Feedback providing improvement strategies and reflection on feedback use: Effects on students’ writing motivation, process, and performance

https://doi.org/10.1016/j.learninstruc.2011.10.003Get rights and content

Abstract

This study investigated the effects of feedback providing improvement strategies and a reflection assignment on students’ writing motivation, process, and performance. Students in the experimental feedback condition (n = 41) received feedback including improvement strategies, whereas students in the control feedback condition (n = 41) received feedback without improvement strategies. Within each feedback condition, half of the students received a reflection assignment on feedback use and the revision (experimental reflection condition), while the other half received a reflection assignment on feedback perception (control reflection condition). Results indicated that in the experimental feedback condition writing performance gained from the control reflection assignment, while in the control feedback condition it gained from the experimental reflection assignment. Improvement strategies negatively predicted self-efficacy beliefs, especially when initial self-efficacy beliefs were low, and positively predicted planning/revising. Reflections on feedback use and the revision positively predicted mastery goal when mastery goal initially was low or moderate.

Highlights

► Feedback providing strategies together with reflection on use of feedback harms performance. ► Feedback providing strategies reduces students’ self-efficacy beliefs. ► Students’ feedback perception explains negative effects on self-efficacy beliefs. ► Feedback providing strategies contributes to students’ planning/revising.

Introduction

Many freshmen’s and graduate students’ writing capacities are insufficient, and therefore writing education is open to improvement (Kellogg & Whiteford, 2009). Several approaches have been suggested to improve writing education, focusing on specific instructional practices that directly aim at writing skills (e.g., strategy instruction) (Graham and Perin, 2007, Kellogg and Whiteford, 2009), but also on practices that enhance students’ writing motivation (e.g., giving students choices about the topic of writing) (Bruning & Horn, 2000).

Feedback is an instructional practice indicated as enhancing both students’ skills and motivation (Brown, 2004, Bruning and Horn, 2000, Crooks, 1988, Kellogg and Whiteford, 2009, Kluger and DeNisi, 1996). Feedback is information provided by an external agent regarding some aspect(s) of the learner’s task performance, intended to modify the learner’s cognition, motivation and/or behavior (Kluger and DeNisi, 1996, Shute, 2008). Despite the fact that teachers spend much time providing students with feedback on their writing (Stern & Solomon, 2006), research on the effects of feedback provided on the adequacy of written products is scarce (Graham & Perin, 2007). For both scientific and practical reasons it is important that this gap is reduced.

Feedback providing improvement strategies is one of the kinds of feedback that is suggested to contribute to students’ writing performance (Stern and Solomon, 2006, Straub, 1996, Straub, 1997). Harris, Alexander, and Graham (2008, p. 88) describe a ‘strategy’ – whether heuristic or algorithmic in character – as “how to” knowledge that can be applied in the solution of a given problem. Strategies may vary on several dimensions and can be cognitive, metacognitive or self-regulative, and general, domain-specific or task-specific (Alexander, Graham, & Harris, 1998). Afflerbach, Pearson, and Paris (2008, p. 368) argued that strategies are deliberate, goal-directed attempts to control and modify a learners’ effort to solve problems, whereas ‘skills’ are automatic actions that occur without awareness of the components and control involved. Thus, the same actions could be either a skill or a strategy, depending on the learners’ awareness, control, and intention (Afflerbach et al., 2008). In this view, when strategies become more and more effortless and automatic, they will become fluent skills. As a result, providing students with writing improvement strategies can be expected to contribute to their writing skills.

Feedback providing improvement strategies suggests how to close the gap between the current product and the goal product, and as such is at the core of formative assessment (Sadler, 1989). In the present study, the effects of feedback providing improvement strategies on students’ writing motivation, writing process, and writing performance are investigated. As feedback has to be paid attention to in order to enhance performance (Anseel et al., 2009, Bangert-Drowns et al., 1991), additionally the effects of a reflection assignment dealing with intentions to use the feedback and the intended approach to the revision are examined.

