SEARCH

SEARCH BY CITATION

Keywords:

  • constructive feedback;
  • plagiarism reduction;
  • tertiary students;
  • writing quality

Abstract

  1. Top of page
  2. Abstract
  3. Defining plagiarism
  4. Why do university students plagiarise?
  5. Strategies to reduce student plagiarism
  6. Method
  7. Results
  8. Discussion
  9. Acknowledgement
  10. References

As Internet resources for assignments become increasingly accessible to university students, the risk that students will misattribute this material and plagiarise it is also on the rise. To address this concern, we trialled several strategies to reduce the incidence of plagiarism and to improve writing and referencing among first-year psychology students across 10 consecutive semesters. Each semester's cohort ranged in size from 950 to 2,000 students. These strategies included the sole use of plagiarism software as a deterrent; several interactive writing and referencing exercises with feedback; and mastery quizzes on writing and referencing. The findings revealed that the most significant reduction in reported plagiarism cases occurred between the ‘no education (a deterrent-only) strategy’ and when a writing exercise with feedback strategy was introduced in the following year. Different patterns in the reduction and spread of various kinds of plagiarism cases are also discussed.

Modern students in any context have at their fingertips an unlimited source of pre-written material in the form of the Internet, such that all educational institutions now face issues relating to authorship, attribution, and plagiarism by students and academics. The diversity of policy responses by institutions suggests that we struggle to deal with these issues (Bermingham, Watson, & Jones, 2010); however, most would agree that plagiarism is undesirable, and it is a threat both to the endeavour of creating unique knowledge in universities and to the challenge of transforming students into reflective critical thinkers. It is satisfying then to see many researchers and educators attempting to address the issue with a variety of educational techniques from the use of online modules (Belter & Du Pre, 2009) to more comprehensive in-class interventions (Elander, Pittam, Lusher, Fox, & Payne, 2010). However, the relative effectiveness of each technique for reducing student plagiarism requires empirical examination. To address this issue, the current study tested a comprehensive battery of different plagiarism reduction strategies over a 5-year period.

Defining plagiarism

  1. Top of page
  2. Abstract
  3. Defining plagiarism
  4. Why do university students plagiarise?
  5. Strategies to reduce student plagiarism
  6. Method
  7. Results
  8. Discussion
  9. Acknowledgement
  10. References

We define plagiarism as the use of another author's words or ideas without appropriate acknowledgement. This means that plagiarism could result from a failure to understand how to acknowledge a source, carelessness in doing so, or a deliberate attempt to present another person's work as your own. The latter case might be termed ‘dishonest plagiarism’, whereas referencing ignorance or carelessness is often described as leading to ‘negligent plagiarism’. Perry (2010) describes a similar dimension in students he calls ‘understanding the rules of the game’ (p. 99), such that a student who has a mature understanding of the rules is potentially an outright cheat, whereas a student with only a superficial understanding is referred to as an ‘accidental infringer’. Some authors also distinguish between ‘serious’ and ‘minor’ plagiarism, based on just how much material is misappropriated as a proportion of the submitted work (e.g., Homewood, 2007, calls >25% serious). Similarly, Bennett (2005) distinguished between minor and major plagiarism as that between copying ‘a couple’ or ‘several sentences’, and copying ‘a paragraph’ or ‘a number of paragraphs’, respectively. In this article, a distinction is also made based on the actual source of the material, such that ‘person-to-person’ plagiarism occurs when students in the same course copy each other's assignments, and ‘resource plagiarism’ occurs when students copy material from set readings or the Internet in general. Understanding these distinctions is important because it is not clear whether one technique for preventing plagiarism will be able to address these diverse dimensions of plagiarism.

Why do university students plagiarise?

  1. Top of page
  2. Abstract
  3. Defining plagiarism
  4. Why do university students plagiarise?
  5. Strategies to reduce student plagiarism
  6. Method
  7. Results
  8. Discussion
  9. Acknowledgement
  10. References

There is an extant literature identifying why students plagiarise (see Bennett, 2005, for a thorough review). Ashworth, Bannister, and Thorne (1997) found that students tend to view plagiarism as a less serious form of cheating. Related to this, a UK study found that more than 50% of undergraduate students reported that plagiarising from the Internet was acceptable (Szabo & Underwood, 2004). Other research has found that students remain confused about what practices constitute plagiarism and in fact have difficulty differentiating between paraphrasing and plagiarism (Carroll, 2004; Roig, 1997). Using external material to write one's own original assignment is an essential but challenging academic skill to develop at a university. It is not sufficient to tell students not to copy other's material; academics must also furnish them with the skills to create their own unique contributions. Importantly, these attitudinal and environmental challenges can presumably be overcome and have in fact been targeted in several plagiarism reduction strategies.

