Investigating academic malpractice within an MBA Marketing module

This paper reports on progress and findings to date on an ongoing action research project designed to address the problem of academic malpractice, notably collusion, amongst postgraduate students on an international MBA programme. Its roots lie in 2008 when it was found that case study assignments had largely designed-out plagiarism but that collusion had taken its place with the result that 132 cases were taken to the University‟s Unfair Practices Committee. Understanding the problem was approached in several ways, including reference to existing literature, which highlighted the importance of students‟ previous cultural and educational experiences, and qualitative research drawing on students‟ views and feelings. As a result, a two-pronged strategy was adopted, one element being a zero-tolerance policy with all MBA assignments submitted to the Turnitin text matching software and wide broadcasting of the dangers and penalties resulting from malpractice. These were coupled with strengthening the induction and Study Skills elements of the programme in order to develop students‟ independent study skills and understanding and ability to comply with UK academic conventions. We conclude that the dual strategy of prevention and cure was effective, resulting in the overall rate of academic malpractice falling from a high of 34.8%, and of collusion of 26.6%, to single figures with both being maintained with subsequent cohorts. The ongoing question of how best to adapt teaching practices to suit the background and expectations of international students remains unresolved. Therefore, further research is being conducted with selected preliminary findings being presented in this paper. Introduction and background This paper explores the incidence and management of academic malpractice amongst students on a Marketing module within an international Masters in Business Administration (MBA) programme. It aims to share a UK University Management Schools‟ experiences to illustrate both why and how academic malpractice may occur and the impact of providing extra student support and the use of Turnitin software to contain the problem. However, it should be noted that this is an action research project and it is anticipated that further findings and learning will emerge over future cycles. International Journal for Educational


Introduction and background
This paper explores the incidence and management of academic malpractice amongst students on a Marketing module within an international Masters in Business Administration (MBA) programme. It aims to share a UK University Management Schools" experiences to illustrate both why and how academic malpractice may occur and the impact of providing extra student support and the use of Turnitin software to contain the problem. However, it should be noted that this is an action research project and it is anticipated that further findings and learning will emerge over future cycles.
The School"s MBA is large, recruiting over 400 students each year predominantly from Southern Asia. However, whilst this research inevitably focuses on the practices and background of this group, it is not suggested that what was found is unique to this population (as confirmed within the literature). The Marketing module is taught in semester two with support provided via the University"s Virtual Learning Environment. Students undertake an Academic Skills module, including guidance on referencing and academic malpractice in their first semester, prior to the start of the Marketing module.
Plagiarism had previously been suspected in theory based essays but was rarely pursued as no systematic detection procedures were in place (the text matching software Turnitin was available but only used for final projects/dissertations) and many staff felt that any concerns were identified at acceptably low levels. However, the new Module Leader introduced new case study based assignments (unique to each cohort) with assignment 1 (weighted at 30%) requiring both group (1a) and individual (1b) elements and assignment 2 (weighted at 70%) asking students to independently develop a marketing plan. The new assignments, in addition to offering pedagogical benefits, were hoped to have plagiarism designed-out (Carroll, 2002) as they require application rather than theory and, as they use unique case studies, no online ready-made answers exist.
However, during marking the first cohort"s (I: semester two, 2008) assignments it was clear that although plagiarism was limited, collusion (i.e., sharing work, ideas or data with a fellow student) seemed common. In order to quantify this, students were requested to retrospectively submit their work to Turnitin (and 96% did). This action was not taken lightly and it was recognised that it raised both ethical and potentially legal issues. Nevertheless, it was felt necessary if we were to understand the degree of collusion and, as one may expect, caused concern amongst both staff and students, some reiterated in this paper. Nevertheless, it did fulfil its purpose by confirming the extent of the problem with one in four submissions being found to merit formal investigation by the University.
Turnitin highlighted what appeared to be a systemic pattern of sharing work. Therefore, it was felt that further investigation, upon which this paper is based, would create a better understanding of the problem and its causes and inform future actions.

