The Psycholinguistic Dimension in Second Language Writing: Opportunities Jor Research and Pedagogy Using Computer Keystroke Logging

The Psycholinguistic Dimension in Second Language Writing: Opportunities for Research and Pedagogy Using Computer Keystroke Logging KRISTYAN SPELMAN MILLER University of Reading Reading, England EVA LINDGREN, KIRK P H SULLIVAN Umeå University Umeå, Sweden This article discusses the use of computer logging as a means of investigating aspects of the second language (L2) writing process as writers are engaged in producing text at the keyboard The observation of writing by means of this method provides researchers with detailed information concerning aspects of the planning, formulation, and revision processes This function is illustrated by reference to a study in Sweden of school-age learners of English as an additional language whose written production was recorded as part of a longitudinal study, and findings from the study are presented The discussion highlights the potential uses of logging, not only not in relation to researching writers’ processes, but also as a pedagogic tool given that its replay facility allows access to information about aspects of the writers’ attention and strategies as they write INTRODUCTION t a time when writing practices are increasingly centered on interaction with keyboard and screen, the research opportunities offered by computer recording of the writer’s actions appear timely and relevant Observation has long been a means of exploring the writing process, and the advent of digital tools for tracking the writing event makes it possible to observe the process without the obvious intrusion of cameras or observer-researchers As a tool for gathering data on aspects of writing, computer recording (or keystroke logging) offers researchers an alternative to introspective methods such as think-aloud protocol and makes A TESOL QUARTERLY Vol 42, No 3, September 2008 433 accessible a mass of detailed information about the processes involved in producing text (Stevenson, 2005; Strömqvist, Holmqvist, Johansson, Karlsson, & Wengelin, 2006; Sullivan & Lindgren, 2006; van Waes, 1991; Wengelin, 2006) The aim of the present article is to evaluate the contribution of computer-logging research with reference to second language (L2) writers and to consider possible applications to language teaching We present data from a longitudinal study of young L2 writers of English in Sweden in order to illustrate the types of insight offered into pausing, fluency, and revision behaviour and to connect these findings to such issues as cognitive capacity, working memory, and automaticity The relevance of this tool within language pedagogy and its potential to prompt recall and self-reflection are then considered in relation to the individualisation of teaching and the promotion of self-assessment, metacognitive awareness, and learner autonomy PSYCHOLINGUISTICS AND WRITING RESEARCH Research into writing and particularly into L2 writing has not always defined itself closely in relation to mainstream psycholinguistic research, with the result that writing is sometimes seen as “a stepchild of psycholinguistics” (Bonin & Fayol, 1996, p 145) The rise in importance in the 1970s and 1980s of the process perspective on writing (e.g., Emig, 1971; Flower & Hayes, 1981; Perl, 1979; Sommers, 1980), however, gave prominence to the study of the cognitive processes underlying written text production and had significant impact on both research and pedagogy, and in particular L2 writing (Raimes, 1985; White & Arndt, 1991) The cognitive perspective established writing as involving a complex set of hierarchically arranged cognitive activities, involving problem-solving and decision-making within clearly defined goals (Kellogg, 1994, p.13) Flower and Hayes’s (1981) model identified three main composing processes—planning, formulating, and reviewing—which are “interactive, intermingling, and potentially simultaneous” (Grabe & Kaplan, 1996, p 91), with transition between these processes and their subprocesses occurring via a monitor Rather than being a linear sequence of processes, then, composing is recursive, with one process potentially triggering and triggered by another More recently, the concern in L2 writing with broader sociocultural and discoursal aspects of composing has distanced some writing researchers (Atkinson, 2003; Kent, 1999; Matsuda, 2003; Tobin, 1994; Trimbur, 1994) from the cognitivist stance in favour of a socially situated perspective on the writing event However, in this so-called postprocess 434 TESOL QUARTERLY era, some researchers also acknowledge that a cognitivist stance does not necessarily imply a reductive, abstract view of writing (Spelman Miller, 2005) Rather, researchers recognize that the internal and external contexts of composing are complementary and mutually reinforcing The early models of composing (e.g., Bereiter & Scardamalia, 1987; Flower & Hayes, 1981; Hayes, 1996), which have been so influential, identified the interacting processes of planning, translating (formulating), and reading/revising which the writer juggles in order to meet the goals and intentions of composing Not only these processes interact more or less simultaneously with one another during the activity of composing, but they also trigger one another (Rijlaarsdam, Couzijn, & Van den Bergh, 2004) For example, after reading parts of the text composed so far, inconsistencies can be noticed which lead to revision Reading can also trigger the discovery of new ideas or relationships in the text and likewise result in revision (Hayes, 2004) The changes made may in turn result in a need to plan the following section, so as to fit it in with the revised text, and the planning may in turn result in more revision Cognitive processing during writing thus forms a recursive pattern in which the different activities—planning, formulation, transcription, reading, and revising—interact These activities can leave traces in the writing process in the form of pauses, insertions, deletions, and movements Taken together, the numerous, detailed, and complex traces form a web of interacting events, constructed and constrained by one specific writer under specific conditions The pattern of cognitive processes created during writing may differ both between writers and within writers, depending on factors such as writing task (Severinson Eklundh, 1994), writing medium (Haas, 1996; Van Waes & Schellens, 2003), writing experience (Bereiter & Scardamalia, 1987), disability (Holmqvist, Johansson, Stromqvist, & Wengelin, 2002; Strömqvist, Ahlsén, & Wengelin, 1999; Wengelin, 2007), and language (first or additional) (New, 1999; Raimes, 1987; Thorson, 2000) In the following section, we briefly consider how we might access information about these cognitive processes, focussing on the insights offered by computer recording COMPUTER-BASED OBSERVATION A number of publications focussing on writing research methodology (Severinson Eklundh & Kollberg, 1996; Spelman Miller & Sullivan, 2006; Strömqvist & Ahlsén, 1999) have already articulated the main features of computer logging as a research tool In brief, this tool records writing sessions and stores in a log file detailed information about the time and occurrence of every keystroke, providing indirect evidence of cognitive THE PSYCHOLINGUISTIC DIMENSION IN SECOND LANGUAGE WRITING 435 FIGURE Logfile from the text by Writer 2, Year activity during online writing As