Tracking plagiarism electronically: First-year students’ perceptions of academic dishonesty and reports of cheating behaviour in the basic communication course

This study explored how electronic submission of course material, intended to deter instances of plagiarism, influenced first-year students’ perceptions of academic dishonesty and reports of cheating behaviour in a large, multi-section basic communication course. Results reveal that electronic submission of course material results in first-year students being less likely to self-report engaging in cheating behaviours and heightens their appreciation and awareness of what constitutes academic dishonesty. Implications for classroom pedagogy, course management, and teacher training are discussed.


Introduction
The transition to college is layered with stress for first-year students as they adjust to new academic responsibilities. This adjustment can function as a potentially invigorating and unsettling time for students as they navigate newly defined academic expectations for college level performance. In fact, first-year students have reported feeling overwhelmed due to the change in environment, workload, and responsibility that accompanies the transition to college (Jorgensen-Earp & Staton, 1993;Smith, Carmack, & Titsworth, 2006). Due to those heightened academic expectations, firstyear students may feel compelled to engage in academic dishonesty as a mechanism through which to handle new academic expectations (Perry, 2010) and earn high grades (Purcell, 2009).
Issues of academic dishonesty at colleges and universities across the globe have recently received much attention in the national media (Blum, 2009;Slattery, 2008;Wasley, 2006;Young, 2010). This attention has heightened the concern of educators and administrators at all educational levels. Indeed, academic dishonesty concerns have been reflected in research in various academic fields including education (Daniel, Blount, & Ferrell, 1991;Karlins, Michaels, & Podlogar, 1988), science (Paldy, 1996), psychology (Marsden, Carroll, & Neill, 2005;Roig & Caso, 2005), criminal justice (Smith, Dupre, & Mackey, 2005), and business (O'Neill & Pfeiffer, 2012;Reisenwitz, 2012;Simha, Armstrong, &Albert, 2012. As Whitley andKeith-Spiegel (2002) note, several reasons justify this concern including a need for equity among student grades, the importance of moral and ethical character development among students, and a desire to uphold the mission of the institution. Additionally, they cite the importance of maintaining high student and faculty morale as well as the need to sustain the reputation of the institution as critical reasons educators must act to deter students from engaging in academically dishonest behaviour. Despite this larger concern with the importance of academic integrity, researchers have found that some educators actually make cheating easy for students by not working proactively to deter such behaviour (Lathrop & Foss, 2005). In one of the most comprehensive assessments of academic dishonesty, McCabe (2005) surveyed approximately 50,000 undergraduates on more than 60 campuses and found that, on most campuses, 70% of students admit to some form of cheating. In addition, nearly one quarter of all participating students admitted to cheating on a test within the past year, while half admitted to engaging in one or more instances of cheating on written assignments.
Definitions of what exactly constitutes cheating have led to a better appreciation for academic honesty in university coursework. Burrus, McGoldrick, and Schuhmann (2004) compared students' cheating behaviour before and after they were provided with a definition of cheating. In general, they found that students' self-reports of cheating increased after they received the definition, suggesting a gain in appreciation for what constitutes academic dishonesty. Those students who underreported their cheating behaviour tended to have lower grades and placed less pressure on themselves to earn high grades. Burrus et al.'s (2004) results suggest that, regardless of grades or academic ability, the frequency of cheating behaviour was largely due to differences in how -cheating‖ was defined for students. Similarly, in a survey of business students, Moberg, Sojka, and Gupta (2008) found that majors and students with lower grade point averages were more likely to engage in academic dishonesty. To address this issue, they called for increased vigilance and effort by faculty members to discourage cheating and research to assess methods that faculty might employ to deter academic dishonesty among college students.
In the communication discipline, Holm (2002) examined students' perceptions of cheating and reports of cheating behaviour in a basic public speaking course and encouraged scholars to -focus on methods for preventing cheating in all classes‖ (p. 74) and create research agendas that emphasise the effectiveness of cheating prevention strategies. Although extant research offers recommendations on disciplinary measures and assignment modifications, Holm (2002) noted that -little to no empirical research on the effectiveness of any of these methods has been conducted‖ (p. 74). This study directly addresses this gap in the literature and examines the effectiveness of electronic submission of course material as a means of deterring academic dishonesty in a large, multi-section basic communication course that is required of all first-year students.

