Understanding the temporal organization of writing is key to studying writing processes. Existing methods to segment writing into phases often rely on arbitrary rules, extensive manual annotation, or focus on numerous transitions. This study aimed to develop an automated segmentation method to detect distinctive transition in the dominant writing process, particularly the transition from first draft to revision. For this, keystroke data (source-based L1 writing (N = 80) and text simplification in L2 (N = 88)) were manually annotated. The BEAST algorithm was applied for Bayesian change point detection, based on five characteristics derived from the annotation criteria: (1) percentage of the final text written so far, (2) distance between typed and remaining characters, (3) relative cursor position, (4) source use, and (5) pause timings. The first three features proved most effective in identifying change points. A rule-based approach was further applied to select one final change point, which resulted in mediocre accuracy ranging from 31% exact agreement to 49% agreement within 60 seconds. To conclude, the BEAST algorithm is useful in detecting a variety of change points in writing processes, yet connecting them to meaningful phases is still quite complex.

Despite the growing interest in the dynamics of the writing process in writing research, publicly available large-scale corpora of keystroke logs have been rare. We introduce KLiCKe, a freely available collection of keystroke logs for around 5,000 argumentative texts written by adults in the United States. The KLiCKe corpus also includes human-rated holistic scores for the essays as well as writers' demographic details, their typing skills, and vocabulary knowledge. We describe our methods for constructing the corpus and present descriptives for different components of the corpus. To illustrate the use of the KLiCKe corpus, we report a study using a subset of the corpus to investigate whether keystroke features are associated with holistic writing quality for L1 and L2 writers. The study shows that higher writing scores are related to shorter pauses in general, shorter between-word pauses, lower proportion of deletions, higher proportion of insertions, and less process variance. The KLiCKe corpus provides a robust resource for researchers to study the dynamics of text production and revision that will help spur the development of process-oriented tools and methodologies in writing assessment and instruction.

Artificial intelligence-based Language Tools (AILTs) are being increasingly used in essay writing in higher education. Its application promotes global and multicultural perspectives in education and plays a critical role in advancing scholarly communication and research dissemination. However, these benefits cannot be measured without also considering student perspectives. This study analyzes the positive and negative aspects identified by students regarding the use of AILTs in their written texts at university. A total of 314 undergraduate and graduate education students were surveyed, and results were analyzed using the Reinert method. The results show that positive aspects are linked to the three pillars of text construction (planning, textualization, and revision). The negative aspects highlight concerns about academic integrity and student competencies. These findings can help guide teachers on how they can promote the responsible and beneficial use of AILTs.

This article reports on a scoping review of the literature exploring the alignment between behavioral and linguistic units in L1 and L2 writing. Behavioral units in writing were assessed using keystroke logging measures of pauses, bursts, and revisions. Linguistic units were operationalized based on lexical and syntactic definitions from the literature. Nine empirical studies met the inclusion criteria. Most of these studies focused on L1 English writing by adult participants, although some explored other first languages, such as German, Dutch, and French. The identified L2 studies focused exclusively on English. Due to the limited number and high heterogeneity of the studies, no definitive conclusions can be drawn. However, meaningful links between behavioral and linguistic units were detected. In addition to confirming some previously known phenomena, the studies provided new evidence on online processing during pauses, revealed certain idiosyncrasies in L1 versus L2 writing cognition, and offered new insights into the nature of revision. We provide a critical interpretation of the results, propose new research directions, and recommend solutions.

Bursts of writing, extracted from online recordings of the writing process, have proved an invaluable vantage point into the cognitive mechanisms at work during written language production. Crucially, they show that writers, much like speakers, produce language through a sequence of small ‘chunks’, patterns-like groupings of words that do not necessarily match the structures of theoretical grammars. As such, they are intriguing objects, whose linguistic properties are yet to be understood. To contribute to this endeavor, we track all instances of French so-called clitic subjects in a corpus of 81 keylogs of short essays written by undergraduate students in experimental conditions. We show that these clitic subjects are attracted to the burst-initial position, favoring resumption of the production after revision events. Moreover, they also act like discursive hubs in that writers are more likely to revise up to a clitic subject and restart from there, possibly relying on an entirely different structure. Therefore, they play the role of landmarks in the writing process, from which information can flow, and to which writers can get back to develop alternative discursive strategies. These results hint that the writing process and the information structure of the product are likely to be intimately intricated.

