TF_Template_Word_Windows_2007 - Lincoln Repository



Developing essay writing skills: an evaluation of the modelling behaviour method and the influence of student self-efficacySocial cognitive learning theory has shown that observational learning positively influences essay writing development in high-school students (Braaksma et al, 2004), and that self-efficacy impacts on motivation (Bandura, 1986, 1989). This study investigated the relative contribution of model observation, model evaluation, post-submission feedback, and factors relating to self-efficacy, as measured through academic confidence, in the essay writing development of 142 University students. The research compared students’ essay marks between two semesters in order to explore whether observational learning had an impact on the development of these complex skills. The results revealed that all students benefit from traditional feedback and higher levels of self-efficacy when developing their essay writing performance. Underperforming students particularly benefited from actual essay writing. However, contrary to the predictions drawn from the literature students in this study did not appear to benefit from observational learning when developing their skills. Limits to social learning theory are discussed.Keywords: social cognitive learning theory, self-efficacy, essay writingIntroduction Learning how to write essays is a complex process that is difficult to master (Rijlaarsdam et al, 2005). Essay writing forms the foundation of many assessments in higher education and mastering essay writing is a requirement for progression within the higher education environment. This paper focuses on predictions derived from social cognitive learning theory (Bandura, 1986, 1989; Braaksma, Rijlaarsdam, van den Bergh & van Hout-Wolters, 2004) regarding the contribution of observational learning from peer models, model evaluation during observational learning and self-efficacy in University students’ essay writing development.It is tradition, and considered to be good practice, within higher education to provide students with post-submission feedback on their essay writing abilities (Gibbs & Simpson, 2004). This approach is based on the premise that reflection on previous experiences is an important component of the learning process (Higgins, Hartley & Skelton, 2002) and that such reflection facilitates students’ learning, although whether or not students engage with the intended reflective process is debatable (Brookhart, 2001). What is evident, however, is that research in relation to high-school students in different cultural contexts has shown that model evaluation during observational learning is very effective for supporting the development of essay writing (Braaksma, Rijlaarsdam, van den Bergh & Hout-Wolters, 2004; Fidalgo, Torrance, Rijlaarsdam, van den Bergh & ?lvarez, 2015). This paper investigates whether University students can also profit from observational learning when developing their essay writing skills, and, importantly, compares the relative contribution that an observational learning approach may have in comparison to traditional post-submission feedback. Observational learning has a rich research history within educational environments. According to social cognitive learning theory observational learning occurs if one watches a role model’s behaviour, receives a verbal description of their behaviour, or views a symbolic representation (e.g. a cartoon) of their behaviour (Bandura, 1986, 1989). As a result of watching a role model learners acquire new behaviours, or decrease unwanted behaviours. According to Bandura’s work observational learning is part of a continuous interaction between cognitive, behavioural and environmental influences. A number of conditions need to be fulfilled in order for observational learning to take place, this includes attention (paying attention to what the models says or does), retention (retaining information in memory in order to be able to enact it in the future), production (being physically capable of producing the models actions) and motivation (students wanting to demonstrate the models actions, this can be supported through reinforcement). According to Bandura if any of these conditions are missing then this form of learning does not occur. Observational learning has been shown to be successful in relation to a range of academic subjects with a range of different age groups (Bandura, 1986, 1997; Rosenthal & Zimmerman, 1978; Sonnenschein & Whitehurst, 1984; Couzijn, 1995, 1999; Schunk, 1987, 1991, 1998; Schunk & Hanson, 1985, 1989a, 1989b; Zimmerman & Kitsantas, 2002; Schunk & Zimmerman, 2007). Rijlaarsdam, Braaksma and colleagues have applied the idea of observational learning to academic writing, where high-school students learnt academic writing skills from models that verbally went through the process of structuring an essay or a manual (Braaksma, van den Bergh, Rijlaarsdam & Couzijn, 2001). Students were also asked to critically discuss the behaviours displayed by the role models. This approach is proposed to have supported the necessary conditions for learning to take place. In contrast to Bandura, model observation and model evaluation, two criteria for observational learning, were shown to be sufficient for having a positive effect on the quality of the academic writing subsequently produced (Rijlaarsdam & van den Bergh, 2002; Braaksma, Rijlaarsdam, van den Bergh & van Hout-Wolters, 2004, 2006; Rijlaarsdam, Braaksma, Couzijn, Janssen, Raedts, van Steendam, Toornenaar & van den Bergh, 2008). Showing that only two criteria are sufficient for observational learning to take place poses the question what the relative contribution of each of these two components is to essay writing development. This is what we investigated in relation to University students.The impact of both observational learning and post-submission feedback on academic writing has been shown to be modified by the performance level of the students involved. Research investigating traditional post-submission feedback has revealed that underperforming higher education students do not use assessment feedback in the same way as high-performing students (Brookhart, 2001). Brookhart demonstrated that high-performing students engaged in self-assessment processes on a regular basis. Post-submission feedback received from tutors was used in this process to identify gaps in understanding or levels of skill and to rectify difficulties encountered in their work. In contrast, post-submission feedback was less effective for underperforming students who had no clear idea of what to do with the feedback received from their tutors (Sadler, 1998). In addition, research on observational learning has shown that high-performing and underperforming students at high-school level only profit from approaches to learning if the ability levels of the models and the learners are matched (Braaksma, Rijlaarsdam & van den Bergh, 2002). If the ability levels of the role models and the learners are not matched, students do not