting strategy. In G. Rijlaarsdam, H. van den Bergh & M. Couzijn (Eds.),Theories, models, and methodology in writing research (pp. 121–141). Amsterdam:Amsterdam University Press.

Galbraith, D. (1999). Writing as a knowledge-constituting process. In M. Torrance &D. Galbraith (Eds.), Knowing what to write: Conceptual processes in text production(pp. 139–159). Amsterdam: Amsterdam University Press.

Garner, R. (1987). Metacognition and reading comprehension. Norwood, New Jersey:Ablex.

Geisler, C. (1994). Academic literacy and the nature of expertise: Reading, writing, andknowing in academic philosophy. Hillsdale, New Jersey: Lawrence Erlbaum.

Greene, S. (1993). The role of task in the development of academic thinking throughreading and writing in a college history course. Research in the Teaching of English,27(1), 46–75.

Greene, S. (1995). Making sense of my own ideas: The problems of authorship in abeginning writing class. Written Communication, 12, 186–218.

Greene, S. (2001). The question of authenticity: Teaching writing in a first-year collegehistory of science class. Research in the Teaching of English, 35, 525–569.

Greene, S. & Higgins, L. (1994). “Once upon a time”: The use of retrospective accounts inbuilding theory in composition. In P. Smagorinsky (Ed.), Speaking about writing: Reflec-tions on research methodology (pp. 115–140). Sage Series In Written Communication,Vol. 8. Thousand Oaks, California: Sage.

Guthrie, J.T., Van Meter, P., Hancock, G.R., Alao, S., Anderson, E. & McCann, A. (1998).Does concept-oriented reading instruction increase strategy use and conceptual learningfrom text? Journal of Educational Psychology, 90(2), 261–278.

Guthrie, J.T. & Wigfield, A. (1997). Reading engagement: Motivating readers throughintegrated instruction. Newark, Delaware: IRA.

Guthrie, J.T. & Wigfield, A. (2000). Engagement and motivation in reading. In M.L. Kamil,P.B. Mosenthal, P.D. Pearson & R. Barr (Eds.), Handbook of reading research, Vol. 3(pp. 403–422). New York: Longman.

Guthrie, J.T., Van Meter, P., Hancock, G.R., Alao, S., Anderson, E. & McCann, A. (1998).Does concept-oriented reading instruction increase strategy use and conceptual learningfrom text? Journal of Educational Psychology, 90(2), 261–278.

Harley, K. (1991). Contrasts in student and faculty perceptions of student writing ability.Paper presented at the Annual Meeting of the 42nd College Composition and Communi-cation Conference, Boston, Massachusetts, March 21–23.

Harpaz, Y. & Lefstein, A. (2000). Communities of thinking. Educational Leadership,58(3), 54–57.

Hartman, D.K. (1995). Eight readers reading: The intertextual links of proficient readersusing multiple passages. Reading Research Quarterly, 30(3), 520–561.

Hartman, J.A. & Hartman, D.K. (1995). Creating a classroom culture that promotesinquiry-oriented discussions: Reading and talking about multiple texts (Technical reportNo. 621). The Center for the Study of Reading, Urbana, Illinois.

Harvey, S, McAuliffe, S., Benson, L., Cameron, W., Kempton, S., Lusche, P., Miller, D.,Schroeder, J. & Weaver, J. (1996). Teacher-researchers study the process of synthesizingin six primary classrooms. Language Arts, 73, 564–574.

Hattie, J., Biggs, J. & Purdie, N. (1996). Effects of learning skills interventions on studentlearning: A meta-analysis. Review of Educational Research, 66(2), 99–136.


Hilgers, T.L., Bayer, A.S., Stitt-Bergh, M. & Taniguchi, M. (1995). Doing more than “thin-ning out the herd”: How eighty-two college seniors perceived writing-intensive classes.Research in the Teaching of English, 29(1), 59–87.

Horowitz, R., Kennedy, C.M. & Cole, B. (1997). Borrowing from sources: Children’slinguistic strategies when extrapolating multiple sources. Paper presented at the AERAConference, Chicago, Illinois.

