153

Chapter 14

A collaborative approach to improving academic honesty

Mark Freeman, Henriikka Clarkeburn and Lesley Treleaven

Faculty of Economics and Business

Academic dishonesty is widely acknowledged in universities as a worsening trend, attributed to an expansion of the internet (Underwood & Szabo, 2003), increased class sizes and decreased personal contact, more reliance on the international student market (Ashworth, Bannister & Thorne, 1997), greater student diversity (Lambert, Ellen & Taylor, 2006), higher tuition costs (Sheard, Markham & Dick, 2003) and increased competition for employment (Underwood & Szabo, 2003). However, despite the strong interest in the public media and across the sector in Australia, Marsden, Carroll and Neill (2005, p. 8) observe that ‘there is no empirical evidence to support the popular contention that dishonesty is on the rise.’ They provide a valuable contribution of self-reported dishonesty. Nevertheless, Jocoy and DiBiase (2006) observe that self-reported figures bear little resemblance to actual cases detected. Lambert et al. (2006) point out a further conundrum: while 5.8% of students report being caught for academic dishonesty, those same institutions’ official records indicate a detection rate of only 0.2%. Clearly academics are choosing to deal with dishonesty through informal channels.

Two critical incidents prompted action on academic dishonesty in the Faculty of Economics and Business at the University of Sydney. First, in 2002 a senior peer review team visiting the Faculty noted that academic dishonesty was a ‘serious problem’. Second, the 2002 Student Course Experience Questionnaire (SCEQ) revealed that as many as 5% of students in one postgraduate program perceived dishonesty as a serious concern.

This chapter reports on a social constructivist approach to addressing these concerns about academic honesty. The research here describes the iterative, evidence-based processes undertaken by the Faculty, over a four year period, developing what has become a holistic strategy to focus strongly on prevention and education. The approach thereby aligns with Park’s (2004) call for institutional responses to protect credibility and reputations in ways that reflect specific cultural contexts. Further, the orientation is also consistent with McCabe (2005) who urges proactive strategies that build a community of trust, where ethical behaviour is valued and academic integrity is the expected behaviour.

With this educative institutional approach to promoting academic honesty in contrast to detecting plagiarism by the individual student, the following section reviews the academic honesty literature. The social constructivist framework and emergent methodology is then described. The three cycles of collaborative action research, employed to systematically address the problem, follow with discussion. Finally, the implications of this approach for change and future directions are outlined.

Literature review

Academic dishonesty is not only considered a form of fraud but fundamentally contradicts core academic values. Students at the University of Sydney are considered to have acted in an academically dishonest manner if they have ‘presented another person’s ideas, findings or written work as his or her own by copying or reproducing them without due acknowledgement of the source and with intent to deceive the 154examiner’ (University of Sydney, 2006, p. 5). However, ambiguity and uncertainty surrounds plagiarism and free-riding. Where does using others’ words or ideas without appropriate acknowledgement become ‘literary theft’ (Park, 2004; p. 291) as opposed to simply poor paraphrasing, referencing or language? Where is the fine line that separates collusion from legitimate collaboration and peer learning? When does inadequate contribution to a group assessment become free-riding rather than simply a different type of contribution (James, McInnis & Devlin, 2002)?

There is a long history of researching academic honesty in North America. Revisiting many of the same institutions surveyed by Bower (1964) in regard to how often and why students cheated, McCabe and Trevino (1993) found that while cheating (in the 12 months preceding the survey) in tests and exams had increased over the thirty years, from 39% to 64% of students self-reporting such behaviour, plagiarism was unchanged (66% in 1964; 65% in 1993).

More recently, McCabe (2005) notes that 51% of students across a sample of over 40,000 students on 68 campuses in North America self-reported plagiarism – a reduction attributed to anonymity concerns introduced with web surveying. McCabe (2005) observes that the free-response comments suggest that this younger cohort is more lenient on both plagiarism and collusion. A perception gap also exists with their teachers who consider both far more serious. Von Dran, Callahan and Taylor (2001) reveal similar concerns, for example 40% of students regard falsifying research results and copying text without referencing the source as minor events worthy only of reduced marks. In contrast, academics felt such practices should result in expulsion and a fail grade for the assessment respectively.