learn from the models. Braaksma et al (2002) refer to the tendency to learn more from peer models that have similar abilities as the similarity hypothesis. The similarity hypothesis has been explained in reference to Bandura’s social cognitive learning theory (1986, 1989): the distance between the learner’s current ability and the ability of a peer model may either motivate learners or reduce their belief in their ability to achieve the required level of competence. Social cognitive learning theory thus assumes that student motivation is an important component supporting the success of observational learning and learning through model evaluation (we return to discuss the issue of student motivation in our discussion of self-efficacy below). Since a student’s ability level is an important modifying factor in relation to both post-submission feedback, observational learning and model evaluation during observational learning, we also investigated the impact of students’ ability levels in relation to essay writing development. As indicated, the present study tries to tease apart the contribution of two key aspects of observational learning in relation to essay writing: role model observation and role model evaluation. In order to consider the contribution of both factors we compared the possible improvements in essay writing skills of students who saw videos of ability-matched peer models and critically evaluated these protocols with students who were only exposed to videos of ability-matched peer models. In all cases students watched videos of their peer models using think-aloud protocols whilst going through the process of preparing an essay plan. Social cognitive learning theory predicts that students who were exposed to both components of the learning approach should demonstrate greater improvements in their essay writing skills compared to students who were only exposed to ability-matched peer models. The success of this approach to learning as compared to post-submission feedback was investigated by also giving traditional post-submission feedback to both experimental groups, and by comparing these groups with a control group who only received the post-submission feedback.Finally, it follows from Brookhart’s (2001) and Bandura’s (1986, 1997) work that even if learners are ability-matched with their models underperforming students should benefit more from the observational learning approach compared to post-submission feedback than high-performing students; as we have seen underperforming students profit less from traditional post-essay submission feedback than high-performing students. If model evaluation during observational learning thus works well for both ability groups underperforming students stand to gain relatively more from this approach for overall essay performance than high-performing students. Our study also investigated a possible contribution of self-efficacy in essay writing. Bandura (1989) suggested that the differential manner in which high-performing and underperforming students react to feedback might be due to the students’ sense of self-efficacy or their belief in their ability to accomplish the goals that they have set for themselves. Self-efficacy was proposed to be a key internal influence for student motivation, and was assumed to direct choices with regard to academic activities (Bandura, 2006, Thomas, Iventosch & Rohwer, 1987, Bandura & Cervone, 1986, Schunk, 1984, Bandura & Schunk, 1981; Zimmerman & Ringle, 1981). The link between actual academic performance and reports of self-efficacy, which are normally measured through confidence ratings, has been supported in relation to a range of academic activities (see, e.g. Schunk, 2003, 2001; Crozier, 1997; Pajares, 1997, 1996, 2002). Perhaps most importantly for the current study, Pajares & Johnson (1996) investigated high school students’ levels of self-efficacy and their essay writing skills. The study revealed that children’s perception of self-efficacy had a direct effect on actual essay writing performance. Taken as a whole, the research findings support the view that a student’s level of self-efficacy for academic activity underpins the learning process by regulating behaviour, thus influencing actual performance. Chemers, Hu & Garcia (2001) suggest that self-efficacy appraisals have more predictive power for current academic performance than previous academic achievements. In summary, it is proposed that low self-efficacy is detrimental to learning because students are less likely to persevere in the face of adversity. In contrast, extremely high self-efficacy is detrimental for learning because students who are too confident will not spend enough time or energy developing their work. Finally, students who have self-efficacy that conforms to the ‘Goldilocks principle’ by being just right, estimated to be slightly above current ability level (Csikszentmihalyi, 1990), will use their beliefs to regulate their academic activity and develop effective coping mechanisms that enable them to achieve better outcomes (Schunk, 2003, Margolis & McCabe, 2006, Nilsen, 2009). Based on social cognitive learning theory we thus expected a positive relation between self-efficacy and essay writing performance. In addition, we considered whether high performing and underperforming students differed in terms of self-efficacy, what impact this had on their performance, and what the underlying factors in self-efficacy were.MethodParticipants142 first year psychology students from the School of Psychology at the University of Lincoln agreed to take part in the study. The ages of the first year participants ranged from 18 to 40 years. The participants were recruited through the first year tutorial programme, and second year models were recruited through their academic tutors. All students received credit points as part of a reciprocal research participation scheme utilised by the School of Psychology. Having enough credits allows students in their third and final year to access first and second year students to participate in their own empirical studies.MaterialsFive videos were produced for the study. These consisted of one video for the control group featuring a postgraduate student discussing the appropriate format and use of references in academic work. This content was drawn from typical comments that tutors made when feeding back to students about their essays. Four videos for the experimental groups were produced with two high-performing and two underperforming peer models discussing their planning for a semester B essay entitled “Are Visual Illusions Merely Amusing Curiosities?” The videos aimed to capture the models thinking processes as they planned out essays for the title. This approach was used as previous studies exploring observational learning proposed that in order for modelling to work it was important that learners were given insight into the strategies that students use when writing.The peer models in the experimental videos first completed online problem solving activities in order to facilitate