Hynd, C.R. (1999). Teaching students to think critically using multiple texts in history.Journal of Adolescent and Adult Literacy, 42(6), 428–436.

IRA & NCTE (1996). Standards for the English language arts. Newark, Delaware:IRA/Urbana, Illinois: National Council of Teachers of English.

Jenson, R.M. (1992). Can growth in writing be accelerated? An assessment of regularand accelerated college composition courses. Research in the Teaching of English, 26,194–210.

Jongsma, K. (1991). Discourse synthesis (questions and answers). The Reading Teacher,44(9), 702–704.

Kantz, M. (1989). Written rhetorical syntheses: Processes and products (Technical reportNo. 1). The Center for the Study of Writing, Pittsburgh, Pennsylvania/Berkeley, Cali-fornia.

Kaspi, M. (Ed.) (1992). Integrated learning: Alternatives for a different organization of thedisciplines in elementary school (In Hebrew). Jerusalem: The Ministry of Education.

Kennedy, M. (1985). The composing process of college students writing from sources.Written Communication, 2, 434–456.

King, A. (1995). Cognitive strategies for learning from direct teaching. In E. Wood & V.Woloshyn (Eds.), Cognitive strategy instruction for middle and high schools (pp. 18–65).Cambridge, Massachusetts: Brookline Books.

Kintsch, W. & Van Dijk, T.A. (1978). Toward a model of discourse comprehension andproduction. Psychological Review, 85, 363–394.

Landis, D. (2002). Reading engaged readers: A cross-cultural perspective. Journal ofAdolescent and Adult Literacy, 45(6), 472–487.

Langer, E. (1993). A mindful education. Educational Psychology, 28(1), 43–50.Lenski, S.D. (1998). Intertextual intentions: Making connections across texts. Clearing

House (Nov/Dec) 72(2), 74–81.Lipson, M.Y., Valencia, S.W., Wixon, K.K. & Peters, C.W. (1993). Integration and thematic

teaching: Integration to improve teaching and learning. Language Arts, 70, 252–263.Lohman, D.F. (1993). Teaching and testing to develop fluid abilities. Educational

Researcher, 22(7), 12–21.Mani, J.E., Fyfe, R., Lewis, G. & Mitchell, E. (1996). Traversing the topical landscape:

Exploring students’ self-directed reading–writing-research processes. Reading ResearchQuarterly, 31(1), 12–35.

McCarthy Young, K. & Leinhardt, G. (1998). Writing from primary documents: A way ofknowing in history. Written Communication, 15(1), 25–68.

McGinley, W.J. (1992). The role of reading and writing while composing from sources.Reading Research Quarterly, 27(3), 226–249.

McGinley, W.J. & Tierney, R.J. (1989). Traversing the topical landscape: Reading andwriting as ways of knowing. Written Communication, 6, 243–269.

Miller-Cleary, L. (1991). Affect and cognition in the writing processes of 11th graders: Astudy of concentration and motivation. Written Communication, 8, 473–508.

Moss, B., Leone, S. & Dipillo, M.L. (1997). Exploring the literature of fact: Reading andwriting through information trade books. Language Arts, 74(6), 418–429.


Murray, S.A. (1993). The role of text structure in composing from sources (reading.writing, information integration). Dissertation, Stanford University, Stanford, California.

Nelson, J. (1988). Examining the practices that shape student writing: Two studies ofcollege freshmen writing across the disciplines. Dissertation, Carnegie Mellon Univer-sity, Pittsburgh, Pennsylvania.

Nelson, J. (1990). This was an easy assignment: Examining how students interpretacademic writing tasks. Research in the Teaching of English, 24(4), 362–396.

Nist, S.L. & Simpson, M.L. (2000). College studying. In M.L. Kamil, P.B. Mosenthal,P.D. Pearson & R. Barr (Eds.), Handbook of reading research, Vol. III (pp. 645–666).Mahwah, New Jersey: Erlbaum.

O’Flahavan, J.F., Hartman, D.K. & Pearson, P.D. (1988). Teacher questioning and feedbackpractices: A twenty-year perspective. In J.E. Readence & R.S. Baldwin (Eds.), Dialoguesin literacy research. 37th Yearbook of the National Reading Conference (pp. 183–208).Rochester, New York: National Reading Conference.