Cheating in the Basic Communication Course
Even though cheating has recently been a popular area of examination in various academic fields, research in the communication discipline generally and the basic communication course specifically has been sparse. In one of the few studies in this area, Holm (2002) found that 54% of the 307 students surveyed reported committing one or more cheating behaviours in a basic public speaking course. In this study, most students reported that their peers were lying about where they found their information, fabricating sources, and making up portions of the bibliographic data for speaking assignments. He also found that students had varied perceptions of what constituted academic dishonesty in scenarios that were directly related to the basic public speaking course. Specifically, the analysis revealed that participants chose either of two extreme positions -definitely cheating or definitely not cheating -most often. However, Holm (2002) noted that what one student considered definitely cheating, another student considered definitely not cheating. This finding leads one to conclude that mixed perceptions of what constitutes academic dishonesty can exist among students, particularly first-year students who are adjusting to newly defined academic expectations (Jorgensen-Earp & Staton, 1993;Perry, 2010;Smith, Carmack, & Titsworth, 2006). Without a doubt, cheating is a socially deviant behaviour and few students would want to get caught for fear of potential repercussions and, in fact, prior research suggests that the risk of detection inhibits cheating (Houston, 1983;Michaels & Miethe, 1989;Tittle & Rowe, 1973). However, Michaels and Miethe (1989) found that even if students perceive a high risk of detection, they may still choose to engage in academic dishonesty. Specifically, they note that those students who consider cheating will weigh the risk of detection and the potential for disciplinary action against the benefits they will receive from engaging in such unethical behaviour. Michaels and Miethe (1989) conclude that students will cheat if they believe the potential benefits will outweigh the probable costs. This assertion is of utmost concern in the basic communication course (and in other general education courses) where first-year students may be given few opportunities to demonstrate a mastery of public speaking techniques. For example, in some basic courses, a large portion of a student's final grade may be determined by evaluating their performance on two or three speeches. As a result, first-year students may contemplate engaging in academic dishonesty not only to earn a high grade on the speaking assignment, but also to earn a high final grade in the course. In response to Holm's (2002) call for action, this study tests the implementation of electronic submission of course material as a means of heightening first-year students' awareness of and ultimately deterring academic dishonesty. The following research questions guided our efforts: RQ 1 : How does the implementation of electronic submission of course material affect first-year students' self-reports of cheating behaviour in the basic communication course? RQ 2 : How does the implementation of electronic submission of course material affect first-year students' perceptions of what constitutes academic dishonesty?

Participants
Participants were 147 first-year undergraduate students enrolled in the basic communication course at a large Midwestern university. The university offers approximately 80 sections of the basic course each semester and enrolls 23 students in each section. Because the course is required of all first-year students as part of the university's general education program, the participants represented various academic disciplines. The sample consisted of 56 males and 91 females, with an average age of 18.16 years (ranging from 18 to 23 years). The racial/ethnic distribution was 82.8% Caucasians, 9.7% African Americans, 5.5% Hispanics, 1.4% Asian Pacific Islanders, and 0.7% other.

Procedures
All procedures were approved through the university's Institutional Review Board. Six sections of the basic communication course were randomly assigned to an experimental group and six sections to a control group. In the first week of classes, each instructor of record informed students that participation in the study was voluntary. Students read an informed consent passage and were instructed to keep it for their records.
Manipulation. Throughout the 16-week semester, experimental group instructors asked their students to submit course materials to an email account that was managed by the researchers. Students were asked to submit outlines from the informative speech, group speech, and persuasive speech, a communication improvement profile paper, a series of communication application essays, and a final course synthesis paper. On a weekly basis throughout the semester, through oral announcements in class and through email messages, the instructors frequently reminded students to submit materials to the email account to allow teachers to assess the presence of academic dishonesty in the course. Through face-to-face meetings and via email, we verified with individual instructors in the experimental group that they were continually asking their students to submit assignments to the email account. We were interested in exploring the deterrent effect of electronic submission of course materials, and as a result, none of the materials were actually submitted to a plagiarism database or other available detection software. The control group participants were not instructed to submit their course materials to the email account.