Strategies to reduce student plagiarism

  1. Top of page
  2. Abstract
  3. Defining plagiarism
  4. Why do university students plagiarise?
  5. Strategies to reduce student plagiarism
  6. Method
  7. Results
  8. Discussion
  9. Acknowledgement
  10. References

Plagiarism detection software as a deterrent

One way to address the challenge of easy access to online material is with plagiarism detection software. Plagiarism software matches the percentage of student text with external sources to indicate the degree of student originality. Informing students that their work would be run through plagiarism detection software has been found to be a highly effective deterrent (Braumoeller & Gaines, 2001; Davis, 2007). However, Devlin (2003) noted that detection software takes no account of the reasons why students plagiarise, and as such, a deterrent-only approach appears more punitive than educative.

Plagiarism detection software is not only used as a deterrent but can also be used by students to self-assess their own work. For example, Davis and Carroll (2009) used plagiarism detection software (Turnitin, http://www.turnitin.com/) as the tool for providing formative feedback to students. Here, students were able to have the software detect plagiarism in several draft assignments, redraft their assignments according to the software feedback, and finally submit a version for assessment 4 weeks later. During this process, students found the software helpful in finding ‘poor’ paraphrasing and ‘patch writing’ (i.e., putting together different texts from different authors without an original voice; Davis & Carroll). Using detection software as a self-assessment tool also has its critics. The main criticism is that students are not encouraged to attempt to create an original work from the beginning if they are given extra chances. Rolfe (2011) recently evaluated such a strategy with students who used ‘Turnitin’, and despite positive staff and students' perceptions, writing quality actually fell significantly, as determined by grades, and the total number of plagiarists remained unchanged. Together, these studies suggest that detection software can be used in more than a deterrent fashion; however, what is also needed is a more proactive strategy that equips students with the knowledge and skills actually needed to write in an effective and original manner.

Educative approaches—Classroom interventions and online modules

The prevalence of eLearning in higher education settings allows for the deployment of relatively straightforward online modules to instruct students about plagiarism. Belter and Du Pre (2009) introduced an ‘academic integrity’ module that defined plagiarism using examples, with strategies to avoid them, outlined penalties, and then assessed students' understanding with an 11-item quiz. Students in the experimental group were required to obtain 100% on this assessment or withdraw from the course. Students in an earlier cohort were exposed to the usual information on plagiarism in their handbook and served as a control. Belter and Du Pre found that 25.8% of the papers in this group contained instances of plagiarism compared with just 6.5% in the group required to complete the online module. Besides the time and anxiety saved imposing penalties, the authors point out that such an approach reduces the probability that otherwise honest students will suffer because they misunderstand citation rules, and of course the clear advantage of an online module is the efficiency with which it is self-administered by students as many times as required.

As previously noted, plagiarism research (Roig, 1997) has shown that some forms may be a result of students lacking the adequate knowledge about citation techniques. Problems with citation techniques are particularly evident among first-year university students, and thus are an important cohort to examine. To address this, Landau, Druen, and Arcuri (2002) found that first-year psychology students who studied examples of plagiarised text were less likely to plagiarise when paraphrasing. Clearly, the Landau et al. classroom exercise shows that providing students with examples on paraphrasing can have positive effects on their knowledge of, and ability to avoid, plagiarism. Indeed, educating students about plagiarism at the same time as they learn about referencing appropriately is in effect teaching them how not to plagiarise negligently.

Students also need to develop an original voice and/or strengthen their authorial identity. A comprehensive program by Elander et al. (2010) implemented a promising intervention that emphasised ‘authorial identity’ in an attempt to prevent negligent plagiarism. Here, authorial identity refers to the sense a student has of himself/herself as an author, an author who can make an original contribution. They contended that students tend not to think of themselves as authors, possibly because they think that this would lead to low marks, where the alternative is to simply be seen as an editor and conveyor of other material. Such an attitude would seem not only to lead to plagiarism but also result in ineffective and unoriginal writing. So Elander et al. defined authorship for students, and they discussed the decisions an author must make, for example, about the central message, emphasis, and interpretation. Examples of different writing styles were given to students to consider, including examples of student writing that contrasted an approach involving an active author making critical decisions, and an approach where the student merely seemed to be editing or reporting on other sources. Following this, Elander et al. gave more high-profile examples of alleged plagiarism, including the so-called author's explanations of what happened. The intervention concluded with a checklist for students to determine if they actually qualified as an author of an assignment they might submit. Elander et al. found that this comprehensive series of activities results in an improvement in knowledge to avoid negligent plagiarism, particularly among first-year students. Unexpectedly, however, there was no reduction in the number of actual cases of student plagiarism. Elander and colleagues suggest that a more sustained intervention may be needed to convert student attitude change about plagiarism into behavioural change.