Methodology
The investigation may be seen as action research in that, as Lancaster (2005) suggests, it explicitly aims to resolve an internal, real-life problem and, as Saunders, Lewis and Thornhill (2009) highlight, is iterative, with future cycles informing our understanding and actions. However, as Saunders et al. (2009) also suggest, the research has wider application outside the School and it is hoped that others will benefit from its findings and learning. A pragmatic approach was adopted with a mixture of quantitative and qualitative techniques being applied as required (Saunders et al., 2009).
As a starting point, a review of the large body of literature, some noted below, significantly broadened our understanding of the issues and, in particular, provided important insights into the educational and cultural issues inherent within those educated in South Asia. In addition, using data derived from Turnitin, this helped to both establish the nature of the initial problem and subsequently the effect of the remedial actions (with longitudinal study tracking long-term trends). Finally, a series of in-depth interviews formed part of both the individual investigations and the qualitative aspects of the research. Subsequently, a survey was initiated in order to identify the experiences, preferred study methods and practices and attitudes in relation to academic malpractices. Whilst this will be separately reported in future papers some tentative preliminary findings are presented.

Understanding malpractice amongst students
There is a large body of literature looking at the occurrence and types of academic malpractice as well as its causes and management. Whilst it is not felt appropriate to review this in any depth, key findings relevant to the study are briefly noted. These relate to the scale of the issue, the particular concerns relating to differing global teaching styles and some differences of opinion.
The scale of the issue has been noted by a number of authors who identify considerable variation in approach. There is an identifiable differentiation between what is actual and reported with reasons why (Badge & Scott, 2008;Carroll, 2007;Carroll & Appleton, 2001;Darbyshire & Burgess, 2006;Errey, 2002;Gururajin & Roberts, 2005;Park, 2003;Tennant & Duggan, 2008).
Appreciation about styles of learning of students from different cultural backgrounds is critical for any depth of understanding and the work of Banwell (2003) and Timm (2008b) helps illuminate the reverence for the words of the teacher and the cultural value given to its replication. These are examples of rote learning still in evidence where a core text is used as the main source and the basis for assessment. Consequently, innovative styles of assessment like reflective logs and peer assessment (where there is the requirement for the authorial voice) are seen as alien.
Unsurprisingly, there is concern that students are poorly equipped given their educational background. Culwin and Lancaster (2000), Errey (2002), Johnston (2003) and Yeo (2006) recognise the border between paraphrasing and plagiarism is often unclear with Alam (2004) finding a lower perceived threshold of acceptability varying amongst students. This is endorsed by Boden and Ahearn"s (2008) discovery that students from some cultural backgrounds believe that general availability of information means that it can be replicated in their work. Errey (cited in Barrett & Malcolm, 2006) found 69% of students from a variety of worldwide destinations believed verbatim copy and paste was acceptable, particularly if the source was cited.
Gathering information about students capabilities in coping with their studies and the challenges faced is also investigated by a range of authors keen to understand approaches taken and reasons why. Here pragmatic and cynical solutions have been discovered by those aware of the purchasing of assessments, the use of software packages, and technologies including translation services (Johnston, 2003;Lancaster & Clarke, 2009). Readily available commercial vendors are easily found via a Google search (Bidgood, Hunt, Payne, & Simonite, 2007) and purchasing behaviour was reported by Johnston (2003) noting students making multiple purchases of assignments.
There is significant uncertainty about whether some students have deliberately set out to break the rules of academic practice or have simply erred through ignorance. As illustrated above, many believe that such malpractice is often due to cultural and experiential factors (Timm 2008a(Timm , 2008b. Premeditated academic malpractice clearly does occur (Park, 2003;Shepherd, 2008), but others such as Carroll (2002Carroll ( , 2007, are concerned that people routinely treat plagiarism and cheating as the same thing, a confusion supported by the findings that follow.

Investigation, remedy and outcomes
Cohort I"s retrospective submission of assignment 1b indicated that, after initial vetting, of 218 submissions, 76 (34.8%) merited formal investigation. Therefore, it was decided that assignment 2 should also be submitted to Turnitin. Additionally, the unacceptability and consequences of academic malpractice were stressed to students and support was given to explain how the checking software works and how to avoid suspicion of malpractice. Even so, the second assignment also displayed high levels of suspected academic malpractice with 44 (20.3%) meriting proceedings.