their use of think-aloud protocols. Once the peer models were comfortable using the protocols, they were given the essay title, copies of two academic articles relevant to the topic (Gregory, 1968, 1997) and one appropriate book chapter (Coren, Ward & Enns, 1994). The peer models were asked to produce an essay plan that would demonstrate how they would approach answering the essay question. The peer models were not subjected to time limits during the task and were provided with paper in order to organise their work. Once the planning was complete (on average of 1 hour and 15 minutes), the peer models were asked to present their essay plan using think-aloud protocols. Video recordings of these presentations were made. Prompts were designed to elicit further information if required. However, a prompt was only used on one occasion. The lengths of the videos ranged from four minutes and three seconds to eight minutes and thirty-three seconds. All videos were recorded using a DVD camcorder. Only the control group video was scripted and subsequently edited to provide a coherent film. Twelve questions were used to facilitate a critical evaluation of the models in the experimental groups who saw the experimental videos and had an active discussion after seeing these videos (see appendix A for the questions and see below for a definition of the experimental groups).In order to measure self-efficacy the Academic Behavioural Confidence Scale (ABC) designed by Sander and Sanders (2006), first published as the Academic Confidence Scale (Sander & Sanders, 2003) and developed in accordance with Bandura’s self-efficacy research (Bandura, 1986, 1989), was administered in paper format (Appendix B). The ABC was chosen as it had been shown to effectively capture students’ self-efficacy by allowing students to rate their confidence in relation to twenty-four general academic activities on a five point scale, ranging from 0% (not confident at all) to 100% (very confident), with 25%, 50% and 75% as discrete intermediate values. ProcedureThe study utilised the first year tutorial system within the university that allocates students to personal academic tutors for the first year of their course. As a part of the tutorial system students submit two formative essays, one in semester A and one in semester B. The marks from these essays do not contribute to the students’ final marks for the year. The academic tutors, 16 in total, marked their own participating students’ essays in both semesters. No guidance on marking was given to the tutors but it was ensured that all tutors marked the same students on both occasions. Marking consistency was thus achieved through the same members of staff marking the same students on the two occasions. Essays were marked in accordance with the School of Psychology categorical marking scheme. The marking scheme provides guidance to tutors regarding the overall quality of academic work expected for each mark and is intended to provide a guide to help tutors balance the strengths and weaknesses of the students work. Under this scheme students are able to attain the following marks: below 5, and 15, 25, 32, 38, 42, 45, 48, 52, 55, 58, 62, 65, 68, 75, 85, and 95. Students were allocated to high-performing or underperforming groups based on the results of their semester A essay. The median essay score in Semester A (58%) was used to allocate students according to ability levels. It was proposed that students in the range of 65% - 55% could be considered to have an average ability (the smallest bandwidth around the median score according to the categorical marking scheme). Therefore students attaining above 65% were considered high performing and students below 55% as underperforming. Students captured in the average ability group were randomly allocated to either the high or low ability groups in order to preserve the variance and statistical power within the data set (please note that if anything, this would go against our hypotheses, and would not favour our hypotheses).Students were assigned to one of three experimental conditions: the control group, the video only group, and the video plus model evaluation group, using an on-line random number generator. The aim of this approach was to allow the researchers to tease apart which aspects of observational learning were necessary for it to take place, e.g. was it enough that students were given insight into the strategies used by others or did they benefit more from also discussing what they were observing with peers.Three weeks prior to the submission of the semester B essay participants were invited to attend one taught session:Control group: Students were invited to take notes whilst they watched the control group video. Students then received a tutorial demonstrating the use of electronic referencing software and electronic databases. Students completed the ABC questionnaire.Video only group: Students watched two videos of peer models matched to their own ability level (high or low). They were prompted to take notes and reflect on what they observed in the videos. Following this, students watched the control video and received a tutorial on how to use the electronic referencing software and electronic databases. Students completed the ABC questionnaire.Video plus model evaluation group: Students received the same tuition as in the video only group, but immediately following the videos students were encouraged to critically evaluate and discuss with each other what they saw on the video. The feedback session was guided by twelve questions (see appendix A). Following this, students received a tutorial on how to use the electronic referencing software and electronic databases. Students completed the ABC questionnaire.All sessions lasted approximately one hour, in order to facilitate equal amounts of time being spent with each student group the researcher provided additional tutorial support on the use of the electronic referencing software and electronic databases to students in the control group. All students were fully briefed prior to participation.The ABC questionnaire was administered immediately after the students had participated in the taught sessions above. In all cases students had received their marks and feedback from their semester A essays. Students were given a full brief of the purpose of the study, contact details of the researchers and paper copies of the questionnaire. They were asked to be as truthful as possible when completing the twenty-four item ABC questionnaire. All students were assured of the anonymity of their responses. The questionnaire took approximately five minutes to complete. Ethical approval for the study was obtained in accordance with the ethical guidelines of the University of Lincoln and the British Psychological Society. ResultsOf the 142 students who agreed to take part seventy-seven participants completed the entire study by finishing all elements: attending the taught session (N=94), completing the ABC questionnaire (N = 88) and submitting both their semester A (N = 142) and semester