The current study

Because of the complex nature of plagiarism, the various types of plagiarism, and diverse motivations to plagiarise, as discussed previously, no single strategy is likely to be completely successful in reducing student plagiarism rates. More specifically, the literature suggests that a one-size-fits-all approach to plagiarism reduction is ineffective (Park, 2004), and what is needed is a multilayered, evidence-based, longitudinal strategy. Therefore, to address this, the current study extends earlier research (White, Owens, & Nguyen, 2008) to both detect and discourage plagiarism among large first-year psychology cohorts across 10 consecutive semesters, and thus is the first study to evaluate the relative and cumulative effectiveness of these various plagiarism reduction strategies. As acknowledged by both Chester et al. (2013) and Knott et al. (2013), supporting and encouraging psychological literacy among first-year students is essential good practice.

In the first intervention in semester 1 of 2007, we required students to submit their assignments online, and began using plagiarism software. Students were informed of the use of this software and reminded of the consequences of plagiarism in lectures and their course manuals. Our standard writing tutorials retained in-depth instructions on correct referencing and the nature of plagiarism. This initial attempt at reducing plagiarism could therefore be considered a ‘deterrent-only’ approach, which will henceforth be treated as the baseline control. In subsequent years, across very similar cohorts, we introduced several interactive online and in-class exercises to better educate students about writing, referencing, and plagiarism, as can be seen in Table 1, and these are described in detail in the method.

Table 1. Strategies to reduce first-year psychology students' plagiarism rates between 2007–2011
YearSemester 1 Essay (PSYC1001)Semester 2 Report (PSYC1002)
2007Plagiarism software introduced and students explicitly warned (deterrent only)Plagiarism software introduced and students explicitly warned (deterrent only)
2008Writing exercise performed in class with feedback/examples onlineOnline feedback (from writing exercise) reposted online and emphasised
2009–2010

Writing mastery quiz introduced for course marks

(this plus all previous interventions maintained in 2010)

Writing mastery quiz repeated with some modifications (this plus all previous interventions maintained in 2010)
2011All of the above plus a new writing tutorial with peer feedback on writing within classAll of the above

In terms of assessing the success of our intervention strategies, the dependent measure is the number of cases of serious plagiarism found in each cohort. We define serious plagiarism as occurring when more than 50% of a submitted work is substantially similar (verbatim or in precise meaning) to another source such as another student's assignment (referred to here as ‘person-to-person’ plagiarism), or sources available on the Internet or the references for the assignment (referred to here as ‘resource plagiarism’). Because the introduction of plagiarism detection software is both a deterrent, and an instrument for assisting in better estimating levels of plagiarism, we cannot hope to test its effectiveness by comparing rates to those plagiarism rates detected before software (pre-2007 detected rates were extremely low); nevertheless, our design allows us to assess the role of deterrence only, by looking forward to see whether that alone is sufficient. Based on our review of the literature that instructs us that plagiarism often results from ignorance and misunderstanding, we expect a large fall in rates once students become more engaged. Given that we retained all previous techniques as we added more, the success of each technique could be determined with a direct comparison with the previous year's outcomes. Because our interventions took many different forms such as special tutorials, mastery quizzes and learning modules, we hoped that, rather than creating redundancy, we might be more likely to capture all students at some point.

As well as addressing overall rates of plagiarism, we are also interested in the nature of the written assessment for which students' work for plagiarism is screened. In semester 1 each year (from 2007 to 2011), students were required to write an American Psychological Association formatted essay. Topics have generally been focused questions, rather than ‘write-about’ or ‘discuss’-type questions, to discourage students from merely summarising sources. This assignment contrasts heavily with the semester 2 assessment run each year (from 2007 to 2011) where students participate in a unique study run in tutorials upon which they are given unique results and required to write an American Psychological Association formatted research report. These two kinds of written assignments may represent different challenges for students and result in different kinds of plagiarism.