The importance of the institutional context, particularly the student culture, is underlined in a number of studies. Institutions with an ‘honour code’ continue to exhibit far less self-reported cheating than those without (McCabe & Trevino, 1993). However, the institution with the lowest dishonesty level in their sample did not have a formal code but rather a strong informal obligation introduced first at orientation and reinforced thereafter. In contrast to individual differences, such as age and gender, which explain only 3% of the total variance, McCabe and Trevino (1997) show that 21% of the variation can be explained by contextual variables with peer disapproval (15.6%) far outweighing both peer behaviour (4.7%) and fraternity membership (0.8%). After 15 years of researching academic integrity, McCabe (2005, p. 30) reiterates the importance of an institutional response ‘rather than investing in detection and punishment strategies [such as] reacting to an increasing number of faculty complaints by simply subscribing to a plagiarism detection service … we would do better to view most instances of cheating as educational opportunities’.

There is no equivalent research in Australia. Whilst overseas research provides valuable background to inform institutional responses, findings cannot be generalised as the tertiary context is fundamentally different. For example, unproctored exams and honour codes have not historically been part of Australian university culture. Furthermore, there is some evidence that Australian students are less likely than American students to regard the use of ideas from a book or paper without adequate acknowledgement as cheating (Waugh, Godfrey, Evans & Craig, 1994 in Marsden et al., 2005).

Business students in North America self-report more dishonest behaviours than engineering, science and humanities students (McCabe & Trevino, 1993). However, in Australia, self-reported behaviour across twelve faculties in four Australian universities 155shows engineering students significantly more likely to cheat (41% of all students admit to cheating) than students in other disciplines (including economics/accounting) and science students more likely to plagiarise (81% of all students admitting to plagiarism) according to Marsden et al. (2005). A disturbing finding is that students further into their program self-reported more dishonesty. Their study concludes that a decision to be dishonest is a function of demographic, situational and personality variables.

Recognising that there are disciplinary differences, Sheard, Markham & Dick (2003) contrast perceptions of postgraduate (coursework) and undergraduate students studying information technology in one institutional context. They find that postgraduate students self-report less cheating, while the top three reasons for preventing cheating are the same (namely, ‘want to know what your work is worth’, ‘pride in your work’, ‘can get good marks without cheating’) and these get stronger with every year of study.

There is mounting evidence of ways forward as some approaches are discounted and others are consistently proposed. Teaching ethics has limited impact on cheating (Brimble & Stevenson-Clarke, 2005). Revising policies to emulate honour codes also has limited impact on cheating (Von Dran et al., 2001). The evidence on informing students of the rules is contradictory with Kerkvliet and Sigmund (1999) claiming an impact and Marsden et al. (2005) showing no significant impact. Where there is agreement is that ‘shifting students’ focus to achieving learning outcomes will promote behaviours that encourage them to engage in learning rather than cheating in order to achieve academic success’ (Sheard et al., 2003, p. 106). Von Dran et al. (2001) propose that staff need to be actively engaged with understanding, communicating (including modelling) as well as enforcing such behaviour. Marsden et al. (2005, p. 9) call for ‘new ways of communicating academic honesty policy.’ This chapter seeks to show how in one site such approaches have been introduced over the last four years.

Methodology

Broadly, the methodology adopted in the Faculty of Economics and Business site can be understood as action research. Action research enables participants to engage directly with others in understanding and acting on issues of concern. While action research (Lewin, 1946) has a long history in education, and especially in the professional development of teachers, its application in higher education for institutional change within university practices has been more limited (Treleaven, 2001).

Throughout the four years of focusing on improving academic honesty, there have been considerable shifts in the approaches to the academic honesty project as leadership has changed and wider understandings of how to effectively engage institutional change have been gained. In summary, these understandings allow the methodology to be reconceptualised from a single intervention to fix a problem (technical action research) to an institutional process adopting collaborative action research processes (Reason & Bradbury, 2001). Cycles of planning, acting, observing and reflecting (Kemmis & McTaggart, 2001) have spiralled outwards dynamically to seek better explanations and options for change. In each cycle, examining underlying assumptions has shifted not only the mode of action research (from technical to collaborative) but also the theoretical framework (from positivist to social constructivist), thereby redefining the scope and appropriate strategic action as Grundy (1982) highlights: 156

At base the different philosophical stances which underpin the various modes, relate to the source and scope of the guiding ‘idea’ of the project and the disposition which determines the type of strategic action (p. 353).

Further, this reconceptualisation as collaborative action research emphasises the emergent nature of the interventions and its commitment to inform subsequent cycles of action on the basis of evidence generated by research, from within and beyond the process itself.