Measurement. The questionnaire was adapted from a study conducted by Holm (2002). The survey consisted of two primary measures: (1) a measure to assess participants' perceptions of academic dishonesty scenarios, and (2) a measure to compile reports of cheating behaviour. The first measure gauged participants' perceptions of academic dishonesty in 11 scenarios described in Table 1. The severity of the scenarios ranged from changing a source's publication date to blatantly copying a friend's outline and research. Participants responded using a seven-point semantic differential scale (this is not cheating at all/this is definitely cheating) to record the degree of violation (rather than a simple relationship -e.g., this is cheating, this is not cheating) in each situation. The second measure asked participants to report if they engaged in various forms of cheating behaviour in the basic communication course during the semester. The severity of the cheating behaviour ranged from using another person's research to using an entire speech written by someone else. Participants responded with -yes‖ or -no‖ to the 10 cheating behaviours presented in Table 2. Given the first measure's design to gauge participants' reactions to 11 different scenarios and the nominal nature of the second measure, alpha reliability estimates are not available (Holm, 2002).
During the first week of the semester, instructors of record for each section in the experimental and control groups administered a pre-test questionnaire. The pre-test was administered before students received instruction in ethical communication, including discussions of plagiarism and academic dishonesty. Although instruction may have varied from section-to-section, students in the experimental and control groups were exposed to the same degree of instruction in ethical communication as all sections at the university utilise similar lesson plans, a common course textbook, and a standardised supplementary materials packet. The pre-test questionnaire contained the first measure that gauged participants' perceptions of academic dishonesty in the 11 scenarios described in Table 1. During the sixteenth week of the semester, the instructors of record administered the post-test questionnaire.
The post-test questionnaire, through the first measure, again gauged participants' perceptions of academic dishonesty in the 11 scenarios. We were specifically interested in differences between the control and experimental groups from pre to post-test. In order to assess pre to post-test differences between the groups, we calculated an overall difference score by subtracting students' pre-test means from their posttest means for each of the 11 items. We then summed the individual items to produce an overall academic honesty score (the closer to 0 the score is, the less students improved over time). The post-test questionnaire also included the second measure where participants reported if they engaged in various cheating behaviours during the semester by responding to ten -yes or no‖ questions (see Table 2). Participants recorded a unique identification number on both questionnaires to allow the researchers to track responses from pre to post-test. The data analyses examined only those students who completed both the pre and post-tests.

Results
The first research question examined the cheating behaviours first-year students reported committing most frequently in the basic course. Specifically, we were interested in differences in self-reported cheating behaviour between first-year students in the control and experimental groups. Overall, chi-square analyses of the 10 cheating items indicate that first-year students in the control group were more likely than their peers in the experimental group to have completely made up a source for a speech, χ 2 = 4.06, df = 1, p < .05, and lied about where they found source information, χ 2 = 7.75, df = 1, p < .05. Analysis of frequency distributions also revealed that a total of 78.5% of cheating behaviours were reported by first-year students in the control group compared to 49.2% reported by their peers in the experimental group. Students' responses, in percentages, can be found in Table 2.  Table 2 First-year Students' Self-Reported Cheating Behaviour Note. Asterisk indicates a significant difference, for that item, between the groups.

Discussion
The transition to college is laden with stress as first-year students adjust to new personal and academic responsibilities (Smith et al., 2006). As part of this adjustment, first-year students must negotiate new roles or identities and learn to manage the academic rigors associated with university level coursework (Jorgensen-Earp & Staton, 1993). In light of these enhanced academic expectations, first-year students may feel compelled to engage in academic dishonesty to manage new academic expectations (Perry, 2010) and achieve high grades in their courses (Purcell, 2009).
Given the prevalence of academic dishonesty among students at various educational levels (Aiken, 1991;Barnett, 1997;Genereux & McLeod, 1995;McCabe, 2005), it is vital for teachers to implement appropriate strategies that dissuade student cheating behaviour. Although a preponderance of scholarship in this arena underscores the importance of dissuading student cheating behaviour and suggests strategies for preventing its occurrence (Holm, 2002;Marsden et al., 2005;Roig & Caso, 2005), little to no empirical research tests the effectiveness of these methods. The present study directly addressed this gap in the literature. The results reveal that asking firstyear students to submit course materials electronically may affect their perceptions of what constitutes academic dishonesty and decrease their self-reported cheating behaviour. These findings are especially salient given previous research suggesting that students will engage in academic dishonesty if they believe the potential benefits will outweigh the probable costs (Michaels & Miethe, 1989 (Houston, 1983;Michaels & Miethe, 1989;Tittle & Rowe, 1973).