an observational tool, keystroke logging shares some of the features of face-to-face observation techniques, such as those used by Matsuhashi (1981, 1987), and more recently by Bosher (1998), Chanquoy, Foulin, and Fayol (1996), Janssen, van Waes, and van den Bergh (1996), Phinney and Khouri (1993), and Schilperoord (1996), who all use video technology Unlike think-aloud protocols, which are widely used in cognitive research, the data are not generated by the writers themselves The use of self-generated data, although still widely used, is not without the problem of reactivity, that is, interference and distortion of the process itself, which can open such data to criticisms (see, e.g., Janssen, van Waes, & van den Bergh, 1996, for a discussion of this issue) Furthermore, the advent of computer recording as opposed to video or direct observation has provided a more versatile and unobtrusive means of data collection because the resident software records all keystroke presses (characters and actions such as deletions, cursor movements, spacebar presses, and so on) and yet appears as a normal text editor A variety of logging programmes are used in keystroke studies, including JEdit, used in the current study (IPLab, 1997; see Severinson Eklundh & Kollberg, 1996), Scriptlog (see Strömqvist & Ahlsén, 1999), and Inputlog (see Leijten & van Waes, 2006) The data file produced by the logging programme (see sample in Figure taken from our study1) presents information concerning the location and duration of pauses 436 See Appendix B for the textual output of this logfile TESOL QUARTERLY (normally identified as those nonwriting occurrences longer than seconds in duration, and indicated as a bracketed < > number), character and punctuation presses, and space bar and other moves including cursor movements and deletions Use of the space bar is indicated by the underscore symbol; the delete key is represented as a crossed block arrow, with the number of deleted keystrokes presented immediately preceding the symbol; the number of spaces moved with the arrow keys and the direction of the movement are also indicated numerically The logfiles contain a wealth of information concerning operations on the text, and this information can be used to replay the writing event revision by revision to elicit later reflections by the writer on his or her actions and intentions In this way, composition may be viewed within the broadest context, that is, in terms of the writer’s response to the demands of a particular task, including decisions concerning genre, audience, and topic development in addition to linguistic choices SECOND LANGUAGE WRITING: APPLICATIONS AND INSIGHTS One of the key issues for exploration is the way in which the writer juggles the complex web of interacting cognitive activities During composition, writers plan content and form, consider audience and style, reread and revise If writers are inexperienced and/or are writing in a foreign language, the complexity of the task increases As a consequence of their lack of automaticity, L2 writers can be forced to focus their attention and cognitive resources on specific aspects of their writing (Graham, Berninger, Abbot, Abbot, & Whitaker, 1997; Graham & Harris, 2000; McCutchen, 1996, 2000) Other aspects might then be left unattended to, and the quality of the final text could suffer Raised awareness of which processes interact during writing in the first language (L1) and in the L2, and how they interact, can assist writers in reducing cognitive load during writing and promote writing development One major difficulty writers have to struggle with during writing is that of distributing their limited cognitive resources between the various constraints of writing (McCutchen, 1996, 2000) Language, topic, content, goals, style, and genre are only a few of the items they need to consider while writing in order to achieve a good text (Chenoweth & Hayes, 2001; Hayes, 1996) In the case of the L2 writer, attention to lower level linguistic demands during writing may take up a large amount of the writer’s working memory resources As a result, less cognitive capacity is available for higher level processing of, for example, content, audience, and style (e.g., Schoonen et al., 2003; Whalen & Ménard, 1995) THE PSYCHOLINGUISTIC DIMENSION IN SECOND LANGUAGE WRITING 437 Insights into the L2 writer’s cognitive activity have been derived from a number of studies using keystroke logging These studies have focussed on a range of variables, including different task types (Severinson Eklundh, 1994; Spelman Miller, 2000a, 2000b), ages of writer (Pålson, 1998), and contexts (quasipedagogic and classroom) (Lindgren, 2004; Sullivan, Kollberg, & Pålson, 1998; Sullivan & Lindgren, 2002; Stevenson, 2005; Thorson, 2000) Keystroke logging has also been used in research on translation (Englund Dimitrova, 2006) Results from these studies have highlighted differences in pausing and formulating behaviour between first and L2 writers, which are consistent with Silva’s (1993) conclusion that L2 writers plan less at the global level (Dennett, 1985; Jones & Tetroe, 1987; Moragne e Silva, 1989; Yau, 1989), and display a slower rate of production and lower productivity (Hall, 1990; Hildenbrand, 1985) Whalen and Ménard (1995) found that writers in L1 (English) and L2 (French) planned and evaluated in their L2 more frequently at a lower linguistic level and less frequently at a higher (textual and pragmatic level) level Spelman Miller’s (2000b) findings concerning two groups of L1 and L2 student writers working on academic tasks in a British university context also confirm group-level differences in fluency, in particular, in the location and duration of microand macroplanning pauses and length of text span produced Planning pauses appeared to be more frequent at phrase-level locations, in particular at the point in the clause where the thematic element is introduced In L1 writers, by comparison, more significant pausing activity occurred at the ends of clauses; this behaviour underpins a difference in the approach to the demands of the task Because L2 writers have to juggle demands on their attention in relation to the linguistic and discourse aspects of the text, planning and revision by L2 writers tend to coincide at sites which are critical in foregrounding or backgrounding information and, therefore, in controlling the development of the topic In this way, the micro analyses reveal sites of conflict and decisionmaking within the construction of text, so facilitating “new insight into the interplay of cognition and context” (Flower, 1994, p 55) Other research into the fluency of L2 writers, both online and offline (Chenoweth & Hayes, 2001; Kellogg, 1994; Ransdell, Levy, & Kellogg, 2002; Schoonen et al., 2003; Whalen & Ménard, 1995; Stevenson, 2005), provides further support for these findings Fluency, defined as efficient access to linguistic knowledge and retrieval of linguistic form (van Gelderen & Oostdam, 2002), is a cognitive construct which presupposes the availability of sufficient linguistic knowledge to formulate ideas into linguistic form (Stevenson, 2005, p 136) When the writer has developed highly proceduralized rules for language production, retrieval can be more rapid, which frees up cognitive capacity from working memory By extension, the lack of fluency occurs when increased cognitive effort is 438 TESOL QUARTERLY devoted to one component process, such as formulation, with the