The analyses and sense-making of the developments in this academic honesty project have drawn on substantial sources of data collected over the four-year research period. The data used can be clustered around three major types: institutional; student and staff feedback; and resource materials. Institutional data includes policy documents; working party and committee reports and minutes; resolutions; and documentary evidence provided in funding applications. Student and staff feedback includes course (SCEQ) evaluations; student and staff interviews; usability studies; student feedback relating to the introduction of the online self-paced learning module, self and peer assessment software (SPARK), and phrase-matching software; and faculty forums. Resource materials provide documentary evidence of action such as the development of online modules and websites to address the problems. Much of this data has been analysed and reported elsewhere, using quantitative analysis of data collected from the self-paced online academic honesty module, SPARK and text-matching software usage; and interviews with students and staff. (See Clarkeburn & Freeman, 2006; Freeman, Hutchinson, Treleaven & Sykes, 2006; Freeman, McGrath-Champ, Clark & Taylor, 2006; Clark & Freeman 2006.)

This chapter seeks to assemble the research findings from this data to present a systematic overview of the emergent and iterative process of organisational change in respect of academic honesty in one site. Such a summative and reflective approach may be suggestive of a model of change that has other applications beyond this problem of academic honesty.

Cycles of development

It is possible to distinguish three cycles in the development processes of academic honesty in the Faculty, consistent with three cycles of planning, acting, observing and reflecting. These are outlined below (see Figure 14.1).

Reactive first cycle

The development of academic honesty procedures commenced in 2003 with a set of polarised assumptions characterised by an ‘us and them’ mentality. Students were regarded as the main actors in academic (dis)honesty, taking ethically unjustifiable actions by plagiarising and free-riding. Although staff recognised the problem, most just wanted someone else to fix it. These positions were supported in the current literature on students engaging in plagiarism and emphasising motivations as the personal realm of the students (e.g., Franklyn-Stokes & Newstead, 1995; Norton, Tilley, Newstead & Franklyn-Stokes, 2001). Accordingly, the Faculty attempted to ‘fix the problem’ by developing singular atomistic actions to detect student dishonesty and increase faculty compliance with minimal impact on teaching staff. This early strategy can be best described as knee-jerk – seeking quick and easy ways to fix the perceived student problem through compliance. 157

Planning, acting, observing and reflecting

A diagram showing the cycles of development. Atthe top in 2004 are the students in singular atomistic actions and then in 2005 is students & academic management in multiple linked activities. In 2006 is students & academic management & staff in an integrated holistic approach. To the left are decsiptions of the Reactive First Cycle, Pragmatic Second Cycle and Integrated Third Cycle.

Figure 14.1: Cycles of development in academic honesty project

A trial of plagiarism detection software was actively pursued, consistent with observations by McCabe and Pavela (2004) of staff requests for enhanced detection methods. However, following a report on the faculty’s detection pilot, the University’s Academic Board rejected continued mandatory use on the basis of protecting students’ 158copyright and intellectual property rights. The faculty complied, albeit with some dismay and disappointment. In these circumstances, staff largely refused to entertain further non-mandatory use, noting that the very students likely to be caught would probably choose not to participate.

As a result, the main strategy was to develop a compulsory stand-alone online module focused on the areas where students were most confused (McCabe, 2003): plagiarism in written work and free-riding in group work. The module served as a way for the faculty to detect compliance by disallowing claims of unintentional plagiarism by students. To be able to track student compliance, it was necessary to fit the module into the faculty’s learning management system. It was also hoped the module might deter students by outlining consequences of dishonesty and highlighting benefits of academic honesty. Additionally, revisions were made to include the faculty policy on academic honesty in the unit of study template and to include a student pledge that work was not plagiarised on individual and group assessment cover sheets.

The online module was piloted with volunteer students in 2004. The second pilot was a compulsory part of a unit of study where students (n=297) were asked to complete a set of self-test questions about academic honesty, then study material on academic honesty, and finally answer the same questions again. The module evaluation revealed three significant pieces of information. First, less than 25% of students were able to correctly identify examples of plagiarism; second, student perceived high levels of plagiarism among their peers; third, the majority of students claimed that they were academically honest because it is ethical or beneficial to them, rather than out of fear of getting caught (Clarkeburn & Freeman, 2006). Students suggested that the module should be compulsory for commencing students. Interestingly, they also found estimating breaches of honesty in other students difficult and uncomfortable (Clarkeburn & Freeman, 2005; Clark, Freeman, Yench & Westcott, 2005).