All cases were investigated under University academic regulations and resulted in 132 cases involving 93 individuals being found against the student (extra cases were discovered during the investigations). Penalties were imposed by scale of severity (ranging from a written warning or re-taking with a 40% cap to cancelling marks and re-taking both the Marketing and Academic Skills modules). The outcomes were widely promulgated in an attempt to warn subsequent cohorts. Whilst Cohort I was not able to see their Turnitin results or re-submit drafts, Cohort II (semester two 2009/10) was, as it was believed that this would help students use Turnitin developmentally to improve their academic skills (cynics also suggested that this merely improves cheating skills). Table 1 shows the Turnitin similarity ratings of over 25% for all four submissions.
As can be seen, the fall in Turnitin software similarity scores is marked across all bands with the overall rate of similarities above 25% falling from an average of 56% to 14.6% and that for ratings above 50% from 26.6% to 0.6%. However, these show overall Turnitin ratings prior to investigation and thus should not be taken as indicative of intentional academic malpractice without further analysis. Table 2 shows the decline in cases felt strong enough to be pursued after investigation of the Turnitin reports for these and the semester two 2009/10 cohort (in each case the first column is assignment 1 and the second assignment 2). The combined results of using Turnitin, the warnings and extra guidance were edifying, in that of Cohort II"s 170 submissions, the total actionable malpractice rate dropped from circa 35% to 7% for 1b and from circa 20% to 9% for 2. As the data illustrate, this fall was sustained (and to date remains so). They also illustrate the higher frequency of collusion as opposed to plagiarism with collusion being more than twice as common. The believed reasons for this are discussed in the following sections.
The nature and types of malpractice largely reflects those identified in the literature. These were that, plagiarism was largely founded upon web-based sources with common business, consultancy and educational sites and non peer-reviewed sources like Wikipedia predominating. Matches with students from other universities were also common but perhaps indicative of using common internet sources rather than copying per se. Some was simple and wholesale copy and pasting often with little concern for relevance (for example, for a holiday sector plan, the Icelandic Government"s Tourism Strategy was submitted with only minor edits). Otherwise, précis and paraphrasing illustrated the patch-writing and omission paraphrasing (anecdotally we heard that students believed that if every seventh word is changed the software will not detect it). Contract cheating and back-translation are both difficult to detect. Some assignments contained elements such as unusual grammar, vocabulary and synonyms or seemingly spurious references, all indicative of the latter. Poor referencing was also evident with bibliography-catch-all and mismatches between in-text and listed references being common. In some cases, entire bibliographies were copy and pasted and, in others, false in-text references inserted.
Collusion similarly followed variable patterns. Some comprised obvious sharing of material, itself sometimes plagiarised (indeed, two students shared the Icelandic Tourism plan). In other cases, the material was original but copied from another student, often with minor edits to context, syntax or structure. In many cases, a network of copying could be traced with several, and in one case six, students being implicated. The sharing of ideas was also commonplace (for example, similar approaches to defining and justifying market positioning or promotional activities). In all cases, both the originator and copyist were taken to be culpable.  Although Turnitin was the primary method of detection used it must be noted that it is not infallible and can only operate within the limitations of its database of material. Therefore, a degree of vigilance and experience were able to detect plagiarism such as the verbatim re-typing (or Optical Character Recognition scanning) of sections from textbooks and the use of online texts (for example Google books). Whilst a web search may identify such sources this is not always the case (especially if the material is an image rather than a text file). Nevertheless, Google searches did locate some, including a textbook which was itself plagiarised from the leading author in marketing, Philip Kotler.
Other forms of malpractice such as back-translation are also unlikely to be spotted by Turnitin although suspicions were often raised by odd vocabulary and syntax, especially if accompanied by impressive lists of references that the student is unlikely to have used. Contract cheating (vendors often promise it to be "Turnitin-safe") is similarly unlikely to be detected except by the experienced mind, raising suspicions regarding the authenticity of the work. Worryingly, there is anecdotal evidence to suggest that it is surprisingly common.