consequence that a greater burden is placed on working memory for other component processes, such as handling higher level concepts In a study of L1 (Dutch) and L2 (English) writers, Stevenson (2005) considered the extent to which higher level processing in L2 writers was inhibited by the need to divert attention elsewhere Levels of fluency were ascertained using both offline and online measures (respectively, the number of words occurring in the final text divided by the total task time and the average number of intact recognisable words produced between pauses of seconds or more, regardless of whether the words occurred in the final product) Higher level processing was investigated in terms of time devoted to processing and the rhetorical structure of the written product Stevenson’s findings confirm that writers appeared somewhat inhibited in their higher level processing in English, with less attention devoted to conceptualising in English and the English text products containing fewer subarguments Extra attention to linguistic processes was identified through the greater number of clause-internal language revisions carried out at the point of inscription, more localised reading of the text, and the strategies to solve language problems Less attention was given to conceptualising in L2 than in L1, and the L2 texts were rhetorically less well developed than the L1 texts However, the study was not able to establish that the inhibition of higher level processing in L2 was related to lack of fluency at a linguistic level This study appears in general to support capacity accounts of writing (Kellogg, 1994; McCutchen, 1996), which assert that inhibition of conceptual processing occurs in L2 writing However, the specific aspects affected by a switch of attention to linguistic processes remain complex and call for further research THE CURRENT STUDY The study which we discuss in this article builds on the research reported in the previous section in considering the cognitive processes of L2 writers using keystroke logging The data we present are part of a larger study comparing the writing of a cohort of young writers composing in both L1 (Swedish) and L2 (English) over an extended period The focus for this particular report is on L2 writing performance over the 3-year data collection period It is concerned with pauses, fluency, and revision, and how they relate to text quality, as measured by a composite score of content, range, complexity, accuracy, and fluency This longitudinal focus adds a new dimension to the studies of L2 process research and allows us to address the following research questions: THE PSYCHOLINGUISTIC DIMENSION IN SECOND LANGUAGE WRITING 439 How does linguistic experience/year of study affect the online writing process in terms of revising, pausing, and fluency? Are there any online writing process variables which can be identified as contributing to text improvement? Participants Seventeen high school students took part in the study The participants were all in Year in the Swedish school system (that is, 14 years old) at the beginning of the study, and they all attended the same school in a Swedish town All were monolingual native speakers of Swedish who, at the start of the study, had been learning English for 3.5 years All the participants were volunteers and their parents or guardians had given written permission for their participation in the study The high school students came from two different school classes: the majority (n = 12) had been on a 50% English immersion program for one term at the start of the data collection; the remaining students (n = 5) studied all their subjects in Swedish The content of the curriculum was the same for all the students: All 17 received three 40-minute English language classes per week The immersion–nonimmersion learning context was originally made a variable, although our results indicate that it is not a significant factor All the students regularly used computers to assist in their schoolwork, including the preparation of project reports Naturally some of the students were more interested in computing and had their own personal computers at home, and some of the students used English more than others outside the classroom, for example, in writing to pen friends or in chatrooms However, as a group, we found the students participating in this study to represent the spectrum of students typically found in Swedish school Year The sample had a small amount of attrition, with 14 students completing all three tasks over the years Materials and Design For this part of the study, each participant wrote one essay in English each year over a 3-year period Following a briefing about the logging tool, the essays were composed in JEdit on an Apple Macintosh computer with an A4 screen, the international standard size of a sheet of paper The participants were taken one by one from their classroom to a small quiet room to undertake the task Although questions relating to JEdit could be posed at any point during the writing session, questions 440 TESOL QUARTERLY relating to English and/or the essay task were not answered The students were allowed to use pen and paper to jot down ideas The instructions given to the students were that they had around 60 minutes to write their essays (although in Year students had the chance to write for longer) and that they were to write approximately one A4 page of text They were explicitly told that they were to decide when their text was complete and that this could be longer or shorter than one A4 page, and could take more or less than 60 minutes (See Appendix A for the descriptive statistics of how long the writers’ wrote From Appendix A it is apparent that the change in instruction did not result in writers limiting their writing to 60 minutes.) The task was then read to the students and given to them in writing The tasks over the years were all descriptive In the first year of the study the writers’ task was to compose an article about a typical day in a Swedish school In the second year, participants completed a task describing holidays in Sweden, and in the third year, the students wrote an article describing their town and surroundings to a time-traveller Examples of the texts produced by one participant are given in Appendix B Coding and Analysis The research questions focus on the potential relationship between the year of writing and a set of variables, which characterize the writing process and the individual The data were coded and analyzed as follows First, information about the writers’ writing processes in terms of online text production, pausing, and revisions was extracted from the keystroke logfiles Second, the grades for the final texts were used as a text quality measure, and third, fluency measures were calculated from information in the logfiles Fluency was calculated on the basis of the definition in Chenoweth and Hayes’ (2001) study of fluency in L1 and L2 writing They define fluency as the number of words written per minute, and define an additional factor, burst, as the number of words written between pauses or revisions In computer-based automatic syntactic analysis, word is a difficult unit to define Therefore, we defined fluency as the number of characters produced per minute and burst as the number of typed characters between pauses and/or revisions: Burst = number of typed characters ÷ (number of revisions + number of pauses) We were also interested in writer fluency during burst This we