This study investigates the impact of peer orthographic revision using a self-reflection tool on orthographic performance in order to improve the understanding and applying of phonological, contextual and morphological rules in third-grade students. Children were assigned to one of three groups: two experimental groups (individual group, dyadic interaction group) and a control group. In the experimental training programme, a self-correction orthographic rubric was used, but while children in the individual group self-corrected the words, children in the dyadic interaction group did it in pairs and interacted in a way such that they should always reach an agreement on the correct spelling. The results showed that although both experimental groups decreased the number of misspellings in the post-test, the dyadic interaction group had the best results, differing significantly from the others, suggesting that self-correction strategies based on rubrics that explicitly display orthographical rules along with collaborative peer learning have a very positive impact on orthographic.

Writing processes constitute a complex interplay of planning, formulation and revision. Ideas take shape through the activation of previous knowledge and, when permitted, also its synthesis with information from sources that help to complement it and resolve doubts and shortcomings arising during writing. The possibility to use external help can be especially useful to those writing in a foreign language, and questions about the nature of the sources consulted can contribute new insights into language processing in the multilingual mind, as well as expand our notion of fluency. While leaving the target text is often considered a distraction, a ‘breakdown’ in fluency, it is, in fact, a part of language processing and text creation.       This article proposes a novel way to use keylogging data from Inputlog (Leijten and Van Waes 2013) to visualise the crosslinguistic nature of solving language and content problems in L3+ writing: creating process graphs to display the temporal dynamics of different types of sources used. The example data comes from a university-level course on Spanish linguistics, where Spanish was a third or subsequent language for the participants. Evidently, the vast majority of their external activity was language-related and brief, and, interestingly, a great part of it recurs to a lingua franca, English. Some social context and reasoning is offered to explain such an observation.

This ethnographic case-study examines the impact of asynchronous written feedback comments on the peer feedback process within one doctorate writing group. The doctorate students were interviewed retrospectively about their perceptions of effective feedback comments. Affective components (e.g. hedging devices) and effective components (e.g. revision comments) within the reviewers’ feedback comments, and external components (e.g. reviewer competency) that influence the peer feedback process were induced from the interview transcripts using a grounded theory approach. Further evidence that these identified components impact the feedback process appreciably was triangulated from the analysis of two other datasets; the participants’ asynchronous written feedback comments and revision plans. The results show that the participants used much affect in their written feedback exchanges, and this affect had a strong impact on the effect of their feedback process. Thus, written affective language can play a significant role in how an author interprets and implements feedback comments. This suggests that affect can play a prominent role in helping to develop more effective feedback practices within writing groups. Helping writing communities develop a better understanding of affect within asynchronous written feedback comments can only help them to develop more useful feedback practices.

This study examined the relationships between writing anxiety, writing self-efficacy, and perceived use of revision strategies in high school students with different achievement goals as they learned argumentative writing in English Language Arts classrooms. Three achievement goal orientation profiles emerged from a sample of 307 American high school students on the basis of their mastery, performance-approach, and performance-avoidance goal orientations: Low on All, Average on All, and High on All. These three profiles of students significantly differed with respect to their writing anxiety and their perceived use of revision strategies. Writing self-efficacy mediated the effect of writing anxiety on the perceived use of revision strategies for students in the Average on All profile only. The findings suggest that students are diverse in their motivational and affective experiences with respect to argumentative writing, and caution against using a one-size-fits-all approach for teaching argumentative writing to students.