B essays (N = 77). Three participants were removed from the final data set for the analyses investigating the effects of the observational learning approach as their essay scores were more than 2.5 standard deviations from the median score of either semester A (median = 58%) or semester B (median = 62%) marks, leaving 74 students on which the analyses comparing the effectiveness of the observational learning approach were conducted.In order to investigate the effectiveness of the two elements of observational learning, model observation and model evaluation, in relation to post-submission feedback, and in order to detect a possible modifying influence from ability level, a 2 (time period: semester A versus semester B) x 2 (ability level: high versus low performers) x 3 (test group: control group versus video only group versus video plus evaluation group) mixed ANOVA was conducted, with time period as a within participant factor and ability level and test group as between factors. In order to reduce the chance of errors occurring in the interpretation of the analysis, the data was checked for equality of variance and corrected figures were used. There were significant main effects of time period (F (1, 71) = 12.235; p = 0.001; MSE = 351.260) (semester A mean = 58.68; SD = 6.1; semester B mean = 61.39; SD = 7.7) and of ability level (F (1, 53) = 47.249; p = 0.000; MSE = 2017.08). There was also a significant interaction effect between time period and ability level (F (1, 71) = 16.356; p = 0.000; MSE = 469.570). Bonferroni post-hoc tests revealed that low ability students significantly increased their essay scores over time (semester A mean = 53.559; SD = 4.37; semester B mean = 60.294; SD 6.51), while high-performing students essay scores remained the same over time (semester A mean = 62.74; SD = 3.62; semester B mean = 62.256; SD = 8.44). There was no main effect for the test group (F (2, 71) = 0.262; p = 0.770; MSE = 7.515) nor were there any interaction effects with the factor test group, indicating that none of the experimental manipulations explained the results (see Table 1 for mean scores).[Insert table one here ]In summary, the results demonstrated no advantage for observational learning over traditional post-submission feedback, and it was not possible to disentangle the relative contribution of model observation and model evaluation for pre-submission feedback. The results did, however, demonstrate that underperforming students improved their essay scores over time while high-performing students’ scores remained the same over the two time periods.The results from the ABC questionnaire were analysed (1) in order to determine the factors that constitute self-efficacy as measured by the questionnaire, (2) in order to determine whether – and if so how – high-performing and underperforming students differed in self-efficacy, and (3) in order to determine whether semester B marks could be predicted on the basis of these confidence scores. All eighty-eight students who had completed the questionnaire and submitted their semester A essay were included in this analysis.In order to determine the factors that constituted self-efficacy an exploratory principal component factor analysis of the questionnaire data was conducted. Initial results showed that the questionnaire had a high level of internal consistency (0.892, Norusis, 2005).After making corrections in accordance with Dancey and Reidy (2004), e.g. suppressing lower value correlations, the results showed that six components provided explanations for correlations between the items within the questionnaire. All six components had values at the acceptable level (Kaiser, 1960) and explained the following amounts of variance: component one 31.54%, component two 9.55% (41.09% cumulative), component three 8.75% (49.84% cumulative), component four 6.06% (55.89% cumulative), component five 4.99% (60.88% cumulative) and component six 4.4% (65.28% cumulative). The first three components offered a stable fit for the questionnaire and also avoided repetition of the questionnaire items within the factors. The three factors were identified as “achievement and motivation”, “understanding and preparation”, and “verbal skills”. Table 2 lists the individual questions for each component.[Insert table two here ]With the three main components identified it was possible to determine whether – and if so how – high-performing and underperforming students differed from each other in terms of self-efficacy. As our interest was purely focused on the differences between the high and underperforming students we removed the middle ability group data (N = 18) from our data set in order to complete these analyses. One-way ANOVAs were carried out on the confidence scores for each of the twenty four questions of the ABC questionnaire, with high ability group (N =38) versus low ability group (N = 32) as a between participant factor. Table 3 presents the questions for which the self-efficacy ratings differed between high and underperforming students (only significant results of p < .05 are reported), and links these differences to the factors revealed during the factor analysis.[Insert table three here]Excluding the questions to which no factor could be assigned suggested that high and underperforming students differed in their confidence ratings in relation to “understanding and performance” (N = 5) as opposed to other factors (N = 3). A Chi-square test, however, showed that this difference was not significant (p < .48). So, although there were differences in self-efficacy between high and underperforming students the questionnaire used did not make it possible to attribute possible causes of these differences to a particular factor although it was possible to attribute the possible ‘causes’ to a set of issues identified in Table 3. These results suggested that a complicated relationship exists between student performance and the skills necessary for essay writing.In order to explore the confidence patterns of the students further we grouped the students’ according their highest factor loading. Table 4 shows the typicality of confidence patterns for both high and underperforming students. In order to compare whether any of the three factors was more prevalent amongst the ability groups chi-square tests were conducted between each of the three factors (motivation, understanding and preparation, and verbal skills). The results of these analyses revealed that for low performing students there were no significant differences between the confidence patterns. However, the results for the high performing students suggested that factor loadings for understanding and preparation were significantly lower to both achievement and motivation (x2 = 6.368; p < 0.0116) and verbal skills (x2 = 9.783; p < 0.0018). [Insert table four here ]Finally, we considered whether the responses to the ABC questionnaire in semester A could be used to predict actual essay writing performance in semester B. A stepwise variable selection multiple linear regression analysis was conducted