calculated as Fluency during burst = (total writing time − total pause time) ÷ (number of revision + number of pauses) The number of characters included spaces as well as characters that were deleted from the final text version Pauses were THE PSYCHOLINGUISTIC DIMENSION IN SECOND LANGUAGE WRITING 441 defined as interruptions to keyboard and mouse activity of more than seconds, and revisions were defined as deletions or insertions Thus, in order to study the relationship between the writers’ text quality, year of writing, fluency, revision, and pausing, the following variables were used: text quality, year of study, fluency, burst, fluency during burst, text length, time on task, revisions (total number), pauses (total number), pause time (as a % of total writing time), pause length (average), and gender In terms of the offline assessment of the text quality, a grading system drawn from Jacobs, Zingraf, Wormuth, Hartfiel, and Hughey (1981) was used that identified the following weighted criteria: content (coverage of topic, detail, and appropriateness to genre) 20%; range (grammatical structures and vocabulary) 15%; complexity (sentence structure and paragraph organisation) 15%; accuracy (grammar, sentence structure and discourse organisation, spelling, punctuation) 30%; and fluency (feel for language, appropriateness, use of idioms) 20% For the purposes of this article, we used the composite score of these criteria Two native speakers with experience of teaching EFL undertook the grading The texts were graded blind and the order of grading was random The interrater reliability was confirmed with Cronbach’s ␣ (␣ = 0.95) Statistical Analysis Two statistical models were used to explore the data: repeated measures and linear regression Only the data from the 14 students who completed all three tasks were included Hence, 42 (14 students × tasks) sets of data points were used First, a one-way repeated measures ANOVA was used to examine the effect of year on the main variables text quality, fluency, burst, fluency during burst, text length, time on task, revisions, pauses, pause time, and pause length Gender was used as between-subjects variables The results are summarised in Table The results show that, in this study, the year of writing did not affect quality or the length of the written text, or the total number of revisions made during the writing session Gender had no effect on any of the tested variables However, the results show that year of writing had an effect on several other variables The writers increased their writing fluency (F (2, 12) = 8.72; p = 0.006) as well as the length of the writing bursts (F (2, 12) = 18.19; p < 0.001) The writers’ time on task decreased (F (2, 12) = 7.03; p = 0.005) (See Appendix A for time on task descriptive statistics.) The total number of pauses decreased (F (2, 20) = 24.48; p < 0.001) as did the total amount of time the writers’ paused during their writing session (F (2, 20) = 9.11; p = 0.002) To further explore which variables affected text quality, two linear 442 TESOL QUARTERLY TABLE Descriptive Statistics and Results of the Repeated Measures ANOVAs Variable Text quality Fluency Burst (characters) Fluency during burst (s) Text length Time on task Revisions Pauses Pause time Pause length M SD M SD M SD M SD M SD M SD M SD M SD M SD M SD Year Year Year 61.6 13.1 34.3 12.1 4.94 2.05 3.60 0.99 2433.1 1062.4 70.6 10.1 5.6 2.5 387.1 92.4 58.1 8.1 6.8 1.7 54.3 11.1 42.4 14.3 6.55 2.11 5.04 1.64 2307.1 846.8 54.9 13.3 5.4 1.9 246.5 95.5 47.8 8.9 7.1 1.8 65.1 16.9 55.4 19.5 9.38 3.43 5.82 1.70 2934.4 927.8 54.6 10.5 5.4 2.4 189.6 55.1 43.5 9.6 7.9 2.6 ANOVA F(2,10) = 2.07, p = 0.173 F(2,10) = 8.72, p = 0.006** F(2,10) = 18.20, p = 0.000** F(2,10) = 10.13, p = 0.004** F(2,10) = 2.03, p = 0.181 F(2,10) = 11.77, p = 0.002** F(2,10) = 0.44, p = 0.653 F(2,10) = 30.17, p = 0.000** F(2,10) = 17.41, p = 0.000** F(2,10) = 0.97, p = 0.411 Note M = mean; SD = significant difference Significance levels indicated as **p < 0.01 regression models were designed with a reduced number of variables Following the results of the ANOVA analyses, gender was excluded because it did not contribute significantly to the understanding of any of the tested variables Furthermore, because the variables fluency, burst, and fluency during burst were calculated using time, text length, and revision measures, collinearity was likely to occur Therefore, two models were created One model included only the automatically generated variables text quality, text length, time on task, revisions, pause time, and pause length The variable pauses was not included in this model because of high collinearity values (VIF = 9.3) and significant correlation (Pearson) with several other variables (pause time r = 0.61, pause length r = 0.43, and time on task r = 0.76) The second model included only the variables burst and fluency during burst, together with year as independent variables The effects of the variables in the two models are presented in Table When interpreting these tables, the power of the models needs to be considered The determination coefficient, r2, for Model is 58.60 and for Model is 55.60 We judge the explanatory power of the models to be reasonable, given the data set and number of variables Neither of the regression models distinguish between the different writers; a model that would account for individual writers would entail too many variables for this limited data set An examination of the results of regression Model shows that (a) the THE PSYCHOLINGUISTIC DIMENSION IN SECOND LANGUAGE WRITING 443 TABLE Results of the Regression Models Variable Constant Year Year Burst Fluency during burst Text length Time on task Revisions Pause time Pause length F-value R square Model Model Y = Text quality Y = Text quality Coefficient t-statistic 38.64 −4.65 1.69 2.89** −1.10 0.32 0.01 −0.38 −0.02 0.36 −0.30 7.88** 58.60 6.22** −2.44* −0.40 1.82 −0.38 Coefficient 50.48 −5.41 −2.68 4.19 −3.65 t-statistic 11.42** −1.49 −0.61 6.39** −3.20** 13.15** 55.60 Note *p < 0.05, **p < 0.01 longer the text, the better the grade, and (b) the less time spent on the task, the better the grade The time spent on the writing task to produce the same length and quality of text is reduced as writers’ writing skills become increasingly automated, and this result is reflected in both of the variables that were found to be significant in Model In Model 2, the fundamental variables were removed, and the variables burst and fluency during burst were added Both of these variables were found to be significant These variables showed that the writers achieving a higher mark wrote more than others during each writing burst and wrote faster DISCUSSION In line with Chenoweth and Hayes’ (2001) findings, the writers in this study increased both their fluency and the amount of text produced between interruptions (pauses and revisions) The writers also required fewer, but equally long, pauses and decreased their total pause time over the period Further, fluency emerged as a strong predictor of text quality Nevertheless, despite the increase in fluency the writers as a group did not increase their text quality This result is due to a considerable dip in text quality in the second year from an average of 61.6 to 54.3 (out of 100) In the third year, the quality increases again to an average of 65.1 Fluency also increased most between the second and third years It is important to bear in mind that these results not account for individual