First, theory concerning feedback effects on performance which includes students’ motivation and task process as mediators will be addressed. Second, feedback providing strategies and reflection on feedback use are discussed. Third, feedback providing strategies and reflection on feedback use are related to writing motivation and the writing process.

Feedback effects on performance have been shown to be highly variable (Kluger & DeNisi, 1996). To account for this variability, Kluger and DeNisi developed Feedback Intervention Theory (FIT). FIT integrates varying theoretical and paradigmatic perspectives (e.g., control theory, goal setting theory), and aims at accounting for known processes that are not addressed by those perspectives.

According to FIT, standards or goals are organized hierarchically. FIT distinguishes three control levels. Meta-task processes involve the self and are at the top of the hierarchy. Task-motivation processes involve the focal task and are located in the middle of the hierarchy. Task-learning processes involve the task-details and are at the bottom of the hierarchy. Higher-level processes can influence lower-level processes. Meta-task processes have the potential to affect task processes through linking higher-order goals (e.g., investing in my career) to task goals (e.g., writing a paper for this course) (Kluger & DeNisi, 1996).

The unique argument of FIT is that the effect of feedback on performance depends on the processes at which feedback directs attention. According to FIT, negative discrepancies between the performance and the standard will generally direct attention to task-motivation processes, leading to more effort. When this does not reduce the discrepancy, attention might shift to components of task execution (task-learning processes) resulting in alternative attempts to execute the task, or attention might shift away from the task to issues involving the self, such as self-esteem and impression management (meta-task processes). In general, feedback cues that direct attention to task-motivation processes or task-learning processes – coupled with corrective information on erroneous ideas or hypotheses – are assumed to enhance feedback effects on performance. Feedback cues that direct attention to meta-task processes are supposed to reduce the effect of feedback on performance, because they divert attention from the task (Kluger & DeNisi, 1996).

FIT has several limitations. First, Kluger and DeNisi (p. 276) acknowledge that FIT lacks very detailed and specific predictions and indicate that research should investigate feedback effects on the indicated mediating processes. Second, we argue that FIT should be more specific on the three control levels. Commonly, motivation literature does not deal with motivation in its general meaning, but with specific “components” of motivation (e.g., self-efficacy beliefs, achievement goals; see for example Murphy & Alexander, 2000). Likewise, several task processes can be distinguished (e.g., planning, revising). Also on the meta-task level distinctions can be made (e.g., impression management, self-esteem). Third, the propositions of FIT are formulated as unidirectional effects (e.g., “feedback effects on performance are augmented by cues that direct attention to task-motivation processes”, p. 268), while Kluger and DeNisi also state that on each control level processes may occur that lead to either negative or positive effects on performance. So, questions to be answered are: (a) which feedback cues relate to specific constructs on the three levels of FIT; (b) what is the direction of the effects of these cues; and (c) do these effects mediate feedback effects on performance?

In the literature several features of feedback messages have been mentioned that would enhance or reduce performance in general or writing performance in particular. These features concern improvement strategies, explanations, questions, specificity, teacher control, tone, amount of feedback, grades, and sign, and can be linked to the three levels of FIT (Duijnhouwer, Prins, & Stokking, 2010). In the present study the focus is on feedback providing improvement strategies. This kind of feedback is related to writing performance, motivation, and process. Feedback features other than strategies are treated as covariates.

Feedback providing strategies for performance improvement suggests means to reduce the gap between the actual performance and the standard (Sadler, 1989). An improvement strategy may for example indicate to schematize the line of reasoning in order to make the text structure explicit to the reader. Providing improvement strategies is a form of instructional scaffolding: the teacher helps the student to internalize routines and procedures to complete the task that is just too difficult to perform alone (Applebee, 1986).

Strategies may concern the writing process, that is, they may aim at planning or revising (Graham, 2006) (e.g., for planning, a strategy may explain how to make an outline in order to clarify text structure; for revising, a strategy may suggest to reread the text solely on spelling). Strategies may also aim at particular text aspects, that is, text content, structure, language, documentation, or layout (e.g., to improve text structure, making an outline can be suggested; to improve language, rereading on spelling can be indicated). Thus, strategies may focus on a particular process and on a particular text aspect. Furthermore, in writing, peer review is a common practice to improve text quality. Thus, strategies can also be distinguished with respect to the involvement of others (e.g., rather than making an outline, the suggestion may be to discuss the text structure with a peer).