O’Neill, P. (1998). From the writing process to the responding sequence: Incorporatingself-assessment and reflection in the classroom. Teaching English in the Two-YearCollege, 26(1), 61–70.

Oyler, C. & Barry, A. (1996). Intertextual connections in read-aloud of information books.Language Arts, 73, 324–329.

Paris, S.G. (2001). Classroom applications of research on self-regulated learning. Educa-tional Psychologist, 36(2), 89–101.

Paris, S.G. & Winograd, P. (1990). How metacognition can promote academic learningand instruction. In B.F. Jones & L. Idol (Eds.), Dimensions of thinking and cognitiveinstruction (pp. 15–51). Hillsdale, New Jersey: Erlbaum.

Pearson, P.D. & Fielding , L. (1996). Comprehension instruction. In R. Barr, M. L. Kamil,P. Mosenthal & P.D. Pearson (Eds.), Handbook of reading research, Vol. 2, 2nd edn.(pp. 815–860). New York: Longman.

Perfetti, C.A., Britt, M.A. & Georgi, M.C. (1995). Text-based learning and reasoning:Studies in history. Hillsdale, New Jersey: Erlbaum.

Perkins, D.N. (1991). Educating for insight. Educational Leadership, 49(2), 4–8.Pressley, M., Van Etten, S., Yokoy, L., Freeben, G. & Van Meter, P. (1998). The metacog-

nition of college studentship: A grounded theory approach. In D.J. Hacker, J. Dunlosky& A.C. Graesser (Eds.), Metacognition in educational practice (pp. 347–366). Mahwah,New Jersey: Erlbaum.

Raedts, M. (2002). The role of self-efficacy beliefs and task knowledge in writing. Paperpresented at the 8th International Conference of the European Association for Researchon learning and Instruction Writing SIG, 10–13 July, Staffordshire University, Stafford,UK.

Raphael, T.E. & Boyd, F.B. (1991). Synthesizing information from multiple sources: Adescriptive study of elementary students’ perceptions and performance of discoursesynthesis (Series No. 45). The Center for the Learning and Teaching of ElementarySubjects, Michigan State University, East Lansing, Michigan.

Rijlaarsdam, G. & Van den Bergh, H. (1996). The dynamics of composing: An agendafor research into an interactive compensatory model of writing: Many questions, someanswers. In C.M. Levi & Ransdell, S. (Eds.), The science of writing: Theories, methods,individual differences, and application (pp. 107–125). Mahwah, New Jersey: Erlbaum.

Risemberg, R. (1993). Self-regulated strategies of organizing and information-seekingwhen writing expository text from sources. Dissertation, City University of New York.


Rosenshine, B., Meister, C. & Chapman, S. (1996). Teaching students to generate ques-tions: A review of the intervention studies. Review of Educational Research, 66(2),181–221.

Sarig, G. (1991). Learning-promoting strategies in composing a discourse synthesis.Paper presented at the Conference on College Composition and Communication, SanFrancisco, California.

Schlumberger, A.L. (1992). The effects of elaboration on community college students’execution of a reading–writing task. Dissertation, The University of Arizona.

Schumacher, G. & Gradwohl, J.M. (1991). Conceptualizing and measuring knowledgechange due to writing. Research in the Teaching of English, 25, 67–96.

Segev-Miller, R. (1989). Linguistic processing in EFL study-summarizing. Unpublishedmanuscript, The School of Education, The Hebrew University of Jerusalem, Jerusalem.

Segev-Miller, R. (1990). Survey of ALT (Academic Literacy Test) assessment of collegeseniors. Tel Aviv: Kibbutzim College of Education.

Segev-Miller, R. (1991). On the differences between holistic and categorical evaluations.Working paper. Kibbutzim College of Education.

Segev-Miller, R. (1996). Writing-to-learn: Conducting a process log. Paper presented atthe 30th International Conference of IATEFL, University of Keele, Stoke-on-Trent, UK.