Another factor of interest in any wide-scale study of plagiarism is sex differences. Underwood and Szabo (2004) found an unexpected trend when they surveyed students about cheating, such that only 35% of males said that they would never commit an act of academic dishonesty under any circumstance, compared with 60% of females. This gap narrowed when students were asked to imagine failure as the only other option, and Underwood and Szabo (2004) do not appear to predict a difference in overall outcome, as while males were more willing to commit offences, females used the Internet more for coursework. Koul, Clariana, Jitgarun, and Songsriwittaya (2009) also found that males were more likely to justify plagiarism by needing to be correct, and they were less likely than females to justify plagiarism with excuses such as being under stress or having a heavy workload. Curiously they also found that males were more likely than females to consider copying from a ‘distant’ source (i.e., a local expert or the library) as plagiarism, and yet no sex differences were found in attitudes to copying from close associates (Koul et al., 2009). Given the mix of different assignment types and possible sources of information in our courses, it is difficult to predict any precise sex differences, but this previous research suggests strong qualitative differences in the factors that influence each sex to plagiarise.

Method

  1. Top of page
  2. Abstract
  3. Defining plagiarism
  4. Why do university students plagiarise?
  5. Strategies to reduce student plagiarism
  6. Method
  7. Results
  8. Discussion
  9. Acknowledgement
  10. References

Participants

Aggregate data comes from those enrolled in one or both of the first-year psychology courses (PSYC1001 and PSYC1002) at the University of Sydney, from 2007 to 2011. We estimate that approximately 70–80% of students each year participated in both courses, thus by the end of second semester, many students would have encountered our strategies twice.

Plagiarism detection software introduced and warnings

In general, our research design is similar to Wilson-Doenges, and Gurung's (2013) type 1 SoTl design in that quantitative student data are gathered before (in 2007) and after (2008–2011) a change in strategy is put into practice. Specifically, in 2007, we required students to submit their assignments online via their eLearning site. Basic plagiarism software (WCopyfind, http://plagiarism.bloomfieldmedia.com/z-wordpress/) then checked for matches between the assignments, and with other materials such as the set references and various websites. The plagiarism section in the course manual included the university and school policies on plagiarism, what constitutes plagiarism (including resource and person-to-person forms), penalties for plagiarising, and the use of plagiarism detection software and its effectiveness. This information on plagiarism remained in all course manuals in subsequent years with regular updates.

In-class writing exercise with online constructive feedback

In 2008, during their essay-writing tutorial, students read a paragraph from a published article and were then asked to write a paragraph response to a focused question relating to information in the article. Students submitted their written paragraph online during class in the same way they would later submit their essay. These paragraphs were then run through WCopyfind. Students who copied multiple sentences to construct just one paragraph were sent an email actually telling them that they had plagiarised, but that there was no consequence because this was an exercise designed to prevent plagiarism in the future. The vast majority of students benefited indirectly from the many examples of plagiarism that were subsequently placed online with the sections of the paragraphs copied verbatim marked in red, an explanation of why this was inappropriate, and an outline of probable consequences and a grade. To provide a constructive experience, examples of good writing were also presented online, as well as other errors such as an over-reliance on quotations, and referencing errors omitting the primary source. Seventeen examples were used to create the online feedback from 676 responses (see White et al., 2008 for examples). In subsequent semesters, this constructive feedback module, created from actual student examples, was made available as an online learning module for students.

Online mastery quizzes

It has already been noted by Dunn et al. (2013) and others that online quiz with immediate feedback is a very effective type of formative assessment. To this end, in 2009, we developed several online quizzes where we attempted to simulate the interactivity of the writing exercise, without the consumption of classroom time and a delay in feedback. Many of the real examples of plagiarism, poor writing and good writing, were transformed into multiple choice and category-matching questions in an online quiz. The online quiz formed part of the students' regular continuous assessments. It was available for a week, and it allowed for unlimited attempts with a forced 3-hr break. It consisted of 11 questions on referencing, using quotations, plagiarism, plagiarism penalties, the use of ‘cited by’, number of references needed, psycINFO (search engine) use, and the sharing of work between students. Also, to address students' need to learn how to use search engines, our final question required a psycINFO search for a particular paper, with a single phrase answer to be entered. These online mastery exercises have featured in all subsequent semesters.

Writing tutorial

By the first semester of 2011, we were satisfied that students were understanding more about plagiarism in an abstract sense, but they had little experience applying this to actual writing. Consequently, we developed a new writing tutorial built around many problem-based learning tenets, namely engaging students in self-directed learning to facilitate the development of critical analysis skills (see Karantzas et al. 2013).