The students' reactions and explanations
As Carroll (2002) suggests, students" views were sought in order to help understand the problem and find means to prevent its recurrence. Inevitably, there was also informal feedback, notably from Cohort I students from whom there was a clear and vociferous outcry about the situation they had unexpectedly found themselves in (with implications for completion dates, costs and visas). This was particularly evident during a Cohort I field trip to Budapest when the Programme Manager was bombarded with pleas of help from shocked and confused students, who failed to understand their errors and expressed concerns regarding what many viewed as unfair entrapment by being asked to retrospectively submit work to the Turnitin software.
Formally, two main methods were used to gather student views. Firstly, three focus groups, each of seven students (a mix of Asian, European and UK), were held during Spring 2009. Secondly, 25 implicated Cohort II students were interviewed (a selfcompletion pre-interview stimulus guide asked them to think about their situation and motives but were not collected or analysed). These resulted in the gathering of much rich data which goes some way to explaining the practices and mindset of students regarding academic malpractice. Figures 1 and 2  Students' reactions to the investigation A strong common view was forthcoming from the student voices. Students believed assignments requiring reference to common sources of information (notably market research reports) meant it was inevitable that similarities would occur (legitimate duplication of relevant data was in fact excluded within the investigations). What students failed to recognise was that whilst using the same source material is acceptable, students selecting the same verbatim passages was both statistically improbable and a breach of academic practice. Confusion regarding the fine line between peer collaboration and collusion was also evident by students explaining that it was both natural and expected that more able students support the weaker ones, not realising that such a collective voice was unacceptable. Collectivism and a culture of mutual support between cohorts were also apparent with one "source" noting that he "...gave my assignment to one junior, he is my cousin... I guess he might have copied".
When directly challenged with the sharing of material, the reaction was mixed. Some claimed innocence, often stating that it must have been taken without their permission, with one student stating 'I still don't understand, if somebody else has used my work why I am being implicated? Catch the person who has done this, why should I be responsible?' That this is possible was supported by a finding that students, especially those living in the same accommodation, often share laptops and pen-drives and seem to have a lax attitude to data security. Others denied all charges, in one case, despite being taken through the evidence line by line, the student simply repeatedly claimed that it was all his own work and that any similarities to another student must be coincidental (a statistical improbability).
Anecdotal evidence of submitting false work, whether a previous students" or purchased assignment, also emerged with one informant suggesting that 40% or more of work was contract-cheating. Indeed, there is evidence of both local and UK-wide networks of students and alumni, offering writing services. Worryingly, a student social networking site openly advertises such services and an abortive attempt was made to gain concrete evidence about a student believed to be running such a network. However, no students would commit evidence to paper and the attempt therefore failed. Other factors to emerge included self-admissions of inadequacy, notably regarding their standard of English, ability to conduct research and write in the required style. Students also recognised a lack of self-directed study and critical thinking skills and a preference for theory based exams rather than the application of theory within practical assignments. Poor subject knowledge was also cited, with one student apparently missing the purpose of learning by stating that 'I have a little knowledge (sic) of the marketing sector and that may be a reason (for malpractice)'. Lack of commercial experience, knowledge and awareness (despite an entry requirement for this) was also given as a reason for the inexperienced to collude with those with work experience. Problems with obtaining material from the Virtual Learning Environment (Blackboard) and Library Resource Centre (including e-textbooks and vital Mintel and Keynote market research reports) due to having outstanding course fees were also noted as forcing reliance on those who had access to pass on relevant material (debtors are allowed to attend lectures and tutorials but are denied such access).
Some students were candid regarding themselves and their peers, with suggestions of low ability, fear of failure, laziness and lack of motivation as being causal factors. Large teaching groups and the lack of personalised lecturer support led to a view that the programme failed to adequately support and monitor students" progress. However, others suggested that some use their study visa merely as an opportunity to reside in the UK (one described it as akin to a finishing school prior to returning to India and an arranged marriage). Poor attendance and low commitment to self study was said to be exacerbated by the demands of mixing study with part-time jobs (often effectively full-time).
Interestingly, the nature and structure of the MBA programme was also held to be partially responsible with suggestions that a (perceived as) light taught timetable (with an expectation for self-directed learning) and a lack of challenge within some modules led some to believe that it lacked rigour and was not taken as seriously as study in their home country.