These results provided an opportunity to reflect on the role and delivery of the academic honesty online module and other academic honesty strategies within the faculty. These reflections were shared in two faculty forums. As a result, the faculty’s academic honesty approach remained focused on the online module with recognition that support was necessary. The module’s function, therefore, shifted to an educational tool beneficial to students, rather than a method of primarily increasing compliance. First, it was made accessible for all students throughout their studies; second, it was designed so that students could freely consult information while answering the self-test questions; and third, the questions regarding peer dishonesty were removed.

Pragmatic second cycle

The pragmatic cycle of developing academic honesty in 2005 extended beyond a focus on students to include academic management. The results from the pilot studies and literature elsewhere (e.g., Carroll, 2002; Varvel, 2005) significantly altered the underlying assumptions. Students were now viewed as uninformed, but willing, participants in the promotion of academic honesty. Staff indicated their support, though many had become disenfranchised by the university’s punitive approach adopted in early 2005, with mandatory reporting requirements and very serious sanctions after a second confirmed case of dishonesty. In response, the faculty’s academic management implemented multiple linked activities to promote academic honesty among students. The main intention was to educate autonomous learners about honest academic writing and group work and to engage staff in active discussion about academic honesty. 159The faculty thus intended to maximise support for students to adopt the faculty understanding of academic honesty.

This approach led to a broader set of strategies. First, in semester 2, 2005 the module was made compulsory for all new students before submitting their first assignment. Second, the academic honesty website was expanded to facilitate further learning and made available to academic staff as well. Additions included PDF take-away summaries and step-by-step paraphrasing support. Videos of students talking about academic honesty, also used in orientation and induction activities, could be viewed online. Third, pledges relating to non-plagiarised work were extended to assessments submitted electronically. Fourth, development work and pilots were initiated for the promotion of text-matching software (SafeAssignments) as a formative tool for supporting students to check their work prior to its submission for assessment. Fifth, further development work was carried out to introduce a tool to promote confidential self and peer assessment (SPARK) to enhance academic honesty in group work.

After a year of faculty-wide use, the academic honesty online module data was analysed and three positive outcomes could be concluded. First, a 17% increase in recognition of plagiarism by students; second, students viewed breaches of academic honesty more seriously; and third, 45% of students believed the module provided them with adequate knowledge about academic honesty (Clarkeburn & Freeman, 2006).

The Student Course Experience Questionnaire (SCEQ) reinforced that progress was being made. Conducted in late 2005, it demonstrated that students perceptions of academic honesty as a problem to be addressed was no longer of concern. In comparison to 2002 when 5% of free-response comments relating to the aspect perceived most in need of improvement related to plagiarism, not a single comment was made in the 2005 surveys.

Integrated third cycle

A focus on learning as a community that included students, academic management and staff characterised the integrated third cycle. Until the start of 2006, assumptions underlying the efforts in the two earlier cycles had been based on a liberal modernist view of the autonomous individual who was, and should remain, free to make life decisions. The approach had been to provide the learner with adequate information upon which to make informed individual choice to maintain or reject academic honesty in their own work.

A new view emerged of students being part of a larger learning community in which promoting and upholding shared communal values was important (Walzer, 1984; McCabe, 2005). With students as members of an academic community, the faculty encouraged adoption of its key community values, most importantly ‘integrity and ethical practice in academic endeavours’ (University of Sydney, 2007). This approach aligns with promoting ‘ethical, social and professional understanding’; one of the University’s five key graduate attributes around which curricula are constructed.

The faculty approach is now based on an integrated holistic approach of continuous improvement involving students, staff and academic management. Students are viewed as part of the academic community, replacing the dichotomy of ‘us and them’. Academics are encouraged to take an active part by reflecting on their own assumptions relating to academic honesty and to design assessments that are less likely to enable dishonest group work or writing practices. As academics are encouraged to share their 160views with students and colleagues, a greater sense of collective responsibility can prevail. The resultant benefit to the culture is consistent with a McCabe’s (2005) village metaphor, one ripe for proactive development strategies that focus on peer accountability driving all community members’ actions.

In practice, this has meant re-writing sections of the academic honesty online module to more explicitly promote shared community values and motivate students to adopt them. A website to support staff in discouraging (and detecting) academically dishonest writing in assessments is under construction. Self and peer assessment software (SPARK) has been piloted and evaluated focusing on formative development of group work assessment within the Faculty (Freeman, Hutchinson et al., 2006), and in other sites (Willey & Freeman, 2006), rather than ex-post adjusting summative marks according to contribution. A new group work website has been launched for both staff and students to promote ethical and productive use of group work as part of the learning process. Further, opportunities for promoting positive group work practices to reduce the likelihood of free-riding have been enhanced by researching and introducing team based learning (Michaelsen, Knight & Fink, 2004; Freeman, McGrath-Champ et al., 2006). Increased effort has also been placed on embedding text-matching software (SafeAssignments). Discouraging cheating in online tests and exams, completed in computer labs, has been adopted by introducing software (Exambient) which isolates computers from the University network and prevents browsing the internet, opening other programs, and printing during online quizzes and exams.