Students received a ‘resources sheet’ consisting of paragraphs from academic sources on particular topics. With one student per computer, a program asked students to write a paragraph based on that material and submit their paragraph online. The program then randomly sent the students' work, anonymously, to another student in the class. At this point, the tutor briefed all students on how to mark the work in front of them. For example, they asked in the early exercises: ‘Is the paragraph a substantial transformation of the resource sheet? If any part of it is identical, then it should be in quotes. If any information has been taken from the resources sheet it needs to be referenced. If not, note for the student that it is plagiarised’. At this point, before students had even begun to give peer feedback, many had already learnt significant lessons, aware that a colleague somewhere else in the room was classifying their work as plagiarised. After 2 min to type feedback based on the work and the tutor's advice, students then received the feedback on their work from their peers. Relying on students' feedback may not always result in good feedback, which is why the mini-'how‘to-mark’ tutor lesson within the exercises was critical, and the software also ensured that students received a different reviewer for each exercise. Depending on how long each exercise took, this process was repeated up to five times in the hour with increasingly challenging tasks all involving formal writing as a response.

Research questions

Overall, we were interested in answering four important research questions: (1) To what extent will our systematic battery of plagiarism reduction strategies from 2007 to 2011 be associated with a decrease in the number of reported cases of plagiarism? (2) Will any decrease be sustained across years with new cohorts and with the same students moving into later years? (3) Will there be sex differences in reported cases of plagiarism? and (4) Will the plagiarism reduction strategies have a different effect on the types (person-to person vs resources; moderate vs serious) of plagiarism cases detected?

Results

  1. Top of page
  2. Abstract
  3. Defining plagiarism
  4. Why do university students plagiarise?
  5. Strategies to reduce student plagiarism
  6. Method
  7. Results
  8. Discussion
  9. Acknowledgement
  10. References

Over 5 years, 10 semesters, and 14,338 students, 109 cases of plagiarism were detected including resources plagiarism (i.e., when students copied material directly from references or other materials) and person-to-person plagiarism (i.e., when students were found to have copied each other), further classified into moderate plagiarism (i.e., cases where 20–49% of the material had been plagiarised) and serious plagiarism (i.e., cases where more than 50% of material had been copied). Figure 1 combines both moderate and serious plagiarism cases to illustrate the trends for each of our assignments over 5 years.

figure

Figure 1. Plagiarism cases for the semesters 1 and 2 assignments from 2007 to 2011.

Download figure to PowerPoint

To evaluate the trend over this period, an initial overall chi-square analysis confirmed the significance of the strong downward shift of rates of plagiarism across all 10 semesters from 2007 to 2011, χ2 = 170.89, p < .005, degrees of freedom (df) = 9. Considering consecutive years, we also found that combined plagiarism rates fell from 2007 to 2008, χ2 = 30.46, p < .005, df = 3; and they also fell significantly from 2008 to 2009, χ2 = 15.13, p < .005, df = 3. However, no further improvements were observed from 2009 to 2010, χ2 = 3.44, p > .05, df = 3; or from 2010 to 2011, χ2 = 2.82, p > .05, df = 3, perhaps as the number of cases were very low at this stage.

A distinct difference between the two types of plagiarism was found when comparing our two different assignments. As can be seen in Fig. 1, the essay assignment led to significantly more resources plagiarism cases than person-to-person cases, χ2 = 41.14, p < .005, df = 1. However, as can be seen in Fig. 1, the research report assignment resulted in an even spread of types of plagiarism cases, χ2 = 0.17, p > .05, df = 1.

Table 2 summarises the combined plagiarism cases across years, but broken down by severity and gender. What is immediately striking from these data presented in Table 2 is that while there is little difference in the number of cases of resources plagiarism that were moderate and serious, χ2 = 0.33, p > .05, df = 1, significantly more person-to-person cases were serious as opposed to moderate, χ2 = 18.00, p < .005, df = 1.

Table 2. Resources and person-to-person plagiarism cases from ten Semesters from 2007 to 2011
 Moderate (20–49%)Serious (>50%)
MalesFemalesTotalMalesFemalesTotal
  1. Note. The average male : female ratio in these introductory psychology courses was approximately 1:2, with n = 5224 males and n = 9114 females in total.