Keystroke loggers facilitate researchers to collect fine-grained process data and offer support in analyzing these data. Keystroke logging has become popular in writing research, and study by study we are now paving the path to a better understanding of writing process data. However, few researchers have concentrated on how to bring keystroke logging to the classroom. Not because they are not convinced that writing development could benefit from a more process-oriented pedagogy, but because 'translating' complex and large data sets to an educational context is challenging. Therefore, we have developed a new function in Inputlog, specifically aiming to facilitate writing tutors in providing process feedback to their students. Based on an XML- logfile, the so-called 'report' function automatically generates a pdf-file addressing different perspectives of the writing process: pausing, revision, source use, and fluency. These perspectives are reported either quantitatively or visually. Brief introductory texts explain the information presented. Inputlog provides a default feedback report, but users can also customize the report. This paper describes the process report and demonstrates the use of it in an intervention. We also present some additional pedagogical scenarios to actively use this type of feedback in writing classes.

Teaching the craft of written science communication is an arduous task that requires familiarity with disciplinary writing conventions. With the burgeoning of technological advancements, practitioners preparing novice research writers can begin to augment teaching and learning with activities in digital writing environments attuned to the conventions of scientific writing in the disciplines. The Research Writing Tutor (RWT) is one such technology. Grounded in an integrative theoretical framework, it was designed to help students acquire knowledge about the research article genre and develop research writing competence. One of its modules was designed to facilitate revision by providing different forms of automated feedback and scaffolding that are genre-based and discipline-specific. This study explores whether and how the features of the RWT may impact revision while using this module of the tool. Drawing from cognitive writing modeling, this study investigates the behaviors of a multidisciplinary group of 11 graduate-student writers by exploring how they interacted with the RWT's features and how this interaction may create conditions for enhanced revision processes and text modifications. Findings demonstrate promising potential for the use of this automated feedback tool in fostering writers' metacognitive processing during revision. This research adds to theory on cognitive writing models by acknowledging the evolving role of digital environments in writing practices and offering insights into future development of automated tools for genre-based writing instruction.

While previous research on peer feedback in L2 writing has stressed the importance of training in giving useful comments on peer’s writing, very few studies have specifically explored metacognitive training in peer review interaction in terms of the perception and actual reviewing practices of L2 learners. This mixed-methods study employed a 12-week intervention course in L2 writing, in which eighteen Secondary One students, aged 12–13 years, received metacognitive training in peer review interaction. The training focused on metacognitive knowledge and metacognitive regulations. The results showed that, first, metacognitive training in peer review interaction helped change the perception of these young learners and increased their level of engagement and collaboration during the five peer review tasks. Second, metacognitive training appeared to encourage students to provide more content-related feedback than language-related feedback during dyadic interactions. Finally, it was found that the students were able to seriously consider their peer feedback when revising the peer-reviewed drafts, but written feedback had a much higher chance of being incorporated than did oral feedback. The pedagogical implications of these findings were discussed.

To date, research into functional descriptions of unfolding language has been almost entirely focused on speech. And whilst writing research has examined the revision of language units, it has backgrounded how these revisions contribute to the unfolding of a text’s meanings. Therefore, using Systemic Functional Linguistics (SFL) as an underlying framework, and keystroke logging software (Inputlog) as a data collection tool, this paper takes a first step toward a dynamic description of written text in terms of the language structures, functions, and systemic choices found in the written revisions of two 2nd year UK undergraduates. More specifically, in detailed textual analysis of four unfolding, digitally composed text, whose end products totalled approximately 1700 words, this paper focuses on the revisions made during consecutive writing sessions, which lasted anything from 8mins to 8hrs 37mins and totalled 56hrs 18mins of recordings. The findings suggest that certain language choices may play a key role when it comes to shaping academic essays, and it is proposed that this new model of analysis can provide an additional perspective on writing behaviour in terms of how meaning-making practices unfold in real time.

In this study we present a comparative report of two effective instructional programs focused on the improvement of upper-primary students’ writing competence through the promotion of revision skills. Both programs shared the main aim but had two different approaches. We contrasted writer-focused instruction with reader-focused instruction. To provide a valid report on the similarities and differences of the two programs, we applied two complementary dimensions. The first dimension, what the researcher intends students to achieve, provides insight into the types of students’ intermediate learning objectives and how they are sequenced. The second dimension, how to teach, includes the instructional design principles which relate the intermediate learning objectives to the specific learning and instructional activities in certain conditions. We analyse similarities and differences between the instructional programs and discuss the implications of using this kind of reporting system as a useful tool for reporting – and designing – writing interventions.