The first research question explored how electronic submission of course material affected first-year students' reports of cheating behaviour. Results revealed that control group participants were more likely than experimental group participants to self-report that they had lied about where they found source information and were more likely to report that they had completely made up a source for a speech. In essence, the findings suggest that electronic submission of course material deterred experimental group students from blatantly fabricating source information, a finding applicable to courses from various disciplines. Our data also reveal an important difference between groups on the total number of reported cheating behaviours. A total of 78.5% of cheating behaviours were reported by first-year students in the control group compared to 49.2% reported by their peers in the experimental groupresulting in a 29.3% difference between groups. Taken together, the findings suggest that electronic submission of course material may deter first-year students from cheating in undergraduate courses.
The second research question asked how the implementation of electronic submission of course material affected first-year students' perceptions of academic dishonesty scenarios. Our findings indicate that students in the experimental group experienced a more significant gain in their appreciation for what constitutes academic dishonesty compared to their peers in the control group. Specifically, experimental group participants improved significantly in their perception of what constitutes academic dishonesty compared to control group participants for the following items: recycling term paper, giving a sales speech for a job, and using another person's idea and visual aids for a speech. Early in the semester, students might have initially perceived these behaviours as acceptable acts; however, throughout the course of the term, they likely developed a better understanding of what comprises academic dishonesty.
In some respects, there are limited effects in this study. The data indicate that firstyear students' scores did not improve on more severe academically dishonest acts such as using a friend's outline and research and recapping an article and calling it a speech. First-year students may have already possessed a significant understanding of these more risky behaviours and, as a result, did not report a gain in appreciation.
Overall though, based on these data, it appears that electronic submission of course material leads to an increase in first-year student awareness of what constitutes academic dishonesty and cheating behaviour in the basic communication course.
Equally important, the control group members' perceptions of academic dishonesty actually decreased from pre-test to posttest on several scenarios, suggesting that their appreciation for what constitutes academic dishonesty actually diminished throughout the semester. This finding is particularly striking and illustrates the importance of employing mechanisms (e.g., electronic submission of course material) that heighten students' awareness of academic dishonesty and dissuade them from engaging in cheating behaviour. Future research might assess whether or not this trend is present in other courses and how first-year students come to better appreciate academically honest behaviour in other content areas (Roig & Caso, 2005;Smith et al., 2005).
With respect to the present study, it seems as though the mere presence of an electronic database led first-year students in the experimental group to develop a more fully crystallised understanding of what constitutes academic dishonesty. Although a single, clear definition of academic dishonesty was not provided for participants in this study, courses, departments, and institutions could benefit from definitions of cheating to foster in students a better appreciation for academic honesty (Burrus et al., 2004).
Taken together, the findings should be of interest to course directors, particularly those who oversee large, multi-section introductory courses primarily designed for first -year students. The results reveal that the implementation of electronic submission of course material can heighten students' awareness of academic dishonesty and potentially deter cheating behaviour. Furthermore, because many institutions require first-year students to enroll in a basic communication course (Gibson, Hanna, & Leichty, 1990;Morreale, Hanna, Berko, & Gibson, 1999), course directors can capitalise on efforts to deter first-year students from engaging in academic dishonesty in the future by implementing electronic submission methods in their courses. This might also help first-year students gain a better appreciation for what constitutes academic dishonesty and, as a result, they may be more inclined to make ethical decisions when completing assignments for other classes. Importantly, the implications of our findings stretch beyond the limits of the basic communication course and any introductory course for that matter. Any administrator could implement electronic methods to track plagiarism within an academic department and potentially deter students from engaging in academic dishonesty in major-related courses (Smith et al., 2005).
Teachers stand on the front line of efforts to deter academically dishonest acts. The results of this study offer course directors an opportunity to provide beginning teachers with important training to guide their first teaching experience, particularly in the areas of course management. Course instructors who are trained to incorporate electronic submission of course material in their classes stand prepared to manage instances of academic dishonesty, promote student awareness of academic dishonesty, and potentially deter students from engaging in cheating behaviours and other student misbehaviours (Meyer, Hunt, Comadena, Simonds, Simonds, & Baldwin, 2008). By including information regarding electronic submission of course material in graduate teaching assistant (GTA) training programs, GTAs and students can jointly benefit-GTAs will be better prepared to establish a classroom environment that is conducive to student learning, and students can gain a better appreciation for academic honesty and potentially be deterred from engaging in cheating behaviour.