on the full sample of 74 student responses. The results revealed that a total of four items from the questionnaire contributed to the prediction of semester B marks: “confidence in the ability to attend tutorials” (unstandardised coefficient = 3.692; SE = 1.358; beta = .301; t = 2.720; p = 0.008), “confidence in the ability to read the recommended background materials” (unstandardised coefficient = -3.360; SE = 0.991; beta = -.384; t = -3.391; p = 0.001), “confidence in the ability to write in an academic style” (unstandardised coefficient = 2.626; SE = 1.139; beta = .256; t = 2.306; p = 0.024) and “confidence in the ability to understand the materials outlined and discussed by lecturers” (unstandardised coefficient = 2.655; SE = 1.243; beta = .229; t = 2.136; p = 0.036). R for regression was significantly different from zero (F (4, 71) = 6.421; p = 0.000) (R (.526), R? (.277) and R? adjusted (.234)), and 28% of the variability in the essay marks was predicted by the four questions. The four questions could not be assigned to one factor discovered by the factor analysis (two questions could not be assigned to any factor, one question could be assigned to “understanding and preparation”, and one question to “achievement and motivation”). Finally, as is demonstrated by the results presented in Table 2 high-performing and underperforming students only differed significantly on two of the four questions: having confidence in the ability to attend tutorials and having confidence to understand the materials.In summary, a regression analysis in relation to self-efficacy scores demonstrated that four questions can be used reliably to predict semester B marks, however, there is not a single underlying factor explaining this prediction, and high-performing and underperforming students do not differ systematically in the way that they had answered these four questions.DiscussionIn summary, the results drawn from this study demonstrate that in contrast to previous research that has supported the utility of the modelling behaviour method for improving writing skills in children (Braaksma et al, 2001, 2002, 2004, 2006), this study did not reveal any effects of pre-submission feedback on essay writing performance. Furthermore in relation to self-efficacy the results support the proposal that there is interaction between students’ levels of self-efficacy for academic activities and actual essay writing performance (Bandura 1997; 2006; Nilsen, 2009). The differences identified between high and underperforming students’ ratings of confidence on the twenty-four items of the questionnaire support the view that confidence ratings are mediated by actual ability level and revealed interesting profiles that may inform on the qualitatively different ways in which these two groups of students approach their work, with underperforming students focusing their attention on outcomes and high performing students focusing their attention on the underpinning skills that support academic achievement such as understanding materials.In contrast to previous research in relation to high-school students (Braaksma et al, 2001, 2002, 2004, 2006; Fidalgo, Torrance, Rijlaarsdam, van den Bergh & ?lvarez, 2015), this study did not reveal any effects of observational learning, with or without model evaluation, on essay writing performance. Several factors may have played a role in mediating the effectiveness of observational learning with University students. It is possible that this study has revealed an underlying developmental trajectory, namely that learning by observation decreases with age. Alternatively, it is possible that the ethos of the higher education environment, where students are encouraged to demonstrate independence in their work, conflicts with the social cognitive learning approach. Given that observational learning has been successful in relation to a range of academic subjects and a range of age groups (Bandura, 1986, 1997; Rosenthal & Zimmerman, 1978; Sonnenschein & Whitehurst, 1984; Couzijn, 1995, 1999; Schunk, 1987, 1991, 1998; Schunk & Hanson, 1985, 1989a, 1989b; Zimmerman & Kitsantas, 2002; Schunk & Zimmerman, 2007) both of these interpretations seem highly unlikely.The most likely reason for an absence of any difference between observational learning and post-submission feedback on essay writing performance in semester B is that the impact of post-submission feedback was strong enough to overshadow any possible effects of the observational learning approach. It can be safely concluded that a lack of any difference between observational learning, either with or without model evaluation, and post-submission feedback cannot be due to observational learning being as strong as post-submission feedback, or stronger. If that were the case, the cumulative effects of model evaluation during observational learning and post-submission feedback would have made the performance of the video plus evaluation group better than the control group. This was not the case. The fact that all experimental groups and the control group received post-submission feedback, and that no difference could be found between any of these groups can only be explained by the impact of post-submission feedback overcoming an influence of observational learning. Finally, it was predicted that underperforming students should profit more from observational learning compared to high-performing students – if both groups would improve. Although underperforming students’ results improved more compared to high-performing students, this effect was not modified by any of the experimental conditions, suggesting that underperforming students profit more from the experience of essay writing – independently of whether observational learning is available – compared to high-performing students.The analyses investigating the structure of the ABC questionnaire revealed that a three factor solution offered the most stable fit for the items included within this instrument. It is noteworthy that although these findings are not entirely consistent with the proposed factors identified in Sander and Sanders’ (2003) original work this particular instrument has been subject to further development including the removal of some items. This further development of the questionnaire occurred subsequent to the completion of the study discussed here (for example, Sander, de la Fuente, Stevenson and Jones, 2011) and therefore it can be proposed that further work needs to be undertaken in order to draw these results into coherence, however, what is clear is that this factor analysis solution was appropriate with this group of students. The analyses investigating self-efficacy appeared to have found support for the claim made by social cognitive learning theory that self-efficacy has a positive relationship with essay writing performance (Bandura 1997; 2006; Nilsen, 2009). Our regression analysis showed that semester B marks could be predicted reliably on the basis of self-efficacy scores obtained in semester A. However, in contrast to what is assumed by social