writers because that would have entailed too many variables for the available data set It is highly likely that individual writers behave 444 TESOL QUARTERLY and develop differently from the general picture presented in our results In order to illustrate the potential of keystroke logging for analysis of individual writers, we have summarised in Table the automatic and manual analyses of three texts produced by one writer, Jonas Jonas’ final texts are presented in Appendix B His marks varied from a text quality score of 57.5 in the first year, to 33.5 in the second, and 68.5 in the third From the final texts, and from the total number of typed characters, it is immediately apparent that his second-year text is shorter, conceptually less developed, and less complex than the other two General fluency is also lower in the second year, although his bursts in the first and second years are similar However, an increase in his fluency during burst can be seen from the first to the second year; and this second-year measure is similar to that for his third-year task Thus, in his second-year task, Jonas produced text between pauses and revisions more quickly than in his first-year task and at a similar rate to his thirdyear task His low general fluency measure indicates, however, that the process of formulating was time-consuming for him A manual analysis of Jonas’s revisions was undertaken in terms of form and concepts, using the LS taxonomy (Lindgren & Sullivan, 2006) Form revisions include changes that not alter the content, such as spelling, grammar, and punctuation, and conceptual revisions address content or stylistic aspects of the text This analysis revealed that the number of revisions and, in particular, the number of conceptual revisions relative to the amount of written text was higher during the writing of the second-year text than during the writing of the other two An automatic TABLE Summary of Jonas’s Writing Process Characteristics Characters Text quality Time on task Fluency (chars/min) Burst (characters) Fluency during burst (s) Pauses, number of Pause time, % of total writing time Pause length, average in seconds Revisions, total number Immediate revisions Distant revisions Form revision Conceptual revision Typographical revisions Year Year Year 1911 57.5 57 34 7.1 4.3 189 (17.5) 66 12 81 57 (2.98) 24 (1.26) 26 (1.36) 17 (0.88) 38 (1.99) 631 33.5 32 20 6.0 8.5 83 (13.2) 48 11.2 37 33 (5.23) (0.63) (1.11) (1.27) 22 (3.49) 2524 68,5 62 41 11.1 9.0 137 (5.43) 45 12.1 91 77 (3.05) 14 (0.55) 21 (0.83) 21 (0.83) 49 (1.94) Note Figures in parentheses refer to number of pauses or revisions per 100 typed characters THE PSYCHOLINGUISTIC DIMENSION IN SECOND LANGUAGE WRITING 445 analysis of the location of the revisions, run using Trace-it (IPlab, 1994), revealed that Jonas executed more immediate revisions—that is, revisions at the point of inscription—while writing the second-year text than while writing the other two Previous studies of revision at the point of inscription, sometimes called precontextual revisions (Lindgren & Sullivan, 2006), show that young writers revise at this point to adjust both form and content Taken together, the manual and automatic analyses of Jonas’s tests suggest that Jonas worked hard on the second-year text, although the markers’ judgment did not reflect his work, as shown in the text quality score Because it seems unlikely that L2 writing competence would decrease from the first to the second year, we may need to look to other possible explanatory factors in relation to this performance It is possible, for example, that affective factors such as motivation, anxiety, or engagement with the topic may be implicated The writers may have perceived the second-year task as more difficult, or they might have been experimenting with more challenging linguistic forms, resulting in lower text quality Further detailed analysis of language errors, coherence markers, and pause position would help to explore these possibilities It is also important to bear in mind when interpreting differences between years that nonidentical tasks were used for each year of the study This procedure complicates interpretation because year is in fact a year–task compound with effects (or lack of them) resulting from some combination of year, task, and year x task interaction The use of a single topic for all three writing tasks would not have removed this year–task compound because the writers would have been affected by their previous solutions to the task, and the recurrence of the task may have caused a decrease in motivation A future study with at least a three-fold increase in cohort size would both improve the power of the statistical analysis and permit each of the tasks to be written every year by a different group of students of the same size as in this study Comparison with an L1 composition would have permitted the teasing apart of those aspects of writing we have detected that are L2 specific and those patterns we have observed in our data that are not L2 specific but are mirrored in the L1 Even if our data cannot answer questions of L2 specificity, our data have, however, shown a range of process effects on quality and process changes over time and task for the L2 writer PEDAGOGICAL APPLICATIONS So far we have discussed the analysis of keystroke data as a means for researchers to understand how L2 writers approach text production and how their production processes develop over time However, this tool 446 TESOL QUARTERLY also has potential pedagogic uses, in particular in relation to developing learner autonomy and noticing the gap Keystroke logging offers a means for both teachers and students to gain insights into student writing For teachers, the automatic function analysis as presented in this article, offers a diagnostic instrument to help monitor students’ writing development For learners, both within the classroom and beyond, logging can provide a means for the writers themselves to increase awareness of the cognitive processes underlying their own use of language and their own writing performance Several recent studies have demonstrated how replaying the writing event can stimulate the writer to reflect on and evaluate his or her thoughts and actions that occurred while composing (Sullivan & Lindgren, 2002; Lindgren & Sullivan, 2003; Lindgren 2004) These studies show how writers’ individual cognitive make-up (see Whalen & Ménard, 1995) may trigger noticing of different features in the production process of which they had been previously unaware Sullivan and Lindgren (2006) argue that replay provides students “with appropriate input” through the use of the student’s own output (p 205) In the TESOL setting, such input could include aspects of English grammar, vocabulary, discourse, and style, which may be taken up in class and discussed A further application, termed peer-based intervention (Lindgren 2004; Lindgren, Sullivan, & Stevenson, 2008; Sullivan & Lindgren, 2006), involves the observation and in-depth discussion of the writing event with a peer, followed by a second draft stage For many students this process assists them in becoming aware “of the linguistic and extralinguistic features necessary for text improvement and language and writing development” (p 206) In these applications, keystroke logging is both a pedagogic tool, supporting individualised learning and whole-class teaching, and one which assists in the exploration of learning theory CONCLUSION The starting