In light of these significant implications, any study must be interpreted within the limitations imposed by the research design. Initially, first-year students in the experimental group knew that their instructors were looking for plagiarism and a social desirability bias may have made them less likely to self-report committing such acts. It may be that electronic submission of course material may deter students from reporting cheating behaviour and not change their actual behaviour at all. However, as extant literature indicates, it is precisely this sensitivity to academic dishonesty that instructors must instill in their students in order to deter such acts (Michaels & Miethe, 1989). Anonymous self-report measures have been widely employed in the study of academic dishonesty (Burrus et al., 2004;Hollinger & Lanza-Kaduce, 1996;Moberg et al., 2008); however, this method has been subjected to criticism, including social desirability bias (Bushway & Nash, 1977;Spiller & Crown, 1995). Despite any potential limitations, self-report methods in the study of academic dishonesty have many advantages. Hollinger and Lanza-Kaduce (1996) noted: a self-administered survey provides the best opportunity to obtain detailed information from students about their academic dishonesty. It also avoids the ethical problems associated with contriving temptations to cheat and then deceiving students about it. Further, a survey instrument can be used to collect information efficiently about different forms of academic dishonesty across a variety of contexts. Surveys also permit students to remain anonymous. In general, confidential self-report surveys about minor forms of deviance among conventionally socialized individuals have been judged to be methodologically valid and reliable. (p. 394) Our findings illustrate the effect of electronic submission of course material on firstyear students' perceptions of academic dishonesty. We would hope that as students become more sensitive to what constitutes academic dishonesty they would avoid such behaviour. As a means of triangulating student self-reports with other sources, scholars might consider in-depth interviews or focus groups with instructors to determine whether or not electronic submission affects teachers' reports of student cheating behaviour. Furthermore, research might couple traditional forms of survey data with electronic data, including students' video portfolios that chronicle their achievement in performance-based classes (Voth & Moore, 1997), to examine academic dishonesty among college students. Future research might also explore whether or not first-year students at two-year institutions differ in their reports of cheating when compared to first-year students at four-year universities (Holm, 2002), and how the transition to a four-year institution might impact transfer students' tendency to engage in academic dishonesty (Owens, 2010). Researchers can also track instances of academic dishonesty through university officials to determine if there are differences in the number of cases adjudicated based on prevention strategies employed in classes.
As we noted, the present study did not actually analyse submissions to the email account with plagiarism detection software. In the future, teachers and researchers might utilise free shareware programs readily available on the Internet or software packages available for purchase to search for instances of plagiarism in the material submitted by students. Electronic platforms, such as Blackboard, WebCT, and Turnitin, naturally function as electronic databases to deter students from cheating (Moberg et al., 2008). It is reasonable to assume that telling students their materials will be analysed using such software would have a larger deterrent effect than merely asking students to submit materials electronically. Scholarship in this venue can assess the effects of these technological platforms on deterring cheating behaviour among college students. Additionally, the use of plagiarism detection software has the added advantage of providing course directors, instructors, and other administrators with the capability to compare submissions from one section to another section, from one course instructor to another course instructor, and submissions at one institution to submissions from a myriad of other institutions.

Conclusion
Without a doubt, academic dishonesty issues are important facets of any academic community. Michaels and Miethe (1989) encourage teachers to implement pedagogical strategies that demonstrate to students that the costs of academic dishonesty outweigh the benefits. Implementing electronic submission of course material is one step in that direction and can help first-year students in particular gain a better appreciation for what constitutes academic dishonesty and potentially deter them from engaging in various types of cheating behaviour. This practice sends a clear message to all students that the instructor, the academic department, and the institution care deeply about fostering an environment of academic integrity where first -year students are able to navigate newly defined standards for academic performance.
Stephen K. Hunt (Ph.D., Southern Illinois University) serves as Professor, Director of Graduate Studies, and Associate Director of the School of Communication at Illinois State University. He also serves as the University's Chair of the American Democracy Project and is a Carnegie Fellow for Political Engagement, leading national efforts to sharpen the political and civic leadership skills of today's college students. Hunt is the author of over 30 scholarly articles, many focused on civic and political engagement, and has authored a public speaking textbook focused on critical thinking and civic engagement that is used by several communication departments throughout the country.