Students' perception and practices regarding sharing work
As collusion emerged as the predominant form of academic malpractice it was felt worthwhile to find out more about students" knowledge, attitudes and practices regarding the sharing of ideas, work and giving and receiving advice. Therefore, a voluntary survey was developed in order to collect data regarding MBA students" background, previous higher educational experiences, preferred assessment methods, knowledge and understanding of unacceptable academic practices and collusion.
The survey was piloted with the semester two 2011 cohort within a plenary session during the induction week when 21 self-completed questionnaires were gathered. Following amendments the questionnaire has to date been administered to a further three cohorts giving a sample size of 182 students (the survey is ongoing and it is intended that a full report of findings will be published in due course).
Whilst the survey gathered much data on a range of topics, only preliminary findings relating specifically to views and attitudes regarding the sharing of work and giving or receiving advice are given here. Part C asked a bank of 14 randomised questions, ten of which related to collusion. The questions used a five-point Likert scale with 1 denoting the respondent very much agrees that the practice described in the statement is OK and 5 very much disagreeing with low scores therefore indicating agreement. All statements were worded in such a way as to test the broad principle and allowed a degree of interpretation regarding the degree of help given with the only indicator of scale being the use of the phrase "… to help in any way they can" within questions C1 to C6. Table 3 shows the mean scores for the ten questions and, as can be seen, all statements received scores above the 2.5 mid-point and therefore respondents broadly felt that such practices are unacceptable. However, it is notable that several statements had means little above this and that significant numbers of students therefore agreed with the statement to a greater or lesser extent. This is illustrated by the frequency distribution of scores, shown in Table 4, where columns 1 and 2 (delineated by the heavy box) show the numbers considering each practice as broadly acceptable (n=182), most notably in relation to receiving help from others (C1/C2), giving help to others (C4) and working in groups to support weaker students (C8), where upwards of 30% of respondents felt these practices to be broadly acceptable.
A second bank of questions, D1-D10 were based on a survey by Brown (1999) where he asked whether respondents had personally performed what may be described as academic malpractices (six-point Likert scale: Frequently to Never), whether they believed others had done so (Yes/No) and whether they considered each to be OK (fivepoint Likert scale showing Not at all OK to perfectly OK). Brown"s original questions were modified to reflect the circumstances of the Table 3: Agreement with statements concerning receiving and giving help with assignments Mean scores: 1= strongly agree to 5 = strongly disagree MBA and current factors such as the use of the internet and social network sites to gather material and assistance. The results are shown in Table 5. Table 4: Agreement with statements concerning receiving and giving help with assignments Mean and frequency distribution: 1= strongly agree to 5 = strongly disagree Table 5: Incidence of self and others carrying out malpractice by type (D1/2/3/5) Self: Mean and frequency distribution: 1= Frequently to 6 = Never; Others: yes/No Questions relevant to the topic of collusion included: D1: Worked with others on an individual assignment. D3: Asked to see the work of someone who has done a similar assignment. D4: Shown your work to someone about to do a similar assignment. D5 Asked on a social networking site (e.g., Facebook) for help with an assignment.
In all cases respondents self-reported low levels of occurrence, although in all cases some individuals admitted to doing so with some frequency (indicated by scores of 1 to 3). In all cases, except for D1 regarding group working on individual assignments, over half claimed never to have undertaken the practice (D1 was 41.8%). However, respondents clearly felt that such practices are undertaken by others with between 25% and 33.2% believing that others do so.
It is aknowledged that the data here presented cannot be taken as definitive as the sample is too small to be statistically reliable and there is (albeit deliberate) ambiguity regarding the degree of help given with some perhaps viewing it to be minimal and others significant. Nevertheless it is felt that these preliminary findings go some way towards illustrating the nature of the phenomenon and its causes and, as the data set becomes larger, will become more robust.
After we considered all of the findings of the research we realised that there was more work to be done. Collusion as an outcome of attempting to reduce plagiarism is worthy of further investigation. We decided to take decisive action.

Actions taken
The findings of this research led to specific changes. These were in Induction design and content, module content, staff support systems and staff development.