Though these student-focused proactive strategies, including those focusing on academic development, are still in their early days compared with some of the elaborate approaches engaging students in North American universities, students are increasingly encouraged to become part of the academic community. Peer mentoring, which has become a popular component in the student transition to university, now pays attention to supporting new students entering the learning community. Orientation and induction activities specifically include students on panels responding in their own words to the importance of an appropriate approach to academic writing and group work.

Moving through these three cycles in a little over three years, the engagement with academic honesty has spiralled outwards to generate a learning community within the faculty that is transforming the initial emphasis on compliance and catching students.

Discussion

The aim of this chapter was to report on institutional learning, undertaking a collaborative action research approach within a faculty addressing academic honesty. Institutional learning problems, like those presented by academic honesty, have no off-the-shelf solutions. An appropriate approach is necessarily contextualised within each learning community and thus we have described the iterative journey taken in the Faculty of Economics and Business that reflects the nature, culture and progress towards valuing learning and teaching in this particular context (Hutchings & Shulman, 1999).

Without considering the context and seeking out appropriate leadership for strategic learning and teaching initiatives, academic honesty is conceptualised as a problem that is inevitably approached technically and atomistically, employing an information-transmission model. Furthermore, innovative leadership recognises that it cannot force change without seeking to engage the community in changing their conceptions. Translating the initiative into the relevant disciplinary language and setting can 161arguably, as Huber and Morreales (2002) have stated, move towards creating institutional change and learning by both staff and students.

The collaborative approach developed in this faculty has involved staff in design, implementation and research, thereby actively following the university’s commitment to strengthen the nexus between research and teaching. Alongside strategic working groups for research-enhanced learning and teaching, the faculty’s efforts have not only begun to facilitate a cultural change in academic honesty but also acknowledged and rewarded staff for their participation in diverse ways. Within a research-intensive environment, engaging in evidence-based decision-making with research outcomes has tangible value.

Further development within the faculty has commenced with a recent teaching improvement grant to help embed honest group work and writing practices in core units of study as part of a graduate attributes project. Other collaborative research initiatives with staff are embedding academic honesty within assessment methods, evaluating students’ approaches to research and referencing and thereby enabling better scaffolding and support for honest academic writing. Research into staff views and the faculty’s approach to academic honesty are not only part of the evidence-based approach to decision-making in the faculty but clearly direct those efforts towards building a positive, educationally-focused community of learners.

The major limitation of this research is the singular site of the study. Although our results within each cycle might not be generalisable because of the specific context, the research should be of interest to others seeking to understand the processes of creating the desirable institutional framework described by Park (2004) and McCabe (2005). Future research opportunities include better understanding the role of staff, both academic and professional, in the processes of generating an integrated holistic approach that focuses on community building and educational development of students rather than catching and punishing students.

The faculty’s integrated holistic approach is being broadened and deepened. The academic honesty module is being shared with a sister faculty at the University of Melbourne as part of an ongoing benchmarking relationship and with other faculties at the University of Sydney. Showcasing the approach with the Australian Business Development Council Teaching and Learning Network is also aimed towards achieving systemic change across the country by working with senior academic managers of Business faculties. As such, the faculty is amongst those who trust the integrity of its students (ACODE, 2005). Our learning endorses the view that ‘… if we have the courage to set our sights higher, and strive to achieve the goals of a liberal education, the challenge is much greater than simply a focus on reducing cheating’ (McCabe, 2005, p. 29).

This collaborative action research model is consistent with the theoretical framework based on a social constructivist approach to both promoting academic honesty as a positive educational practice and engaging students and staff in institutional change. Just as McCabe’s work demonstrates a transition from emphases on cheating to integrity and thence to learning, so this collaborative action research project has shifted its focus from detect, deter, and deal (Carroll, 2002) to a commitment to develop a leading, learning community of practice in the Faculty of Economics and Business. A significant feature of such a learning community centres around the core ethical values of academic honesty and is located within institutional practices that support, encourage and demonstrate such practices. 162