Writing is a complex, recursive, and strategic process that requires metacognitive competencies. Skillful writers have a high level of metacognitive strategy knowledge (MSK) and use strategies effectively. MSK about writing describes a person’s verbalizable knowledge and awareness of memory, comprehension, and higher order processes that underlie skillful writing. Measurement instruments assessing students’ MSK about academic writing in higher education that can be used for group settings and large samples are lacking. The aim of this article is to describe the development of a new MSK test instrument. The MSK test consists of three different writing scenarios related to the three self-regulated writing phases: planning prior to composing full text, monitoring the writing during composition, and subsequent revision. The findings of a pre-study (N = 51) and two studies (N = 23; N = 113) showed that the new MSK test is economical in use, is reliable and has high content validity. Further, the findings demonstrated external validity of the new instrument in terms of relationships with students’ metacognitive strategy use and writing performance. Implications for future research and educational practice are discussed.

Peer assessment is a technique with many possible benefits for instruction across the curriculum. However, the value obtained from receiving peer feedback may critically depend upon the relative abilities of the author and the reviewer. We develop a new model of such relative ability effects on peer assessment based on the well-supported Flower and Hayes model of revision processes. To test this model across the stages of peer assessment from initial text quality, reviewing content, revision amount, and revision quality, 189 undergraduate students in a large, introductory course context were randomly assigned to consistently receive feedback from higher-ability or lower-ability peers. Overall, there were few main effects of author ability or reviewer ability. Instead, as predicted, there were many interactions between the two factors, suggesting the new model is useful for understanding ability factors in peer assessment. Often lower-ability writers benefitted more from receiving feedback from lower-ability reviewers, while higher-ability writers benefitted equally from receiving feedback from lower-ability and higher-ability reviewers. This result leads to the practical recommendation of grouping students by ability during peer assessment, contrary to student beliefs that only feedback from high ability peers is worthwhile.

This study investigated how struggling adult writers solve a writing task and what they know about writing and themselves as writers. The writing process of the adult writers was examined by combining three elements: the observation of collaborative writing tasks, analyses of their written texts, and structured individual interviews that included both retrospective and prospective parts. This methodical approach provides productive tools to assess writing processes and writing knowledge of struggling adult writers. The triangulation of data from the different sources is visualized in a case study. Findings from the case study suggest both similarities and differences between struggling adult and younger writers. Concerning the writing process of both groups, planning and revision play a limited role. However, alongside these similar limitations in their writing process, struggling adult writers distinguish themselves from their young counterparts through their relatively extensive knowledge about themselves as writers.

One hundred and three (N=103) peer review studies contextualized in L2 composition classrooms and published between 1990 and 2015 were reviewed. To categorize constructs in research studies, this researcher used Lai’s (2010) three Ps dimensions (perceptions, process, and product). Perceptions are the beliefs and attitudes of peer review. Process refers to the learning process or implementation procedures of peer review. Product is the learning outcomes of peer review. A thematic analysis of the studies’ constructs showed that perception studies examined learners’ general perceptions/attitudes, Asian students’ perceptions/attitudes (cultural influences), and learner perceptions of peer feedback in comparison to self and/or computerized feedback. Process studies discussed the effects of training, checklists/rubrics, writer-reviewer relationships, the nature of peer feedback, communicative language, timing of teacher feedback on peer feedback, grouping strategies, as well as communicative medium. Product research, on the other hand, investigated peer feedback adoption rates and ratio of peer-influenced revisions, effects of peer review on writers’ revision quality, effects of peer review on reviewers’ gains, and effects of peer review on writers’ self revision. In light of this review, research gaps are identified and suggestions for future research are offered.