point for this article was the argument that L2 writing should be sited firmly in the domain of psycholinguistic research The development of theory and pedagogic practice depends on researchers and instructors gaining a clearer understanding of the complex cognitive activity underpinning the production of text within a defined sociocultural context Through computer logging unobtrusively recording the writing event, we have proposed an instrument for eliciting in-depth data concerning the planning, formulation, and revising processes of writers at a given time The flexibility of the tool enables the data collected to be stored and analysed and, through replay, used for reflection THE PSYCHOLINGUISTIC DIMENSION IN SECOND LANGUAGE WRITING 447 and interpretation Not only does this method offer the researcher an alternative to other methods of data collection; it also can be exploited as a pedagogic tool supporting learning development The detailed information available through computer logging offers a visualisation of the otherwise hidden processes of text production The fine-grained measurement of pausing and spans of text formulation as they emerge in the evolving text, and the identification of type and occurrence of revision activity provide means of exploring fluency and revision during online production As illustrated through the findings presented in this article, this method may allow consideration of interactions between aspects of the writing process, and comparison with other offline measures such as text quality In the case of the 3-year L2 study we presented, changes in fluency, amount of text produced, and pause and revision behavior are evidenced in the data, although an association with the quality of the final output was not clearly supported These results confirm earlier research concerning constraints on processing and the deployment of attentional resources (Chenoweth & Hayes, 2001; Stevenson, 2005), and set the agenda for further exploration of the interaction between cognitive processes, and factors such as task, context of writing, individual affect, and the product of writing The potential uses of keystroke logging, then, span the research, teaching, and learning contexts: for researchers, teachers, and individual writers Its flexibility in all these contexts has been demonstrated in a number of the studies discussed earlier, since it may be combined with other methods such as concurrent verbalisation or retrospective interview to enable the user to gain deeper insights into the reasoning behind the writer’s actions This study goes some way to counter arguments that the observation of cognitive processes of writing is abstract and reductive in nature or that it provides partial and indirect evidence that, though detailed, is removed from the context of production Rather, through combination with complementary means of data collection, keystroke logging allows us to bring together cognitive, sociocultural, and affective aspects of the writing event to explore how writers deal with the constraints of task, genre, style, and language in real time THE AUTHORS Kristyan Spelman Miller is a senior lecturer in applied linguistics and faculty director of teaching and learning for the Faculty of Arts and Humanities, University of Reading, England Her research interests are in written text analysis, L1 and L2 writing, and written communication within a digital environment Eva Lindgren is Head of the Dean’s Office, Faculty of Teacher Education, Umeå University, Sweden She holds a doctorate in English linguistics Her research inter448 TESOL QUARTERLY ests include L1 and L2 writing, keystroke logging, revision, self-assessment, writing development, and foreign language learning with a particular focus on young learners Kirk P H Sullivan is an associate professor in the Department of Language Studies, and a researcher in the Department of Child and Youth Education, Special Education and Counselling, Umeå University, Sweden His research interests include language teaching and learning, writing, and forensic linguistics REFERENCES Atkinson, D (2003) L2 writing in the post-process era: Introduction Journal of Second Language Writing, 12, 3–15 Bereiter, C., & Scardamalia, M (1987) The psychology of written composition Hillsdale, NJ: Erlbaum Bonin, P., & Fayol, M (1996) Producing isolated words from pictures, from orally and visually presented words: An on-line study of naming and writing In G Rijlaarsdam, H van den Bergh, & M Couzijn (Eds.), Current research in writing: Theories, models and methodology (pp 145–158) Amsterdam: Amsterdam University Press Bosher, S (1998) The composing processes of three Southeast Asian writers at the post-secondary level: An exploratory study Journal of Second Language Writing, 2, 205–241 Chanquoy, L., Foulin, J N., & Fayol, M (1996) Writing in adults: A real-time approach In G Rijlaarsdam, H van den Bergh, & M Couzijn (Eds.), Current research in writing: Theories, models and methodology (pp 36–43) Amsterdam: Amsterdam University Press Chenoweth, N A., & Hayes, J R (2001) Fluency in writing: Generating text in L1 and L2 Written Communication, 18, 80–98 Dennett, J (1985) Writing technical English: A comparison of the process of native English and native Japanese speakers (Doctoral dissertation, University of Colorado, 1985) Dissertation Abstracts International, 46, 3275A Emig, J (1971) The composing processes of twelfth graders Urbana, IL: National Council of Teachers of English Englund Dimitrova, B (2006) Segmentation of the writing process in translation: Experts vs novices In K P H Sullivan & E Lindgren (Eds.), Computer key-stroke logging: Methods and applications (pp 189–201) Oxford: Elsevier Flower, L (1994) The construction of negotiated meaning: A social cognitive theory of writing Carbondale: Southern Illinois University Press Flower, L., & Hayes, J (1981) A cognitive process theory of writing College Composition and Communication, 22, 365–387 Grabe, W., & Kaplan, R B (1996).Theory and practice of writing London: Longman Graham, S., Berninger, V W., Abbot, R D., Abbot, S P., & Whitaker, D (1997) Role of mechanics in composing of elementary school children: A new methodological approach Journal of Educational Psychology, 89, 170–182 Graham, S., & Harris, K R (2000) The role of self-regulation and transcription skills in writing and writing development Educational Psychologist, 35(1), 3–12 Haas, C (1996) Writing technology: Studies on the materiality of literacy Hillsdale, NJ: Erlbaum Hall, C (1990) Managing the complexity of revising across languages TESOL Quarterly, 24, 43–60 THE PSYCHOLINGUISTIC DIMENSION IN SECOND LANGUAGE WRITING 449 Hayes, J R (1996) A new framework for understanding cognition and affect in writing In C M Levy & S Ransdell (Eds.), The science of writing: Theories, methods, individual differences, and applications (pp 1–27) Mahwah, NJ: Erlbaum Hayes, J R (2004) What triggers revision? In L Allal, L Chanquoy, & P Largy (Eds.), Revision: Cognitive and instructional processes (pp 9–20) Boston: Kluwer Academic Hildenbrand, J (1985) Carmen: A case study of an ESL writer (Doctoral dissertation, Columbia University, 1985) Dissertation Abstracts International, 46, 3637A Holmqvist, K., Johansson, V., Stromqvist, S., & Wengelin, Å (2002) Studying reading and writing on-line In S Stromqvist (Ed.), The diversity of languages and language learning (pp 103–123) Lund, Sweden: Lund University, Centre for Languages and Literature IPLab (1997) JEdit (Version 0.9.3b20) [Computer software] Stockholm: Royal Institute of Technology, Department of Numerical Analysis and Computer Science Available from http://www.jedit.org/index.php Jacobs, H L., Zingraf, S A., Wormuth, D R., Hartfiel, V F., & Hughey, J B (1981) Testing ESL composition: A practical approach Rowley, MA: Newbury House Janssen, D., van Waes, L., & van den Bergh, H (1996) Effects of thinking aloud on writing processes In C M Levy & S Ransdell (Eds.), The science of writing (pp 233–250) Mahwah, NJ: Erlbaum Jones, S., & Tetroe, J (1987) Composing in a second language In A Matsuhashi (Ed.), Writing in real time (pp 34–57) Norwood, NJ: Ablex Kellogg, R T (1994) The psychology of writing Oxford: Oxford University Press Kellogg, R T (1996) A model of working memory in writing In C M Levy & S Ransdell (Eds.), The science of writing: Theories, methods, individual differences and applications (pp 57–72) Mahwah, NJ: Erlbaum Kent, T (Ed.) (1999) Post-process theory: Beyond the writing-process paradigm Carbondale, IL: Southern Illinois University Press Leijten, M., & van Waes, L (2006) Inputlog: New perspectives on the logging of on-line writing processes in a Windows environment In K P H Sullivan & E Lindgren (Eds.), Computer key-stroke logging: Methods and applications (pp 73–93) Oxford: Elsevier Lindgren, E (2004) The uptake of peer-intervention in the writing classroom In G Rijlaarsdam, H van den Bergh, & M Couzijn (Eds.), Effective teaching and learning of writing (pp 259–272) Dordrecht, The Netherlands: Kluwer Academic Lindgren, E., & Sullivan, K P H (2003) Stimulated recall as a trigger for increasing noticing and language awareness in the L2 writing classroom: A case study of two young female writers Language Awareness, 12, 172–186 Lindgren, E., & Sullivan, K P H (2006) Analysing on-line revision In K P H Sullivan & E Lindgren (Eds.), Computer key-stroke logging: Methods and applications (pp 157–188) Oxford: Elsevier Lindgren, E., Sullivan, K P H., & Stevenson, M (2008) Supporting the reflective language learner with computer keystroke logging In B Barber & F Zhang (Eds.), Handbook of research on computer enhanced language acquisition and learning (pp 189–204) Hershey, PA: IGI Global Matsuda, P K (2003) Process and post-process: A discursive history Journal of Second Language Writing, 12, 65–83 Matsuhashi, A (1981) Pausing and planning: The tempo of written discourse production Research in the Teaching of English, 15, 113–134 Matsuhashi, A (1987) Writing in real time: Modeling production processes Norwood, NJ: Ablex 450 TESOL QUARTERLY McCutchen, D (1996) A capacity theory of writing: Working memory in composition Educational Psychology Review, 8, 299–325 McCutchen, D (2000) Knowledge, processing, and working memory: Implications for a theory of writing Educational Psychologist, 35(1), 13–23 Moragne e Silva, M (1989) A study of composing in a first and second language Texas Papers in Foreign Language Education, 1, 132–151 New, E (1999) Computer-aided writing in French as a foreign language: A qualitative and quantitative look at the process of revision Modern Language Journal, 83, 80–97 Pålson, E (1998) Revision strategies in young foreign language writing (TRITA-NAP9802, IPLab-143 Report) Stockholm: Royal Institute of Technology, Department for Numerical Analysis and Computer Science, Interaction and Presentation Laboratory Perl, S (1979) The composing processes of unskilled college writers Research in the Teaching of English, 13, 317–336 Phinney, M., & Khouri, S (1993) Computers, revision, and ESL writers: The role of experience Journal of Second Language Writing, 2/3, 257–277 Raimes, A (1985) What unskilled ESL students as they write: A classroom study of composing TESOL Quarterly, 19, 229–258 Raimes, A (1987) Language proficiency, writing ability and composing strategies: A study of ESL college student writers Language Learning, 37, 439–468 Ransdell, S., Levy, C M., & Kellogg, R T (2002) The structure of writing processes as revealed by secondary task demands L1-Educational Studies in Language and Literature, 2(2), 141–163 Rijlaarsdam, G., Couzijn, M., & van den Bergh, H (2004) The study of revision as a writing process and as a learning-to-write process Two prospective research agendas In L Allal, L Chanquoy, & P Largy (Eds.), Revision: cognitive and instructional processes (pp 21–38) Dordrecht, The Netherlands: Kluwer Academic Schilperoord, J (1996) The distribution of pause time in written text production In G Rijlaarsdam, H van den Bergh, & M Couzijn (Eds.), Current research in writing: Theories, models and methodology (pp 21–35) Amsterdam: Amsterdam University Press Schoonen, R., van Gelderen, A., de Glopper, K., Hulstijn, J., Simis, A., Snellings, P., et al (2003) First language and second language writing: The role of linguistic knowledge, speed of processing and metacognitive knowledge Language Learning, 53, 165–202 Severinson Eklundh, K (1994) Linear and nonlinear strategies in computer-based writing Computers and Composition, 11, 203–216 Severinson Eklundh, K., & Kollberg, P (1996) Computer tools for tracing the writing process: From keystroke records to S-notation In G Rijlaarsdam, H van den Bergh, & M Couzijn, M (Eds.), Current Research in writing: Theories, models and methodology (pp 526–541) Amsterdam: Amsterdam University Press Silva, T (1993) Toward an understanding of the distinct nature of L2 writing: The ESL research and its implications TESOL Quarterly, 27, 657–677 Sommers, N (1980) Revision strategies of student writers and experienced adult writers College Composition and Communication, 31, 378–388 Spelman Miller, K (2000a) Academic writers on-line: Investigating pausing in the production of text Language Teaching Research, 4, 123–148 Spelman Miller, K (2000b) Writing on-line: Temporal features of first and second language written text production Unpublished doctoral thesis University of Reading, England Spelman Miller, K (2005) Second language writing research and pedagogy: A role for computer logging? Computers and Composition, 22, 297–317 THE PSYCHOLINGUISTIC DIMENSION IN SECOND LANGUAGE WRITING 451 Spelman Miller, K., & Sullivan, K P H (2006) Keystroke logging: an introduction In K P H Sullivan & E Lindgren (Eds.), Computer key-stroke logging: Methods and applications (pp 1–9) Oxford: Elsevier Stevenson, M (2005) Reading and writing in a foreign language: A comparison of conceptual and linguistic processes in Dutch and English Unpublished doctoral thesis University of Amsterdam, The Netherlands Strömqvist, S., & Ahlsén, E (Eds.) (1999) The process of writing: A progress report Göteborg, Sweden: Göteborg University, Department of Linguistics Strömqvist, S., Ahlsén, E., & Wengelin, Å (1999) The production process in speech and writing In S Strömqvist & E Ahlsén (Eds.), The process of writing: A progress report (pp 9–24) Göteborg, Sweden: Göteborg University, Department of Linguistics Strömqvist, S., Holmqvist, K., Johansson, V., Karlsson, H., & Wengelin, Å (2006) What keystroke logging can reveal about writing In G Rijlaarsdam (Series Ed.) and K P H Sullivan & E Lindgren (Eds.), Computer key-stroke logging: Methods and applications (pp 45–71) Oxford: Elsevier Sullivan, K P H., Kollberg, P., & Pålson, E (1998) Trace-it—A computer tool with application to the second language classroom BABEL: Journal of the Australian Federation of Modern Language Teachers Associations., 33, 22–27 Sullivan, K P H., & Lindgren, E (2002) Self-assessment in autonomous computeraided