Resources82735123042
Person-to-Person134141428

When we consider the number of males and females plagiarising, there is a proportional representation of the sexes based on the actual mix in these courses. While Table 2 reveals that there were 35 cases of male students plagiarising compared with 72 females, this ratio does not differ significantly from what we might expect given the overall numbers of males (n = 5224) and females (n = 9114) enrolled, χ2 = 0.65, p > .05, df = 1. Also, while the data do suggest that if males are involved, it tends to be in serious cases, this trend does not reach significance, χ2 = 2.27, p > .1, df = 1. Interestingly, however, when we combine moderate and serious cases, males are overrepresented in person-to-person plagiarism compared with resources plagiarism, χ2 = 4.16, p < .05, df = 1.

Discussion

  1. Top of page
  2. Abstract
  3. Defining plagiarism
  4. Why do university students plagiarise?
  5. Strategies to reduce student plagiarism
  6. Method
  7. Results
  8. Discussion
  9. Acknowledgement
  10. References

Our systematic regime of various plagiarism reduction strategies between 2007 and 2011 is strongly associated with decreased plagiarism. The most significant reduction in the overall number of reported cases of plagiarism occurred between 2007 and 2008 when an in-class writing exercise with online constructive feedback module was introduced in parallel with plagiarism detection software. Importantly, this significant decrease in the number of plagiarism cases continued between 2008 and 2009 when a writing mastery quiz was introduced and added alongside the previous plagiarism reduction strategies. The extremely low incidence of plagiarism cases continued into 2011, with numbers so low that achieving any further statistically significant decreases was near impossible.

The results also highlight the central role that the nature of the assignment plays in the type of plagiarism cases detected. Our semester 2 research report, based entirely on a unique study and unique data, might have been expected to reduce levels of plagiarism from external sources. The proportion of detected ‘resources’ cases for the research report were certainly lower than for the essay, but only because the incidence of person-to-person cases was higher. Thus, while Hughes (2009) suggests changing the nature of the assignment as a way to reduce plagiarism, we have seen that this enforced requirement for originality simply changes the type of student plagiarism. Ever since we first noticed this pattern, our various support materials, such as mastery quiz questions, have included clear warnings for students not to share their work.

Distinctions between types of plagiarism run deeper than their incidence depending on assignment categories. We also found that person-to-person plagiarism cases were more likely to be serious than moderate plagiarism, and that males were more likely to engage in this kind of plagiarism. Person-to-person plagiarism involves copying from a colleague's work, and it could be considered a consciously dishonest act. However, end-of-semester interviews with all students who were found to have been involved in person-to-person plagiarism revealed that in most cases assignments were willingly handed over from one student to another, usually to help a ‘friend’ who claimed to be sick or struggling. In fact, in only a few cases (n = 3, of 32) of person-to-person plagiarism were students' assignments stolen without their knowledge. Nevertheless, these interviews also revealed that once in possession of a complete work, students were aware that what they were doing was wrong, which explains why most cases of this type were serious (>50% plagiarised), as there was a deliberate strategy. While almost all those students caught copying from another student admitted that it was wrong, they generally blamed their actions on illness, language issues, or the difficulty of the assignment. Even their colleagues who had given away their assignments seemed fully aware of potential consequences and responded by saying things like: ‘I knew I shouldn't have emailed it to them, but they promised me they'd change every word’; or ‘I knew I wasn't supposed to show them, but they were very sick that week’. These straightforward admissions in the person-to-person cases contrasted heavily with students found copying from resources, who usually blamed their own carelessness or ignorance of referencing rules, and were considerably more upset with themselves, as opposed to being upset that they were caught.

It is important for us to consider these types of plagiarism separately, as they appeared to be produced by different kinds of assignment, at distinct levels of severity, and seem to be caused by completely different factors—that is, poor referencing knowledge versus outright dishonesty. This complex interplay of factors also supports a multifaceted approach to discouraging plagiarism, such as the one we have implemented. Specifically, measures intended to educate students about consequences and ease of detection are needed to directly discourage person-to-person plagiarism, whereas measures educating students about referencing and writing are essential to prevent negligent plagiarism from resources. Of course, where such educational measures form a cohesive package encouraging students to put more effort into their writing, they should also be less tempted to copy their colleague's work.

Several studies have found sex differences in attitudes to plagiarism and self-reported cheating (e.g. Underwood and Szabo, 2004), which suggest that males engage in such practices more often; however, in our large-scale longitudinal study of actual plagiarism cases, we found no overall difference between the proportion of males and females involved. Koul et al. 's (2009) findings about what males and females consider to be plagiarism showed no sex differences in interpretations of copying from colleagues, but our one significant sex difference finding was that males were slightly overrepresented in the person-to-person category of plagiarism, a more dishonest and serious version of plagiarism. Otherwise, our lack of obtained differences suggests that either the large sex differences found in self-reported cheating may not generalise to actions by males and females, or perhaps whatever gap there was between the sexes in academic misconduct (e.g. Davis, Grover, Becker & McGregor, 1992) no longer exists.