cognitive learning theory our factor analysis revealed that apart from “achievement and motivation” (Bandura 1997; 2006, also Schunk, 2003, Margolis & McCabe, 2006, Nilsen, 2009) confidence in the following areas was also important: “understanding and preparation”, and “verbal skills” and were also linked to self-efficacy. The factors related to self-efficacy in this case are thus not limited to “motivation”, but also appear to be based on individual coping styles and traits. What is more, the factors that we identified as being linked to self-efficacy could not be used to predict essay writing performance. Nor was it possible to explain the differences between high-performing and underperforming students in relation to self-efficacy on the basis of the factors identified by our factor analysis. Finally, the four questions measuring self-efficacy that predicted performance in general or the difference between high-performing and underperforming students in particular could not be assigned to a single factor identified during our factor analysis. So, although the positive relationship between self-efficacy and essay writing performance is congruent with social cognitive learning theory, the factors we found to define self-efficacy in this context and the complex influence of self-efficacy factors on performance were not directly predicted by social cognitive learning theory.In this study we used the ABC questionnaire (Sander & Sanders, 2006) to measure self-efficacy. This questionnaire taps into self-efficacy through measuring confidence in relation to a number of academic tasks and in relation to learning in general terms. It is proposed that such a general measure is useful not only because some items are able to predict academic performance in relation to essay writing, but also because other items may be able to predict performance on other academic tasks (e.g. presentation skills). It is proposed that further work could aim to explore which other academic skills this measure can predict. Whilst the results of this work did not support predictions made by social learning theory, a number of factors were identified as having an impact on students’ development of essay writing skills: ability in interaction with practice, feedback and self-efficacy. The data contained in Table 3 is particularly useful for developing ideas that may help to support students and therefore the following recommendations for supporting students are proposed.RecommendationsFirstly, students, particularly those that are underperforming, do benefit from practicing essay writing skills and therefore it is important to include opportunities for formative assessment in this area. Such opportunities support students in the development of both writing skills and their confidence to undertake such work effectively.This research has helped to highlight the importance of post-submission feedback, this does help students when they are developing their essay writing skills and therefore it is important for tutors to continue the good practice that already exists by providing detailed feedback on students work and opportunities for students to discuss this.The issues related to self-efficacy suggest that one approach to supporting students’ development of skills may be achieved through mentoring programs. For example, underperforming students may benefit from learning important skills such as managing their workload, techniques to improve their understanding of the materials that they are reading and the skills needed to follow themes and debates. Such mentoring may not only help students to manage their workload more effectively but it may also help to raise students’ self-efficacy in this area and as our research shows students’ self-efficacy can be linked to their actual academic performance. Furthermore underperforming students may also benefit from mentoring that encourages them to focus more on the ‘input’ aspects of academic work (e.g. reading and understanding materials) rather than the end goals such as achieving good marks.Finally, Table 3 highlights that attendance, or at least confidence in the ability to attend both taught sessions and tutorials are important aspects for high performing students and are areas that these students feel more confident in; therefore it is proposed that underperforming students should be encouraged to attend more frequently and are supported in obtaining realistic confidence levels, as confidence can be expected to have a positive impact on essay writing.In conclusion, this research revealed the highly complex nature of learning essay writing skills. It revealed that such factors as ability in interaction with practice, feedback and self-efficacy influenced a student’s essay writing performance. Furthermore, it also revealed that traditional post-submission feedback as opposed to pre-submission feedback plays an important role in University students’ development. ReferencesBandura, A. (1986) Social foundations of thought and action: A social cognitive theory. Englewood, NJ: Prentice Hall.Bandura, A. (1989) Social Cognitive Theory, in R. Vasta (Ed.), Annals of Child Development. Vol 6. Six theories of child development (pp.1-60). Greenwich, CT:JAI Press.Bandura, A. (1997) Self-efficacy. The exercise of control. New York: W. H. Freeman and Company.Bandura, A. (2006) Guide for constructing self-efficacy scales. In F. Pajares & T. Urdan (Eds.), Self-efficacy beliefs in adolescents (pp. 307-337). USA: IAP-Information Age Publishing Inc.Bandura, A., & Cervone, D. (1986) Differential engagement of self-reactive influences in cognitive motivation. Organizational Behaviors and Human Decision Processes, 38, 92-113.Bandura, A., & Schunk, D. H. (1981) Cultivating competence, self-efficacy, and intrinsic interest through proximal self-motivation. Journal of Personality and Social Psychology, 41, 586-598.Black, P., Harrison, C., Lee, C., Marsahll, B., & Wiliam, D. (2003) Assessment for Learning: Putting it into practice. Berkshire: Open University Press.Braaksma, M. A. H., Van den Bergh, H., Rijlaarsdam, G., & Couzijn, M. (2001) Effective learning activities in observation tasks when learning to write and read argumentative texts. Journal of Psychology of Education, 1, 33-48.Braaksma, M. A. H., Van den Bergh, H., Rijlaarsdam, G., & Couzijn, M. (2001) Effective learning activities in observation tasks: a post-hoc analysis. European Journal of Psychology of Education, 1, 33-48.Braaksma, M. A. H., Rijlaarsdam, G., & Van den Bergh, H. (2002) Observational learning and the effects of model-observer similarity. Journal of Educational Psychology, 94(2), 405-415.Braaksma, M. A. H., Rijlaarsdam, G., Van den Bergh, H., & Van Hout-Wolters, B. H. A. M. (2004) Observational learning and its effects on the orchestration of the writing process. Cognition and Instruction, 22(1), 1-36.Braaksma, M. A. H., Rijlaarsdam, G., Van den Bergh, H., & Van Hout-Wolters, B. H. A. M. (2006) What observational learning entails: a multiple case study. Educational Studies in Language and Literature, 6(1), 