The purpose of this study was to examine how six middle-school students used Automated Essay Evaluation (AEE) technology to revise their writing. Students in a combined 7th and 8th grade Literacy class at one school participated in two in-depth think alouds and semi-structured interviews as they used AEE technology to revise their writing on two separate writing tasks. Constant-comparative analysis of data, including think alouds, semi-structured interviews, and student writing along with a separate quantitative analysis of student revisions revealed themes in three areas: (a) student use of AEE feedback to make revisions; (b) student motivation to revise their writing when using AEE technology; (c) and student understanding and application of AEE feedback during revision. Findings indicated that students who received low scores used AEE feedback to prompt non-surface revisions whereas students with high scores did not. Further, students who used AEE feedback to prompt non-surface revisions made more overall non-surface revisions, revised for different reasons, made more t-unit level revisions, and had more revisions rated as major successes than students who did not use the feedback. Students who used the AEE feedback, MY Editor, were often confused by the grammar and punctuation feedback and had a low success rate using it. However, students were more successful with the spell checker only feedback. In addition, findings show that students were motivated to revise because of the numerical scores the technology assigned their writing. Moreover, knowledge that they would receive a score prompted students to do extensive revising prior to submitting their writing for scoring. Finally, student understanding of the AEE feedback was varied. Implications for classroom use of AEE technology and directions for future research are discussed.

It has been established that in the Netherlands, as in other countries, a majority of students do not attain the desired level of writing skills at the end of elementary school. Time devoted to writing is limited, and only a minority of schools succeed in effectively teaching writing. An improvement in the way writing is taught in elementary school is clearly required. In order to identify effective instructional practices we conducted a meta-analysis of writing intervention studies aimed at grade 4 to 6 in a regular school setting. Average effect sizes were calculated for ten intervention categories: strategy instruction, text structure instruction, pre-writing activities, peer assistance, grammar instruction, feedback, evaluation, process approach, goal setting, and revision. Five of these categories yielded statistically significant results. Pairwise comparison of these categories revealed that goal setting (ES = 2.03) is the most effective intervention to improve students’ writing performance, followed by strategy instruction (ES = .96), text structure instruction (ES = .76), peer assistance (ES = .59), and feedback (ES = .88) respectively. Further research is needed to examine how these interventions can be implemented effectively in classrooms to improve elementary students’ writing performance.

Test writers sometimes complain they cannot perform to their true abilities because of time constraints. We therefore examined differences in terms of process and product between texts produced under test and non-test conditions. Ten L2 postgraduates wrote two argumentative essays, one under test conditions, with only forty minutes being allowed and without recourse to resources, and one under non-test conditions, with unlimited time as well as access to the Internet. Keystroke logging, screen capture software, and stimulated recall protocols were used, participants explaining and commenting on their writing processes. Sixteen writing process types were identified. Higher proportions of the processes of translation and surface revision were recorded in the test situation, while meaningful revision and evaluation were both higher in the non-test situation. There was a statistically significant difference between time allocation for different processes at different stages. Experienced teachers awarded the non-test texts a mean score of almost one point (0.8) higher. A correlational analysis examining the relationship between writing process and product quality showed that while the distribution of writing processes can have an impact on text quality in the test situation, it had no effect on the product in the non-test situation.

Given the widespread use of English for the international dissemination of scholars’ research results, numerous intercultural analyses have been undertaken in the field of English for Academic Purposes in diverse genres. Rhetorical and discursive conventions across languages and cultures have been studied to help non-native English scholars to be successful in the difficult endeavour of being granted publication in international English-medium publications. The increasing competition to get one’s research published in international journals in English has resulted in the authors’ need to clearly spell out what their contribution to their discipline is, a rhetorical convention which seems to be currently crucial especially in some fields. It is the aim of this paper to trace statements of contribution in the Introduction and Conclusion sections of research articles published in two international journals in finance and to compare the results with those obtained from an analysis of three manuscripts written in English by a team of Spanish scholars sent to the same journals but which received major revision or rejection reports. Reference to these statements made by reviewers in their reports will also be analysed to explore to what extent (non) compliance with this rhetorical convention may influence their final decision (not) to recommend publication.