When feedback is given with the aim to enhance writing performance, the assumption is that it evokes reflection on the content and the process of writing. Mindful reception of the feedback promotes learning or performance (Bangert-Drowns et al., 1991, Salomon and Globerson, 1987). Reflection is a process that starts with the identification of a problem and the decision to seek a solution. A plan or decision to act is made up through cognitive activities (Rogers, 2001). Examples of cognitive activities are comparing, analyzing, evaluating, questioning, concluding, and intending (Stokking, Van der Schaaf, Leenders, & De Jong, 2004). As a result of feedback on writing, reflection would start with the identification that the text is not (yet) what it should be, and – in case the text has to be revised – the decision to seek a solution for the problems identified in the feedback. A student could for example analyze the text based on the teacher’s comments, evaluate the feedback (“Do I agree with the teacher’s point of view?”), conclude whether or not action should be taken, and make a plan (formulate an intention) on how to approach the revision. This example illustrates that, as Stokking et al. propose, not only cognitive activities can be distinguished, but also objects of these activities (the text in its present form, the feedback content, and the task approach for the subsequent draft; other objects might for instance be the first draft writing process and the way the feedback is communicated). The example also shows that feedback can instigate reflection, as well as be object of reflection.

Reflection on the feedback is supposed to add to the effect of feedback on performance because it directs the feedback recipient’s attention to the task level (cf. Anseel et al., 2009). One of the formal activities that may trigger and structure the students’ reflection process is asking questions (Seibert, 1999, Seibert and Daudelin, 1999, Smits et al., 2009). Instruction that provides students with questions to ask themselves during problem-solving has been shown to enhance performance (Ge & Land, 2003). Instructional procedures that pose questions that progress from what to why to how follow the reflection process as described. A problem is clarified, ideas on the causes generated, and action planned (Seibert & Daudelin, 1999). Since the utility of a reflection instigating procedure depends on the match between what is activated and what is required (Salomon & Globerson, 1987), we expect that a reflection assignment that asks for students’ opinion on the feedback, their intentions to use the feedback, and their intentions on the approach to writing the subsequent draft poses effective reflection enhancing questions.

According to social cognitive theory, self-efficacy beliefs are a core motivational construct, as they are a major determinant of effort and persistence (Bandura, 1977). With regard to writing tasks, students’ self-efficacy beliefs concern their judgment of their capability to perform actions leading to the required text (cf. Bandura, 1986, Schunk and Swartz, 1993). These beliefs are predictive of students’ writing performance (McCarthy et al., 1985, Meier et al., 1984, Shell et al., 1989, Zimmerman and Bandura, 1994).

Self-efficacy beliefs are based on previous performances, observation of others performing the task, persuasion, and physiological and emotional states (Bandura, 1977, Van Dinther et al., 2011). Self-efficacy beliefs, once established, are assumed to be resistant to temporary failures (Bandura, 1997). However, the stability of self-efficacy beliefs has been rarely investigated (Bong & Skaalvik, 2003). With regard to feedback, it has been found that self-efficacy beliefs are susceptible to change, even after a single episode of particular teacher feedback (Duijnhouwer et al., 2010).

As feedback including improvement strategies suggests a means to reach a better performance, it brings performance improvement under learner control (see Bruning & Horn, 2000). It communicates that although the task perhaps is difficult, it can be accomplished. Therefore, this feedback can be expected to enhance students’ self-efficacy beliefs. This is in line with the notion that strategy instruction enhances students’ self-efficacy beliefs (Walker, 2003). While feedback is a form of persuasion and thus can be expected to affect students’ self-efficacy beliefs, students’ reflection on feedback use and the approach to the revision is not a source of efficacy information, and thus would not be expected to affect their self-efficacy beliefs.