L2 writing ELT Journal, 56, 258–266 Sullivan, K P H., & Lindgren, E (Eds.) (2006) Computer key-stroke logging: Methods and applications Oxford: Elsevier Thorson, H (2000) Using the computer to compare foreign- and native-language writing processes: A statistical and case study approach Modern Language Journal, 84, 55–70 Tobin, L (Ed.) (1994) Taking stock: The writing process movement in the ‘90s Portsmouth, NH: Heinemann Trimbur, J (1994) Taking the social turn: Teaching writing post-process College Composition and Communication, 45, 108–118 van Gelderen, A., & Oostdam, R (2002) Improving linguistic fluency for writing: Effects of explicitness and focus of instruction L1—Educational Studies of Language and Literature, 2, 239–270 van Waes, L (1991) De computer en het Schrijfprocess De Invloed van de Teksverweker op het Pauze—en Revisiegedrag van Schrijvers Enschede, The Netherlands: Univsiteit Twente van Waes, L., & Schellens, P J (2003) Writing profiles: The effect of the writing mode on pausing and revision patterns of experienced writers Journal of Pragmatics, 35, 829–853 Wengelin, Å (2006) Examining pauses in writing: Theory, methods and empirical data In K P H Sullivan & E Lindgren (Eds.), Computer key-stroke logging: Methods and applications (pp 107–130) Oxford: Elsevier Wengelin, Å (2007) The word level focus in text production by adults with reading and writing difficulties In M Torrance, L van Waes, & D Galbraith (Eds.), Writing and cognition: Research and applications (pp 67–82) Amsterdam: Elsevier Whalen, K., & Ménard, N (1995) L1 and L2 writers’ strategic and linguistic knowledge: A model of multiple-level discourse processing Language Learning, 44, 381– 418 White, R V., & Arndt, V (1991) Process writing Harlow, England: Longman Yau, M (1989, March) A quantitative comparison of L1 and L2 writing processes Paper presented at the 23rd Annual TESOL Convention, San Antonio, Texas, United States 452 TESOL QUARTERLY APPENDIX A Time Spent Writing (in Minutes) Writer Year Year Year 3 10 11 12 13 14 15 16 17 77 57 67 69 85 71 94 55 71 65 80 71 75 64 60 68 50 32 55 33 70 70 66 51 60 66 80 54 53 38 60 51 45 45 62 47 67 56 68 66 52 64 55 56 33 61 47 40 Note This table includes those writers who failed to undertake all three tasks They are Writers 9, 13, and 17 APPENDIX B Sample Texts Produced by the Same Writer (Jonas) Across Years Jonas, Year First at the day I arive with the buss then I walk to our home calssroom and some times we wait for the lesson to start, but some times the buss come late to school and we have to hurry up to the lesson At the lesson I work mostly At the brake I maby go to the labery or to UG, some times I sit down and chat with my friends And at the second break we have mithle meal and I go to the dinner room, there I go and take usely two sandwitches and a glas with milk and then I sit down and eat When I have eaten I go with my friends somewere At the PE I am some times outside and some times inside and have for example foot ball, hand ball, ice hocky, and basket.Once in the weak we have 40 minutes break, at that break I dont so very much becouse the labery is closed and UG opens a little bit laiter Some days maby somebody come over to the school and talks about something who we can learn Nerly allways at the start on a lesson I haveto go to the home classroom were my locker is and get my books and my pencil We have a computer lesson once a weak and thats a funny lesson At the English lessons we write and read in english books At the lunch break I go to the dinner room again and haveto wait a little time on the people who is before me sometimes, when I goten my food, I sit down and eat nearly all the time with my friends After I have eaten I go somewere, maby to the labery At the end of the school I go to my locker and take out my bag and put the books (who I have homework in) in my bag, and I walk to the jacket room and get my jacket and then I go out and wait for the buss or for my parents (becouse they are taxi drivers both of them) when for exampel the buss comes I show my buss card and step into the buss Jonas, Year Hello I think you should come and visit me at Christmas here in Sweden I have allredy buyd many presents for you and if you come i dont need to send them to you At Christmus we can eat very THE PSYCHOLINGUISTIC DIMENSION IN SECOND LANGUAGE WRITING 453 good food at the Christmus table There we have for example meatballs, fish and very good sallads Then we can go and doo some snowboarding At new year we can buy fier crackers and blow some postboxes high up in the air and some rockets who can be pritty butiful And we can go to a fjäll wre we can have fun and you can get to see the snow Why dont you get down here and have some fun here in Sweden Good bye Jonas, Year Hello I come from the year 1999 and I have traveled with a time-machine to your time and I am about to tell you how it is and looks in the twentieth century XX is a little city in Sweden near YY In XX they have a fine school named ZZ and there I am studing right now, or before I traveled with the time-machine witch I was given by some aliens I met in XX And this school is a very fine school We have in this school some hi-tech computers, maby not so higly developed as yours, but how would I know, maby you just have not build any greater computers But anyhow we alot of work on the computers and then we use to use internet when we search for facts to our school-works And we have pretty fine air here in the year 1999, not like the air here in 400 years into the futur were it is poluted and thick and not so pleasant XX is a very small city that have some shops but not any big shops so it can get pretty more expensiver than in shops in bigger cities And I live in a tall lilttle village called XY and there it is not so much to do, but this summer I will go fishing, because this village has a small lake named YZ and there you can have much fish, if you now what fish is? Well it is a kind of a animal that live in the water (in my time there were running water a little bit everywere) and they taste good, but this fish that I was fishing is so bony so I give it to the cat, but some times you get very fine fish witch is good to eat And now before I traveled here I was trying to fix up my boat so it would get nice and then I can have a relaxed summer In my time we have airplanes that is not so advanced as yours but you can get to other countries with them A bad thing with my time is that it is war in a countrie named Kosovo and then there is an organisation called NATO that is bombing them because they dont whant them to start a war so to prevent Kosovo from starting war NATO starts war and that sound pritty sick, and that it is, but now the clock is ticking so if anywon whants to come with me into my clean and good time and leave your time that is about to collaps just hurry up and thats it 454 TESOL QUARTERLY ... Miller, 2000a, 2000b), ages of writer (Pålson, 1998) , and contexts (quasipedagogic and classroom) (Lindgren, 2004; Sullivan, Kollberg, & Pålson, 1998; Sullivan & Lindgren, 2002; Stevenson, 2005;... observation techniques, such as those used by Matsuhashi (1981, 1987), and more recently by Bosher (1998) , Chanquoy, Foulin, and Fayol (1996), Janssen, van Waes, and van den Bergh (1996), Phinney... Theories, models and methodology (pp 145–158) Amsterdam: Amsterdam University Press Bosher, S (1998) The composing processes of three Southeast Asian writers at the post-secondary level: An exploratory