This article seeks to formulate translator profiles based on process data from keylogging and eye-tracking, while at the same time identifying features which are shared by all translators in a sample consisting of both students and professionals. Data have been collected from 12 professional translators and 12 graduate students translating three texts of varying complexity. We found that individual behavioural characteristics with respect to initial orientation in the source text (ST), online ST reading, and online and end revision remained relatively constant across texts of varying complexity, supporting our hypothesis that translator profiles can be observed which are independent of the difficulty of the translation task. The analysis of the data also indicated that translators could be grouped into broad categories of locally-oriented and globally-oriented translation styles, which are partly, though not entirely, comparable to styles known from writing research. We also identified shared features with respect to reading and revision behaviour during drafting. Common to all translators was that they looked beyond the source text word they were about to translate, and that they made revisions while drafting the translation.

This study investigated situational and writer characteristics that influence the revision process. Thirty-four students who scored high on print exposure and 32 students who scored low on print exposure had 10 or 70 seconds to think about each of 2 prompts before beginning to write (prewriting time) the essays on a computer. A keystroke-logging program captured writing and editing behavior, including pauses, edits (deletions, substitutions, insertions), and prompt reviews. Quality was measured using an 8-factor, 3-point analytic scoring rubric. Results indicated that high print exposure students wrote longer and higher quality essays than low print exposure students. In addition, the short prewriting time increased prompt reviewing and average pause lengths. High and low print exposure writers showed differential responses to the prewriting time manipulation in terms of total pause-associated edits during writing. The complexity of the revision process and the importance of understanding multiple immediate variables in the writing situation are discussed.

The inclusion of peer feedback activities into the academic writing process has become common practice in higher education. However, while research has shown that students perceive many features of peer feedback to be useful, the actual effectiveness of these features in terms of measurable learning outcomes remains unclear. The aim of this study was to investigate the linguistic and review features of peer feedback and how these might influence peers to accept or reject revision advice offered in the context of academic writing among L2 learners. A corpus-based machine learning approach was employed to test three different algorithms (logistic regression, decision tree, and random forests) on three feature models (linguistic, review, and all features) to determine which algorithm offered the best predictive results and to determine which feature model most accurately predicts implementation. The results indicated that random forests is the most effective way of modeling the different features. In addition, the feature model containing all features most accurately predicted implementation. The findings further suggest that directive comments and multiple peer comments on the same topic included in the feedback process seem to influence implementation.

The object of this research is a learning method that uses email correspondence to promote the development of narrative writing skills in year 4 and year 5 students. Focusing on the written production of episodes of adventure novels and peer review, this learning method was applied to four classes in the Paris region over a 1-year period. The classes were paired in such a way that some students were required to read and analyze texts produced by correspondents (advice givers) while others carried out revisions using peer advice and suggestions (advice receivers). To describe the dynamics of writing, revision and learning, a qualitative analysis of the texts and suggestions given or received by the student partners is carried out. A statistical analysis comparing the texts produced by students in both groups is used to corroborate the findings of the initial analysis. Students showed an increasing awareness and consideration of the key characteristics of the practiced genre, resulting in an improvement of the quality of the texts in the course of the revision process and throughout the year. The texts produced by the advice givers improved more than the texts produced by the advice receivers. The findings are attributed to greater self-reflection and successive reformulations fostered by the elaboration of advice and suggestions.

Peer review has been viewed as a promising solution for improving students' writing, which still remains a great challenge for educators. However, one core problem with peer review of writing is that potentially useful feedback from peers is not always presented in ways that lead to revision. Our prior investigations found that whether students implement feedback is significantly correlated with two feedback features: localization information and concrete solutions. But focusing on feedback features is time-intensive for researchers and instructors. We apply data mining and Natural Language Processing techniques to automatically code reviews for these feedback features. Our results show that it is feasible to provide intelligent support to peer review systems to automatically assess students' reviewing performance with respect to problem localization and solution. We also show that similar research conclusions about helpfulness perceptions of feedback across students and different expert types can be drawn from automatically coded data and from hand-coded data.