their five‐year durations. These studies reinforce the conclusions of the reviews by Marusic et al. (2016) and Stoesz and Yudintseva (2018) that skills‐based interventions, accompanied by text‐matching software, appear to be effective in reducing academic misconduct.
In summary, it appears that the evidence for skills‐based interventions and text‐matching software is more substantial and provides clearer demonstrations of impact than the implementation of honor codes. This conclusion accords with some important evidence that students often plagiarize because they do not know how not to plagiarize (Delvin and Gray, 2007). Moreover, only assuming academic integrity reflects honor, character, and morality fails to account for the fact that plagiarism can be inadvertent (Barnhardt, 2016; Mulholland, 2020). Taken together, it could be argued that students do not develop character, in the context of academic integrity, just by being told to be of good character; they must be taught how to be of good character. This is not to say that higher education should abandon honor codes as the evidence for their effectiveness in reducing cheating, especially when honor codes are regularly made salient, is persuasive (Tatum and Schwartz, 2017). However, it is, at least, important to accompany honor codes with the education students need in the practical skills of academic writing to allow them to meet the terms of these codes.
CONCLUSIONS ON TRENDS IN THE PREVALENCE OF PLAGIARISM AND CHEATING: 1990–2020
In this chapter, I have reviewed the best available evidence of trends in the prevalence of plagiarism and cheating over the 30 years from 1990–2020. Specifically, I aggregated and compared the results of three time‐lag studies of plagiarism and cheating, where each study surveyed similar student groups with similar questions, allowing for like‐with‐like comparisons. These studies indicated that students’ engagement in plagiarism and cheating has trended downward since at least 1994. Two main explanations were considered for the downward trend in cheating and plagiarism: 1) whether students are switching to commercial contract cheating to avoid “detectable” forms of plagiarism with the rise of text‐matching software and 2) whether interventions designed to reduce plagiarism and cheating are effective and, if so, what types?
On reviewing the best available like‐with‐like evidence, there was no indication of a clear trend in the prevalence of students’ engagement in commercial contract cheating in the last 30 years. This result is consistent both with crime displacement theory and findings that people are less likely to break moral rules that they consider to be more serious (Ariely, 2012). However, there is considerable evidence that educational interventions, combined with text‐matching software, were effective in reducing plagiarism and cheating. In addition, there was less clear evidence that honor codes may also be beneficial in lowering plagiarism and cheating. In sum, these results suggest that the work higher education institutions and staff have been doing to try to reduce plagiarism and cheating has had a positive effect. The general finding that educational interventions work to reduce academic misconduct accords with arguments that, in an educational setting, education is preferable to punishment to elicit future academic integrity behaviors of students (Bertram Gallant and Stephens, 2020).
It is important to point out that despite the downward trend in plagiarism and cheating, only one data‐point in Figure 1, Stiles et al.'s (2018) 2014 survey, shows less than 50 percent of students engaging in some form of academic misconduct. For all of the other time‐lag studies, most students engaged in some form of plagiarism or cheating at least once. Therefore, the finding that plagiarism and cheating is trending downward should not make higher education providers complacent. Every year, new students start their learning journey in colleges and universities, and every year these new students must be trained to understand and apply the expected standards of integrity in their work. In addition, it is worth considering emerging threats to academic integrity that may stop or reverse the downward trend.
CURRENT AND FUTURE THREATS TO ACADEMIC INTEGRITY
One recent development that may contribute to the future recorded trends in the prevalence of academic integrity breaches is the Covid‐19 pandemic, which seriously disrupted usual higher education assessment and teaching practices. There are four main reasons to believe that the Covid‐19 pandemic may result in an increase in the prevalence of plagiarism and cheating for some years to follow. First, many higher education institutions made a rapid move away from in‐person proctored exams to various alternatives, such as unproctored take‐home exams and online exams. Clearly, proctored exams significantly reduce students’ opportunities to cheat; indeed, evidence suggests that students achieve marks 10–20 percent better in unproctored online exams, and some of this improvement may be attributable to cheating (Daffin and Jones, 2018). Second, it is reasonable to think that a rapid switch to online teaching will not have been executed with a high degree of skill or planning in all instances. Students are likely to find poorly‐delivered online courses dissatisfying, and dissatisfaction with the learning and teaching environment predicts engagement in cheating (Anderman and Won, 2019; Bretag et al., 2019). Third, the Covid‐19 pandemic is likely to have caused many students significant emotional stress, including fear for their health and the health of their loved ones, social isolation, and potential bereavement. Negative emotions, such as anxiety, increase unethical behavior per se (Kouchaki and Desai, 2015). More specifically, Tindall and Curtis (2020) reported that negative emotional experiences were related to more positive student attitudes toward plagiarism, and Birks et al. (2020) note that mental health concerns are often reported as a factor by students who have engaged in academic misconduct. Moreover, Eaton and Turner (2020) argue that steps taken to address academic integrity during the Covid‐19 pandemic, such as e‐proctoring, have unknown impacts on students’ mental well‐being. Finally, essay mills and other contract cheating services have directed advertising at students focused on “helping” students to overcome study problems attributable to the pandemic (McKie, 2020). Indeed, research emerging in 2020–2021 suggests that the prevalence of academic misconduct increased during the Covid‐19 pandemic (e.g. Lancaster and Cotarlan, 2021).
There are several reasons to think that some of the factors associated with the Covid‐19 pandemic that may increase academic misconduct prevalence will persist. Optimistic estimates initially predicted an end to the Covid‐19 pandemic in late 2021, but the emergence of variants of the virus and the need for worldwide vaccination suggests that it will persist as a public health threat for much longer than initially hoped (Charumilind et al., 2020). Numerous commentators have speculated that some substitution of face‐to‐face with online delivery, and proctored with unproctored assessments, will persist for years beyond the pandemic and may become “the new normal” (e.g. Champagne and Granja, 2021; McMurtrie, 2021). It is, therefore possible that students may develop new habits of cheating and collusion on unproctored tests that will solidify as new norms of behavior that are subsequently conveyed to other students.
Another driver of future trends in the prevalence of plagiarism and cheating is the ongoing technological tug‐of‐war between those seeking to engage in academic misconduct, or allow others to engage in academic misconduct, and those seeking to deter and detect academic misconduct. Curtis and Vardanega (2016) suggested that advances in technology that allow students to cheat and plagiarize are potentially matched by technologies that allow cheating to be detected in a process analogous to a co‐evolutionary arms race (Vermeij, 1992). The launch of Turnitin™ in 2000 and the gradual uptake of systems for automated text‐matching align historically with the downward trend in plagiarism over the past 30 years. However, an important question is, what technological changes will the future hold?
One possible battleground in the next arms race between technologies that facilitate cheating and technologies that may deter or detect cheating is artificial intelligence and machine learning. It has been reported that artificial intelligence can be used to create “passable” academic writing automatically (e.g. Abd‐Elaal et al., 2019). Automatic paraphrasing tools can potentially revise unoriginal work sufficiently to evade detection by text‐matching software (Rogerson and McCarthy, 2017). Artificial intelligence may also