Feedback providing strategies may also affect students’ purposes in the achievement context, that is, their achievement goal (Ames, 1992). Achievement goals are cognitive representations of what individuals try to do or want to achieve (Pintrich, Conley, & Kempler, 2003) with respect to achievement tasks (Pintrich, 2000). Achievement goals focus on competence (Dweck & Leggett, 1988). A distinction is made between mastery goals and performance goals (e.g., Dweck and Leggett, 1988, Elliot and Thrash, 2001). As students have a stronger mastery goal they have a stronger focus on developing their competence (Ames, 1992). As students have a stronger performance goal they have a stronger focus on getting their competence positively judged (Elliott & Dweck, 1988). A student adopting a mastery goal for writing tasks in class could for example evaluate his writing competence according to whether he improved in creating a good text structure. A student adopting a performance goal could for example compare his grade for a text with other students’ grades. Mastery goals and performance goals have been further distinguished based on their approaching or avoiding character: the desire to reach a positive result (approach) versus the desire to avoid a negative result (avoidance) (Elliot & McGregor, 2001; for a review see Elliot, 1999). The present study restricts itself to students’ mastery approach goal and performance approach goal (further called mastery goal and performance goal, respectively).

Achievement goals have been assumed to be relatively stable learner characteristics. This appears from studies that have focused on how stable student characteristics may cause students’ achievement goals (Senko & Harackiewicz, 2005), for example students’ theories on intelligence (Dweck & Leggett, 1988). However, research has shown that students’ achievement goals can change over a semester (Muis & Edwards, 2009), over weeks (Fryer & Elliot, 2007), and even within one laboratory session (Senko & Harackiewicz, 2005), as a result of new information, such as feedback (Fryer and Elliot, 2007, Senko and Harackiewicz, 2005). As feedback providing improvement strategies offers information to improve writing skills and performance, this kind of feedback may focus students on skill development and thus enhance their writing mastery goal. Indeed, feedback providing strategies is perceived by students as having a developmental focus (Lizzio & Wilson, 2008). We do not expect feedback providing improvement strategies to affect students’ performance goal. Reflection on feedback use and the approach to the revision makes students focus on their goals for the achievement task. Reflective questions that focus on using feedback may enhance both students’ mastery and performance goal. Students may strengthen their goal to learn or their goal to get a positive judgment of their text.

Improvement strategies concern the task process. With regard to writing, three major processes have been distinguished in Hayes and Flower’s (1980) writing model. During the planning process, information is retrieved from memory and the task environment in order to make a writing plan. This writing plan guides the translating process, in which information is transformed to text. During the reviewing process, the text produced is read and edited. The processes do not function as stages, but can be returned to recursively. The altering of processes is determined by the monitor (Hayes & Flower, 1980). In a revision of the model, Hayes (1996) added working memory and emphasized its role in the writing process. Working memory provides a limited capacity to process information and to temporarily store information (Kellogg, 1996) and is essential for the functioning of the cognitive processes involved in writing (Hayes, 2006). In addition, help-seeking is of interest, as socio-cognitive writing models (e.g., Hayes, 1996, Zimmerman and Risemberg, 1997) emphasize the social environment in which writing takes place. In academic writing, peer review is a common practice. It has been shown that supportive interaction with peers contributes to generating persuasive text content (Harris, Graham, & Mason, 2006). In the present study, improvement strategies suggest planning, revising, and help-seeking, and, therefore, effects on these writing processes are investigated. Additionally, effects on students’ effort (including concentration) are examined.

Research on writing instruction has shown that strategy-focussed writing instruction stimulates text planning activities (Braaksma et al., 2004, De La Paz and Graham, 2002, Fidalgo et al., 2008) and contributes to text quality (Fidalgo et al., 2008, Graham and Perin, 2007). Research on feedback that provides learners with improvement strategies is however scarce. The use of writing revision strategies was modeled in a study of Zimmerman and Kitsantas (2002). Students who saw a model making and correcting revision errors and improving in the use of revision strategies, reported higher self-efficacy beliefs and performed better on a writing revision task than students who saw a model flawlessly implementing revision strategies and than students who saw no model. So the combined information of revision strategies and dealing with errors was most beneficial. While in Zimmerman and Kitsantas’ study students watched a model using improvement strategies, it may be more effective to provide feedback that includes improvement strategies. Feedback including improvement strategies was provided in a study by Narciss and Huth (2006). Children who received strategic information on the correction of subtraction errors reported higher motivation and performed better than children who did not receive this information. In another study, Narciss (2004) found that students who received strategic information for the correction of errors in concept identification tasks (dealing with the identification of attributes and rules) did not perform better than students who did not receive this information. However, if the available time-on-task was restricted, strategic information did contribute to students’ effort and task engagement.