31-62.Brookhart, S. M. (2001) Successful students’ formative and summative uses of assessment information. Assessment in Education, 8(2), 153-169.Brown, S. (2004) Assessment for Learning. Learning and Teaching in Higher Education, 1, 81-89.Coren, S., Ward, L. M., & Enns, J. T. (1994) Sensation and Perception (Fourth Edition) Fortworth: Harcourt Brace College Publishers.Cooper, D. (2006) Talk About Assessment: Strategies and Tools to Improve Learning. Toronto, ON: Thomson Nelson.Couzijn, M. (1995) Observation of writing and reading activities. Effects on learning and transfer. Unpublished dissertation. University of Amsterdam.Couzijn, M. J. (1995) Learning to write by observation of writing and reading processes; effects on learning and transfer. Learning and Instruction, 9(2), 109-142.Couzijn, M., & Rijlaarsdam, G. (1996) Learning to read and write argumentative text by observation. In G. Rijlaarsdam, H. van den Bergh, & M. Couzijn (Eds.), Effective teaching and learning of writing. Current trends in research (pp.253-273). Amsterdam: Amsterdam University Press.Dancey, C. P. , & Reidy, J. (2004) Statistics without Maths for Psychology (Third Edition). Harlow: Pearson Prentice-Hall.Fidalgo, R., Torrance, M., Rijlaarsdam, G., van den Bergh, H., & ?lvarez, M. L. (2015). Strategy-focused writing instruction: Just observing and reflecting on a model benefits 6th grade students. Contemporary Educational Psychology, 41, 37-50.Gibbs, G., & Simpson, C. (2004) Conditions under which assessment supports students’ learning. Learning and Teaching in Higher Education, 1, 3-31.Gregory, R. L. (1968) Perceptual illusions and brain models. Proceedings of the Royal Society of London. Series B, Biological Sciences, Vol. 171, No. 1024, A Discussion on the Logical Analysis of Cerebral Functions (Dec. 13, 1968), 279-296.Gregory, R. L. (1997) Knowledge in perception and illusion. Philosophical Transactions of the Royal Society of London: Biological Science, 352, 1121-1128.Higgins, R., Hartley, P., & Skelton, A. (2002) The conscientious consumer: reconsidering the role of assessment feedback in student learning. Studies in Higher Education, 27(1), 53-64.Kaiser, H. F. (1960) The application of electronic computers to factor analysis. Educational and Psychological Measurement, 20(1), 141-151.Margolis, H. , & McCabe, P. P. (2006) Improving self-efficacy and motivation: what to do, what to say. Intervention in School and Clinic, 41(4), 218-227.Nilsen, H. (2009) Influence on student academic behaviour through motivation, self-efficacy and value-expectation: an action research project to improve learning. Issues in Informing Science and Information Technology, 6, 545-556.Norusis, M. J. (2005) SPSS 14.0 Statistical Procedures Companion. Harlow: Prentice Hall.Pajares, F. (1996) Self-efficacy belief in academic settings. Review of Educational Research, 66(4), 543-578.Pajares, F., & Johnson, M. J. (1996) Self-efficacy beliefs in the writing of high-school students: A path analysis. Psychology in the Schools, 33, 163-175.Pajares, F., & Schunk, D. H. (2002) Self and self-belief in psychology and education: an historical perspective. In J. Aronson (Ed.), Improving Academic Achievement. New York: Academic Press.Pajares, F. (2003) Self-efficacy beliefs, motivation, and achievement in writing: a review of the literature. Reading & Writing Quarterly, 19, 139-158.Rijlaarsdam, G., Braaksma, M., Couzijn, M., Janssen, T., Kieft, M., Broekkamp, H., & van den Bergh, H. (2005) Psychology and the teaching of writing in 8000 and some words. BJEP Monograph Series II, Number 3 - Pedagogy - Teaching for Learning, 1(1), 127-153. Rijlaarsdam, G., Braaksma, M., Couzijn, M., Janssen, T., van Steendam, E., Toorenaar, A., & van den Bergh, H. (2008) Observation of peers in learning to write: practise and research. Journal of Writing Research, 1(1), 53-83.Rosenthal, T. L., & Zimmerman, B. J. (1978) Social Learning and cognition. New York: Academic Press.Sadler, D. R. (1998) Formative assessment : revisiting the territory. Assessment in Education, 5(1), 77-84.Sander, P., & Sanders, L. (2003) Measuring confidence in academic study: a summary report. Electronic Journal of Research in Educational Psychology and Psychopedagogy, 1(1), 1-17.Sander, P., & Sanders, L. (2006) Understanding Academic Confidence. Psychology Learning and Teaching, 12(1), 29-39.Sander, P. de la Fuente, J., Stevenson, K., & Jones, T. (2001) A validation of the Academic Behavioural Confidence Scale with Spanish Psychology Students. Psychology Learning and Teaching, 10(1), 11-24. Sanders, L., & Sander, P. (2007) Academic Behavioural Confidence: A comparison of medical and psychology students. Electronic Journal of Research in Educational Psychology, 13:5(3), 633-650.Schunk, D. H. (1984) The self-efficacy perspective on achievement behavior. Educational Psychologist, 19, 199-218.Schunk, D. H. (1987) Peer models and children’s behavioural change. Review of Educational Research, 57, 149-174.Schunk, D. H. (1991) Learning theories: An educational perspective. New York: Merill.Schunk, D. H. (1998) Teaching elementary students to self-regulate practice of mathematical skills with modeling. In D. H. Schunk & B. J. Zimmerman (Eds.), Self-regulated learning, from teaching to self-reflective practice (pp. 137-159). New York: The Guilford Press.Schunk, D. H. (2001) Social cognitive theory and self-regulated learning. In B. J. Zimmerman & D. H. Schunk (Eds.), Self-regulated learning and academic achievements: Theoretical perspective (2nd Edition., pp. 125-151).Mahwah, NJ: Erlbaum.Schunk, D. H. (2003) Self-efficacy for reading and writing: influence of modelling, goal setting and self-evaluation. Reading & Writing Quarterly, 19, 159-172.Schunk, D. H., & Hanson, A. R. (1985) Peer models: Influence on children’s self-efficacy and achievement. Journal of Educational Psychology, 77, 313-322.Schunk, D. H., & Hanson, A. R. (1989a) Influence of peer-model attributes on children’s beliefs and learning. Journal of Educational Psychology, 81, 431-434.Schunk, D. H., & Hanson, A. R. (1989b) Self-modeling and children’s cognitive skill. Journal of Educational Psychology, 81, 155-163.Schunk, D. H., & Zimmerman, B. J. (2007) Influencing children’s self-efficacy and self-regulation of reading and writing through modeling. Reading & Writing Quarterly, 23, 7-25.Sonenschien, S., & Whitehurst, G. J. (1984) Developing Referential Communication. Child Development, 55, 1936-1945.Thomas, J. W., Iventosch, L., & Rohwer, W. D. (1987) Relation-ships among student characteristics, study activities, and achievement as a function of course characteristics. Contemporary Educational Psychology, 21, 19-41.Zimmerman, B. J., & Kitsantas, A. (2002) Acquiring writing revision and self-regulatory skill through observation and emulation. Journal of Education Psychology, 94(4), 660-668.Zimmerman, B. J., & Ringle, J. (1981) Effects of model persistence and statements of confidence on children’s efficacy and problem solving. Journal of Educational Psychology, 73, 485-493.Appendix AThe twelve questions generated for discussion and to