Segev-Miller, R. (1997). Cognitive processes in discourse synthesis, Dissertation. TheHebrew University of Jerusalem, Jerusalem.

Segev-Miller, R. (2000). Retrospective and on-line verbal reporting as instruments ofresearch into reading and writing processes. The Annual UTELI (University Teachersof the English Language in Israel) Conference, 1 February, University of Haifa.

Segev-Miller, R. (2001). The effect of explicit instruction on college students’ writing-from-sources: products and processes. Paper presented at the 8th International Confer-ence of the European Association for Research on Learning and Instruction, Writing 02SIG, 10–13 July, Staffordshire University, Stafford, UK.

Segev-Miller, R. (2002). The effect of explicit instruction on college English majors’writing from sources: Processes and products. Paper presented at the Annual Confer-ence of Ayla – The Israeli Educational Research Association, 2–3 October, Bar-IlanUniversity, Ramat Gan.

Segev-Miller, R. (2004). Writing to learn: Conducting a process log. In G. Rijlaarsdam,H. van den Bergh & M. Couzijn (Eds), Effective teaching and learning of writing (pp.533–546). Dordrecht, The Netherlands: Kluwer Academic Publishers (in press).

Segev-Miller, R. (in press). Cognitive processes in discourse synthesis: The case ofintertextual processing strategies. In D. Galbraith, M. Torrance & L. van Waes (Eds),Recent developments in writing process research. Dordrecht, The Netherlands: KluwerAcademic Publishers.

Segev-Miller, R. (in progress). The long-term effects of explicit instruction of intertextualprocessing strategies: A follow-up.

Smith, M.C. & Stahl, N.A. (1993). The use of the reading diary for obtaining readingbehavior data among adults. In J.L. Johns (Ed.), Literacy: Celebration and challenge.Illinois Reading Council, Silver Anniversary Monographs (#ED 362844).

Spires, H.A. & Donely, J. (1998). Prior knowledge activation: Inducing engagement withinformational text. Journal of Educational Psychology, 90, 249–260.

Spivey, N.N. (1990). Transforming texts: Constructive processes in reading and writing.Written Communication, 7, 256–287.

Spivey, N.N. (1997). The constructivist metaphor: Reading, writing, and the making ofmeaning. New York: Academic Press.


Spivey, N.N. & King, J.R. (1989). Readers as writers composing from texts. ReadingResearch Quarterly, 24, 7–26.

Stahl, S., Hynd, C., Britton, B., McNish, M. & Bosquet, D. (1996). What happens whenstudents read multiple source documents in history? Reading Research Quarterly, 31,430–457.

Sternglass, M. (1988). The presence of thought: Introspective accounts of reading andwriting, Advances in Discourse Processes, vol. XXXIV. Norwood, New Jersey: Ablex.

Sternglass, M. (1993). Writing development as seen through longitudinal research: A casestudy exemplar. Written Communication, 10, 235–261.

Sullivan, K. & Hall, C. (1997). Introducing students to self-assessment. Assessment andEvaluation in Higher Education, 22(3), 289–305.

Williams, J.M. & Colomb, G.G. (1993). The case for explicit teaching: Why what youdon’t know won’t help you. Research in the Teaching of English, 27(3), 252–264.

Zerger, S. (1992). Literacy in college: The use of evidence in reading and writing of under-graduates in three disciplines (discipline specificity). Dissertation, University of Kansas,Lawrence, Kansas.

Zimmerman, B.J. & Bandura, A. (1994). Impact of self-regulatory influences on writingcourse attainment. American Educational Research Journal, 31(4), 845–862.

Yagelsky, R.P. (1995). The role of classroom context in the revision strategies of studentwriters. Research in the Teaching of English, 29, 216–238.

Kibbutzim College of EducationThe English Department and The Center of Academic Literacy149 Namir RoadTel Aviv 62507IsraelE-mail: [email protected]

Writing from Sources: The Effect of Explicit …web.macam.ac.il/~offprint/88896.pdfWRITING FROM...

Documents

Transcript of Writing from Sources: The Effect of Explicit …web.macam.ac.il/~offprint/88896.pdfWRITING FROM...