So, if feedback providing strategies offers suggestions with regard to planning and revising, this feedback can be expected to increase planning and revising, as well as effort and performance. If strategies focus on consulting others, feedback providing strategies can be expected to increase help-seeking. A reflection assignment on feedback use, whether or not this feedback includes strategies, makes the student formulate intentions on planning and revising. Therefore, a reflection assignment on feedback use can be expected to contribute to planning and revising, as well as to performance.

The present study addresses the following research questions: (a) What is the effect of feedback providing improvement strategies on students’ writing performance, writing motivation, and writing process? (b) What is the effect of a reflection assignment asking for students’ opinion on the feedback, their intentions to use the feedback and their intentions on the approach to writing the subsequent draft on students’ writing performance, writing motivation, and writing process?

For the first research question it was hypothesized that the mere provision of improvement strategies (i.e., comparison of experimental feedback condition and control feedback condition) as well as an increasing number of provided improvement strategies (within the experimental feedback condition) would lead to better performance (Hypothesis 1a), more positive self-efficacy beliefs (Hypothesis 1b), a stronger mastery goal (Hypothesis 1c), more planning and revising (Hypothesis 1d), more effort (Hypothesis 1e), and more help-seeking (Hypothesis 1f). No effects of improvement strategies were expected on the strength of students’ performance goal (Hypothesis 1g). It was explored whether effects of the improvement strategies depend on the initial level of performance, motivation, or writing process (i.e., before strategies were provided).

For the second research question it was hypothesized that reflecting on feedback use (experimental reflection condition) in comparison to reflecting on feedback perception (control reflection condition) as well as an increasing number of reflections on feedback use (within the experimental reflection condition) would lead to better performance (Hypothesis 2a) and more planning and revising (Hypothesis 2b). It was explored whether reflection conditions differed on self-efficacy beliefs, mastery goal, performance goal, effort, and help-seeking, and, within the experimental reflection condition, whether the number of reflections was related to self-efficacy beliefs, mastery goal, performance goal, effort, and help-seeking. It was also explored whether effects of the reflections depend on the initial level of performance, motivation, or writing process (i.e., before students reflected on the feedback).

Section snippets

Participants

The study took place at a university in the Netherlands, in a 9-week graduate course on competence-based education and assessment. Ninety-six students (81% women), ranging in age from 20 to 56 years (M = 26.85, SD = 6.52), participated in the course. All students were asked to fill in four questionnaires during the course and 12 students were asked for an interview after the course had finished (of these 12, 1 dropped out due to personal circumstances). All students participated in the study

Manipulation checks

Students in the experimental feedback condition received improvement strategies (M = 6.93, SD = 3.26), whereas students in the control feedback condition generally did not receive improvement strategies (M = 0.17, SD = 0.44), t(41.47) = −13.16, p < 0.01. So, the feedback manipulation was successful.

Students in the experimental reflection condition reflected more than students in the control reflection condition on utilization of the feedback (MeR = 1.68, SDeR = 0.74; McR = 0.11, SDcR = 0.31), t(59.50) = −12.88, p < 

Discussion

The aim of the present study was to investigate the effects of feedback providing improvement strategies and of a reflection assignment on students’ writing performance, writing motivation, and writing process. In terms of FIT, feedback providing improvement strategies was hypothesized to direct students’ attention to their task motivation as well as to their task process, and, as a result, to affect their performance. Particularly, it was expected that feedback providing improvement strategies

References (80)

  • L.A. Stern et al.