prompt peer feedback in the video plus evaluation condition.?Comparing the two models, how well do you think they both addressed the essay question??How would you do this??What do you think about the general essay plans that the models generated??How would you do this??What did you think about the model’s plans for their introduction??How would you do this??What did you think of the model’s plans for the main body of their essays??How would you do this??What did you think of the model’s plans for the conclusions to their essays??How would you do this??What are your ideas about how the models used referencing??How would you do this?Appendix BSander and Sanders (2006) Academic Behavioural Confidence Scale (ABC)Please rate your confidence on the five point scale (100% = very confident – 0% = not confident at all) for the following 24 questions as honestly as you can.How confident are you that you will be able to:1.Study effectively on your own in independent / private studyVery Confident Not confident at all100%75%50% 25%0%2.Produce your best work under exam conditionsVery Confident Not confident at all100%75%50% 25%0%3.Respond to questions asked by a lecturer in front of a full lecture theatreVery Confident Not confident at all100%75%50% 25%0%4.Manage your workload to meet coursework deadlinesVery Confident Not confident at all100%75%50% 25%0%5.Give a presentation to a small group of fellow studentsVery Confident Not confident at all100%75%50% 25%0%6.Attend most taught sessionsVery Confident Not confident at all100%75%50% 25%0%7.Attain good marks in your workVery Confident Not confident at all100%75%50% 25%0%8.Engage in profitable debate with your peersVery Confident Not confident at all100%75%50% 25%0%9.Ask lecturers questions about the materials that they are teaching, in a one to one settingVery Confident Not confident at all100%75%50% 25%0%10.Ask lecturers questions about the materials that they are teaching, during a lectureVery Confident Not confident at all100%75%50% 25%0%11.Understand the materials outlined and discussed with you by lecturersVery Confident Not confident at all100%75%50% 25%0%12.Follow the themes and debates in lecturesVery Confident Not confident at all100%75%50% 25%0%13.Prepare thoroughly for tutorialsVery Confident Not confident at all100%75%50% 25%0%14.Read the recommended background materialVery Confident Not confident at all100%75%50% 25%0%15.Produce coursework at the required standardVery Confident Not confident at all100%75%50% 25%0%16.Write in an appropriate academic styleVery Confident Not confident at all100%75%50% 25%0%17.Ask for help if you don’t understandVery Confident Not confident at all100%75%50% 25%0%18.Be on time for lecturesVery Confident Not confident at all100%75%50% 25%0%19.Make the most of the opportunity of studying for a degree at UniversityVery Confident Not confident at all100%75%50% 25%0%20.Pass assessments at the first attemptVery Confident Not confident at all100%75%50% 25%0%21.Plan appropriate revision schedulesVery Confident Not confident at all100%75%50% 25%0%22.Remain adequately motivated throughoutVery Confident Not confident at all100%75%50% 25%0%23.Produce your best work in coursework assignmentsVery Confident Not confident at all100%75%50% 25%0%24.Attend tutorialsVery Confident Not confident at all100%75%50% 25%0%Table 1. The mean and standard deviation scores for the two ability group (low and high) for both semester A and semester B essay scores across the three experimental conditions investigated in the study (control group, video only group and video plus model evaluation).AbilityGroupMeanStd. DeviationNSemester A mark LowControl53.862.4414Video only53.884.328Video plus model evaluation54.602.6310 HighControl62.812.8316Video only62.405.5410Video plus model evaluation62.753.1316Total 74 Semester B mark LowControl61.147.0314Video only60.138.568Video plus model evaluation59.304.9210 HighControl62.068.1816Video only59.906.3410Video plus model evaluation62.508.3716Total 74Table 2. The individual questions of the Academic Confidence Scale loading on the main three components in the factor analysis and their respective values.Self-Efficacy QuestionsComponent 1Achievement and MotivationComponent 2Understanding and PreparationComponent 3Verbal SkillsProduce work at a required standard.798Write in an appropriate academic style.738Pass assessments at the first attempt.714Attain good marks in your work.596Remain adequately motivated throughout.414Follow the themes and debates in lectures.744Manage your workload to meet coursework deadlines.455Understand the materials outlined and discussed with you by lecturers.655Ask for help if you don’t understand.410Make the most of opportunity of studying for a degree at University.617Prepare thoroughly for tutorials.565Respond to questions asked by a lecturer in front of a full lecture theatre.817Give a presentation to a small group of fellow students.780Engage in profitable debate with peers.732Ask lecturers questions about the materials that they are teaching during a lecture.714Table 3. The questions for which high and underperforming students differed significantly (p < .05) in their self-efficacy scores, together with the means, SE’s and factors corresponding to these questions. UP refers to “understanding and preparation”, AM to “achievement and motivation”, and VS to “verbal skills”.Question – How confident are you that you will be able to:Ability GroupMeans and SE’s FactorManage your workload to meet coursework deadlinesHigh 70.39% (3.38)UPUnderperforming31.3% (4.2)Attend most taught sessionsHigh92.11% (2.13)Underperforming60.16% (5)Ask lecturers questions about the materials that they are teaching in a one to one settingHigh73.37% (3.75)VSUnderperforming56.25% (3.89)Understand the materials outlined and discussed with you by lecturersHigh70.4% (2.47)UPUnderperforming35.94% (3.54)Follow the themes and debates in lecturesHigh74.3% (2.58)UPUnderperforming63.28% (3.36)Be on timeHigh88.16% (3.86)Underperforming73.44% (4.6)Produce your best work in coursework assignmentsHigh72.37% (3.09)Underperforming57.81% (3.79)Attend tutorialsHigh91.45% (1.95)Underperforming66.4% (3.48)Produce your best work under exam conditionHigh52.63% (3.5)Underperforming61.72% (2.96)Respond to questions asked by a lecturer in front of a full lecture theatreHigh30.92% (3.7)VSUnderperforming41.41% (3.8)Attain good marks in your workHigh63.82% (2.1)AMUnderperforming 81.25% (3.18)Ask lecturers questions about the materials that they are teaching during a lectureHigh32.89% (4.4)UPUnderperforming73.43% (4.88)Make the most of the opportunity of studying for a degree at UniversityHigh80.92% (3.19)UPUnderperforming91.41% (2.13)Table 4. The distribution of high and underperforming students highest factor loadings from the three factors identified by the principle components factor analysisHighest Factor LoadingNumber of Underperforming StudentsNumber of High Performing StudentsAchievement and Motivation1019Understanding and Preparation94Verbal Skills1315 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download