The introduction of a revised Induction programme that included formative assessments specifically on plagiarism was a direct result of the experience of the collusion on the Marketing module. The Induction included specific literacy and numeracy testing for support mechanisms to be put in place by offering supplementary classes and ongoing tutorial support (provided by the University"s department supporting students from overseas) plus a revised academic Core Skills Module to be taught throughout the duration of the first semester to specifically address the difficulties experienced by these international students. This was seen as being particularly beneficial because the exercises were designed to complement existing assessments and would improve capability before embarking on the Research Methods module where assessments would contribute to overall results.
Staff training has taken place to improve greater awareness amongst the teaching team. This has taken the form of discussion sessions amongst the teaching team. The tendency is that staff have now become over-reliant on the results from the software rather than their own observations because they feel that their word is not taken as evidence and this too is being addressed by recording conversations with students when concerns about the ownership of their work is expressed by tutors.

Conclusions and further research
Our research clearly reflects and reinforces the findings of much of the literature, in particular in relation to the cultural issues (Hayes & Introna, 2005a), the lack of academic and writing skills (Timm, 2008b), the need to be alert to unintentional malpractice (Carroll, 2007a) and, it must be said, the proclivity of some to cheat (Johnston, 2003). In addition, it illustrates the efficacy of a combination of early support and guidance, a zero-tolerance policy and the use of the Turnitin text-matching software to identify and discourage malpractice (Carroll, 2002(Carroll, , 2007bTennant & Duggan, 2008;Timm, 2008b;Badge & Scott, 2008).
Particular value was found in providing earlier and more focused support designed to develop students" understanding of the UK approach to higher education and to develop research, referencing and study skills. Malpractice was addressed via plenary and workshop sessions which explained about plagiarism and collusion, clearly differentiating the latter from collaboration and provided clear guidance on what is and is not acceptable. A key element was found to be sessions showing how the Turnitin software is used and interpreted and allowing students to submit a piece of work as formative assessment with students found to be at risk of plagiarising receiving further support and counselling.
But developmental activity was not limited to students. We found some staff had an ambivalent attitude to malpractice and detection: reportage was, at best, patchy with some ignoring it and others imposing their own arbitrary penalties, sometimes just a warning. As noted in the literature, others felt students to be unwitting victims of their own academic inadequacies (Carroll, 2002;Magyar, 2009) or were reluctant to report cases as they felt that they were inconsistently dealt with (Tennant & Duggen, 2008) or just felt that it imposed too much extra work (personal communication). Furthermore, there was a combination of suspicion and lack of understanding about how the Turnitin detection software works and how to interpret its results. Therefore, an equally important part of the initiative was to remedy such attitudes and build understanding of how to use the software. This was aided by a streamlining of the investigatory process and made "non-negotiable" by a policy that all appropriate MBA assignments should be submitted to Turnitin.
Subsequently, and with support from the Vice Chancellor for Learning and Teaching and Registry, the MBA model has been further developed and is being considered for adoption across the University. Other important organisational initiatives have been the creation of an Unfair Practice Working Group and revisions to the Academic Regulations to reflect the range of malpractice observed and clarify the investigatory processes and penalties. Staff development has not been ignored with Schools ensuring staff awareness of the new regulations and policies and establishing a core of staff trained to uniformly interpret Turnitin"s findings.
However, if we wish to look for a cause rather than a cure we must look to students" educational and cultural experiences and the major impact they have on their ability to cope with the UK style, self-directed learning environment (often on top of living in a new country, away from friends and family). Whilst the preliminary survey findings cast some light on such issues and also students" understanding and attitudes to malpractice, there is a need to know and understand more. In order to do so the survey will be administered to further cohorts and will be supplemented by focus groups designed to deepen our understanding of these vital issues.
Other issues also remain and, as is the nature of action research, will be further investigated. Amongst them is the inability of Turnitin to detect some sources, particularly back-translation and contract cheating. Finally, but perhaps aptly for a marketing course, it is noteworthy but frustrating that supply has followed demand and the num-ber of online offers of contract cheating help has risen whilst the price has dropped, making purchase easily affordable to most students.
The most significant issue is to what extent it will be necessary for us to change rather than the students. Is it unrealistic for us to entirely overturn deep seated cultural factors and past academic practices with the limited time and resources we have available. We will also need to further adapt our teaching methods to the circumstances of the students and plan more pragmatic means of assessment.