Aside from our observations of the small number of plagiarists caught between 2007 and 2011, the overwhelming finding of our study is that initially high plagiarism rates in two large university courses were reduced considerably over 5 years thanks to a steady rollout of educational interventions. Engaging students in interactive exercises, both online and in-class, clearly increased their awareness of the nature of plagiarism and how to avoid it. Key to the success of these interventions must have been their central integration into the courses. These were not peripheral modules to be completed as an afterthought. In-class interventions were given substantial class time, and online activities such as the mastery quizzes were awarded assessment marks. These integrated curriculum strategies informed all students that they should be aware of the possibility of plagiarism from the beginning of their writing through to submission of their assignment.

Most importantly, once reduced, plagiarism rates remained extremely low, even to the extent that this made further interventions harder to assess. Additional analyses revealed that these low plagiarism rates of first-year students were sustained in the second- and third-year courses that these same students subsequently moved into with only a total of nine cases detected since 2008. In these later courses, only plagiarism software and warnings were utilised (as in our initial 2007 first year intervention); however, almost all our students progressing to second and third year had experienced our more comprehensive interventions in first year. This subsequent success across all years strongly suggests first year is the most appropriate time to strongly emphasise writing and referencing practices, as the lessons learnt can be carried forward with minimal additional maintenance.

While we would like to think that our own education-based approaches have encouraged students to create original passages from scratch, it may be that their increased awareness of how plagiarism software works (i.e. essentially by verbatim matches with a small degree of flexibility) has simply allowed them to sharpen their techniques to escape detection. However, even if some component of our reduction in observed plagiarism was caused by dishonest plagiarists being better informed, the system would still force such plagiarists to reword their entire assignments. The clearest message we should send to all those students who spend hours rewording paragraphs they never wrote is that their time is better spent composing assignments from scratch because that is where high marks are scored. Assignments must be designed to ensure that dull, reworded summaries do not obtain strong marks.

While is it difficult to disentangle the independent contribution of each of our strategies, we believe that this multifaceted approach to reducing rates of plagiarism has been a success precisely because it captures all students at all levels of ability and attitude. On the one hand, deterrence strategies are an essential component because educational strategies in isolation will not affect students who are committed to cheating. On the other hand, an educational approach to both writing and referencing is essential for those students who genuinely have little understanding of referencing rules. As we face an increasing threat of plagiarism from better technology, we should use technology to create interesting and engaging resources to assist our students to become better writers.

Acknowledgement

  1. Top of page
  2. Abstract
  3. Defining plagiarism
  4. Why do university students plagiarise?
  5. Strategies to reduce student plagiarism
  6. Method
  7. Results
  8. Discussion
  9. Acknowledgement
  10. References

This research was generously supported by a University of Sydney Teaching Improvement and Equipment Scheme (TIES) Grant to Fiona White and Caleb Owens.