    Effective faculty feedback: the road less traveled

    Assessing Writing

    (2006)
  • J.W. Strijbos et al.

    Content analysis: what are they talking about?

    Computers and Education

    (2006)
  • M. Van Dinther et al.

    Factors affecting students’ self-efficacy in higher education

    Educational Research Review

    (2011)
  • B.J. Zimmerman et al.

    Becoming a self-regulated writer: a social cognitive perspective

    Contemporary Educational Psychology

    (1997)
  • P. Afflerbach et al.

    Clarifying differences between reading skills and reading strategies

    The Reading Teacher

    (2008)
  • P.A. Alexander et al.

    A perspective on strategy research: progress and prospects

    Educational Psychology Review

    (1998)
  • C. Ames

    Classrooms: goals, structures, and student motivation

    Journal of Educational Psychology

    (1992)
  • A.N. Applebee

    Problems in process approaches: toward a reconceptualization of process instruction

  • S. Askew et al.

    Gifts, ping-pong and loops – linking feedback and learning

  • A. Bandura

    Self-efficacy: toward a unifying theory of behavioral change

    Psychological Review

    (1977)
  • A. Bandura

    Social foundations of thought and action: A social cognitive theory

    (1986)
  • A. Bandura

    Self-efficacy: The exercise of control

    (1997)
  • R.L. Bangert-Drowns et al.

    The instructional effect of feedback in test-like events

    Review of Educational Research

    (1991)
  • M.A.H. Braaksma et al.

    Observational learning and its effects on the orchestration of writing processes

    Cognition and Instruction

    (2004)
  • M. Bong et al.

    Academic self-concept and self-efficacy: how different are they really?

    Educational Psychology Review

    (2003)
  • S. Brown

    Assessment for learning

    Learning and Teaching in Higher Education

    (2004)
  • R. Bruning et al.

    Developing motivation to write

    Educational Psychologist

    (2000)
  • T.J. Crooks

    The impact of classroom evaluation practices on students

    Review of Educational Research

    (1988)
  • S. De La Paz et al.

    Explicitly teaching strategies, skills, and knowledge: writing instruction in middle school classrooms

    Journal of Educational Psychology

    (2002)
  • H. Duijnhouwer et al.

    Progress feedback effects on students’ writing mastery goal, self-efficacy beliefs, and performance

    Educational Research and Evaluation

    (2010)
  • C.S. Dweck et al.

    A social-cognitive approach to motivation and personality

    Psychological Review

    (1988)
  • A.J. Elliot

    Approach and avoidance motivation and achievement goals

    Educational Psychologist

    (1999)
  • A.J. Elliot et al.

    A 2 × 2 achievement goal framework

    Journal of Personality and Social Psychology

    (2001)
  • A.J. Elliot et al.

    Achievement goals and the hierarchical model of achievement motivation

    Educational Psychology Review

    (2001)
  • E.S. Elliott et al.

    Goals: an approach to motivation and achievement

    Journal of Personality and Social Psychology

    (1988)
  • J.H. Flavell

    Metacognitive aspects of problem solving

  • J.W. Fryer et al.

    Stability and change in achievement goals

    Journal of Educational Psychology

    (2007)
  • T. Garcia et al.

    Assessing students’ motivation and learning strategies in the classroom context: the Motivated Strategies for Learning Questionnaire

  • X. Ge et al.

    Scaffolding students’ problem-solving processes in an ill-structured task using question prompts and peer interactions

    Educational Technology Research and Development

    (2003)
  • S. Graham

    Strategy instruction and the teaching of writing: a meta-analysis

  • Cited by (90)

    • Effectiveness of online regulation scaffolds on peer feedback provision and uptake: A mixed methods study

      2022, Computers and Education
      Citation Excerpt :

      As they included sufficient details about the performance, the students might not have needed the scaffolds for further processing and planning before revision. Similarly, Duijnhouwer et al. (2012) reported redundancy of reflection on feedback that involved improvement strategies. Moreover, in the second task, students received a few implementable feedback comments, which might have made using the ORS for feedback processing unnecessary.

    View all citing articles on Scopus
    View full text