References

  1. Top of page
  2. Abstract
  3. Defining plagiarism
  4. Why do university students plagiarise?
  5. Strategies to reduce student plagiarism
  6. Method
  7. Results
  8. Discussion
  9. Acknowledgement
  10. References
  • Ashworth, P., Bannister, P., & Thorne, P. (1997). Guilty in whose eyes? University students' perceptions of cheating and plagiarism in academic work and assessment. Studies in Higher Education, 22, 187203.
  • Belter, R. W., & Du Pre, A. (2009). A strategy to reduce plagiarism in an undergraduate course. Teaching of Psychology, 36, 257261.
  • Bennett, R. (2005). Factors associated with student plagiarism in a post-1992 university. Assessment and Evaluation in Higher Education, 30, 137162.
  • Bermingham, V., Watson, S., & Jones, M. (2010). Plagiarism in UK law schools: Is there a postcode lottery? Assessment & Evaluation in Higher Education, 35, 114.
  • Braumoeller, B., & Gaines, B. (2001). Actions do speak louder than words: Deterring plagiarism with the use of plagiarism detection software. Washington DC: American Political Science Association. Retrieved from http://www.apsanet.org/PS/dec01/braumoeller.cfm
  • Carroll, J. (2004). Six things I did not know four years ago about dealing with plagiarism. In H. Marsden , M. Hicks , & A. Bundy (Eds.), Educational integrity: Plagiarism and other perplexities. Proceedings of the first Australasian Educational Integrity Conference (pp. 1218). Adelaide, SA: University of South Australia.
  • Chester, A., Burton, L. J., Xenos, S., & Elgar, K. (2013). Peer mentoring: Supporting successful transition for first year undergraduate psychology students. Australian Journal of Psychology, 65(1), 3037.
  • Davis, M. (2007). The role of Turnitin in the formative process of academic writing: A tool for learning and unlearning? Brookes e-Journal of Learning and Teaching, 2(1). Retrieved from http://www.bejlt.brookes.ac.uk/article/the_role_of_turnitin_within_the_formative_process_of_academic_writing/
  • Davis, M., & Carroll, J. (2009). Formative feedback within plagiarism education: Is there a role for text-matching software? International Journal for Educational Integrity, 5, 5870.
  • Davis, S. F., Grover, C. A., Becker, A. H., & McGregor, L. N. (1992). Academic dishonesty: Prevalence, determinants, techniques, and punishments. Teaching of Psychology, 19(1), 1620.
  • Devlin, M. (2003). Policy, preparation, prevention and punishment: One Faculty's holistic approach to minimising plagiarism. In H. Marsden & M. Hicks (Eds.), Educational Integrity: Plagiarism and other perplexities, refereed proceedings of the inaugural educational integrity conference (pp. 3947). Adelaide: University of South Australia. 21–22 November.
  • Dunn, D. S., Saville, B. K., Baker, S. C., & Marek, P. (2013). Evidence-based teaching: Tools and techniques that promote learning in the psychology classroom. Australian Journal of Psychology, 65(1), 513.
  • Elander, J., Pittam, G., Lusher, J., Fox, P., & Payne, N. (2010). Evaluation of an intervention to help students avoid unintentional plagiarism by improving their authorial identity. Assessment and Evaluation in Higher Education, 35, 157171.
  • Homewood, J. (2007). An active approach to preventing plagiarism and increasing generic skills in new undergraduate students: The PSY104 project. Paper presented at the Higher Education Academy Conference, July 3–5, in Harrogate, UK.
  • Hughes, C. (2009). Assessment as text production: Drawing on systemic functional linguistics to frame the design and analysis of assessment tasks. Assessment and Evaluation in Higher Education, 34, 553563.
  • Karantzas, G. C., Avery, M. R., Macfarlane, S., Mussap, A., Tooley, G., Hazelwood, Z., & Fitness, J. (2013). Enhancing critical analysis and problem-solving skills in undergraduate psychology: An evaluation of a collaborative learning and problem-based learning approach. Australian Journal of Psychology, 65(1), 3845.
  • Knott, V. E., Mak, A. S., & Neill, J. T. (2013). Teaching intercultural competencies in introductory psychology via application of the Excellence in Cultural Experiential Learning and Leadership model. Australian Journal of Psychology, 65(1), 4653.
  • Koul, R., Clariana, R. B., Jitgarun, K., & Songsriwittaya, A. (2009). The influence of achievement goal orientation on plagiarism. Learning and Individual Differences, 19(4), 506512.
  • Landau, J. D., Druen, P. B., & Arcuri, J. A. (2002). Methods for helping students avoid plagiarism. Teaching of Psychology, 29, 112115.
  • Park, C. (2004). Rebels without a clause: Towards an institutional framework for dealing with plagiarism by students. Journal of Further and Higher Education, 28, 291306.
  • Perry, B. (2010). Exploring academic misconduct: Some insights into student behaviour. Active Learning in Higher Education, 11, 97108. doi: 10.1177/1469787410365657.
  • Roig, M. (1997). Can undergraduate students determine whether text has been plagiarised? Psychological Record, 47, 113122.
  • Rolfe, V. (2011). Can Turnitin be used to provide instant formative feedback? British Journal of Educational Technology, 42, 701710.
  • Szabo, A., & Underwood, J. (2004). Cybercheats: Is information and communication technology fuelling academic dishonesty? Active Learning in Higher Education, 5, 180199.
  • Underwood, J., & Szabo, A. (2004). Plagiarism: Is this a problem in tertiary education. School of Science, The Nottingham Trent University. Retrieved, 3.
  • White, F. A., Owens, C., & Nguyen, M. (2008). Using a constructive feedback approach to effectively reduce student plagiarism among first-year psychology students. Symposium Proceedings, UniServe Science Conference, The University of Sydney, Australia, 127132.
  • Wilson-Doenges, G. & Gurung, R. A. R. (2013). Benchmarks for scholarly investigations of teaching and learning. Australian Journal of Psychology, 65(1), 6370.