Perceptions of Academic Dishonesty: Insights from the University of Tehran

In recent times, Iran has seen an increase in various forms of academic dishonesty. The frequency of academic fraud, plagiarism, and cheating has led to efforts to rebuild the global reputation of Iranian academic institutions. We argue that academic dishonesty adversely affects not only the academic sphere but society as a whole, and addressing it requires an understanding of its various contributing factors. To this end, we undertook a study involving students from three different faculties at the University of Tehran. The study included 300 undergraduate students (182 females and 118 males), aged between 17 to 34 years (average age 20.55, standard deviation 2.04), from the Faculty of Psychology and Educational Sciences, Faculty of Management, and Faculty of Social Sciences. While only minor differences were observed among students from these three faculties, significant variations were noted in the perceptions of academic dishonesty among students from different academic years, including their views on professors' and peers' dishonest behaviors, and the justifiability of academic dishonesty.


Introduction
Academic dishonesty and misconduct has been one of the main issues that academia.Being too tough on it will disperse students (clients?)and overlooking it will likely culminate in debacle.This is why the majority of universities take a middle stance which is more likely inclined towards loosening monitoring and measures.Therefore, whether it is in the form of plagiarism, ghost authorship, cheating or fabricating data, etc., academic dishonesty or misconduct is prevalent all over the world.The advent of Internet has been a double edge sword (see for example Shahghasemi et al., 2023;Sarfi et al., 2021;Aeini, Zohouri & Mousavand, 2023).On one hand we can now track and find instances of academic dishonesty across languages and cultures, and on the other hand, the very nature of Internet has made it much easier to become a successful but unethical academician.This is not simply about the "copy-paste" capability of electronic resources or even "automatic paraphrasing" which make it almost impossible to track-down and hunt instances of plagiarism, but about a complex set of social, economic, political, and maybe racial factors that now govern the industry of academic writing and publishing.For example, as the prosperous world is now connected to the third world in a "high speed" manner, intellectual works are now easily and openly outsourced to the third world researchers (see for example Kapoor (2012); Havens & Lotz (2017)), and therefore tracking them down has become complicated.The introduction of AI-assisted article writing has only made this problem more complicated (Zohouri, Sabzali, & Golmohammadi, 2023).
One major concern in the academia has always been student academic dishonesty.Students are the future of the science and educating and correctly evaluating them is vital in continuity of scientific flourishment.This is why universities continuously update their terms of academic conduct, particularly in regard to utilizing new communication technologies, though this does not stop here.Now, rules are increasingly extended to other related fields like professors' relationships with students, receiving money in return for educational-research work, or self-plagiarism (see for example Roig (2015); Roig (2014); Pellegrini (2018) and Shahghasemi & Akhavan (2015)).
Iran is known as a culture in which "knowledge" has a special place.For centuries, Iranian and Persian literature has been full of poems, admonitions and advices that the human life is worth nothing without knowledge, and in our time Academic dishonesty and misconduct remain significant challenges in the academic world.Adopting an overly strict approach can alienate students (or clients), while neglecting the issue may lead to serious consequences.Consequently, most universities adopt a moderate approach, often erring on the side of relaxed monitoring and enforcement.Academic dishonesty, manifesting as plagiarism, ghost authorship, cheating, or data fabrication, is a global issue.The internet, while useful in detecting academic dishonesty across various languages and cultures, has also simplified the process of engaging in unethical academic practices.The challenges are not just due to the ease of copying and pasting or automatic paraphrasing that makes detecting plagiarism difficult, but also due to a complex mix of social, economic, political, and perhaps racial factors influencing academic writing and publishing.For instance, the easy accessibility of outsourcing intellectual work to researchers in developing countries complicates the tracking of original sources (refer to Kapoor (2012), Havens & Lotz (2017)).
Student academic dishonesty is a particular concern, as students represent the future of science.Ensuring their proper education and evaluation is crucial for the continued advancement of knowledge.Universities are constantly updating their academic conduct policies, especially in light of new communication technologies.These updates extend beyond mere student conduct, encompassing aspects such as faculty-student relationships, financial exchanges in educational research, and self-plagiarism (refer to Roig (2015), Roig (2014), Pellegrini (2018), Shahghasemi & Akhavan (2015)).
In Iran, where knowledge holds a special cultural significance, the quest for academic degrees is prominent.With around 4.5 million students in Iranian universities and another 100,000 studying abroad, the country's emphasis on education is notable, especially given its economic challenges and limited international relations.However, this rapid expansion has led to a shift from quality to quantity in education (see Varij Kazemi and Dehghan Dehnavi (2017)).Recent years have seen the Iranian academic community suffer from international incidents of plagiarism and misconduct.Notable instances include accusations by Declan Butler (2009) in Nature, and subsequent heightened scrutiny by organizations like Plagiarism Watch.These incidents have tarnished the reputation of Iranian academia, leading to the expulsion of students and professors and prompting initiatives like Professors against Plagiarism to monitor and report academic misconduct in international journals.
We believe that the issue of academic dishonesty should be tackled but we don't think expelling cheaters or whistle blowing is enough.We rather think providing an education which is rigorous and help students to become aware of what ethical writing and publication is and how to do it, is a vital and practical step in removing two main causes of academic misconduct in Iran -namely, lack of education about academic misconduct, and lack of fear of punishment.Certainly, when students are aware of what academic dishonesty is, professors will become more cautious not to cross red lines of ethical writing.Doing research on this subject will help us bring more light on this problem and authorities will probably be forced to "do something" about it.

Review of Literature
Several Iranian researchers have recently focused on academic misconduct and dishonesty, publishing their findings in international journals.These works, written in English, are accessible to the readers of this journal.However, in this section, we will only discuss studies published in Persian journals.We identified at least 24 such studies, but for brevity, we will only highlight five that are representative of the broader findings.Nakhaei & Nikpour (2005) conducted a study to evaluate the prevalence of research cheating among medical students using a comprehensive sampling approach.All 104 final-year medical students at their institution participated, completing a questionnaire that explored seven main types of academic misconduct.This questionnaire was developed based on methodology literature and focus group discussions with researchers.The students were asked to rate the frequency of each misconduct type and their personal views on a Likert scale.The analysis revealed that 37% of students reportedly fabricated data, while 40% manipulated data to yield desired results.Additionally, it was estimated that 25 to 50 percent of theses contained plagiarized material.Nakhaei and Nikpour warned that if these findings were indicative of a national trend, it would represent a significant issue in the educational system.University.They used Categorical Randomized sampling method to select 263 students.Their results showed that there was a significant difference between female and male attitudes towards Plagiarism.Based on students' major, there were also differences between the consequences of the Plagiarism and students' attitudes towards Plagiarism.In this study there was a strong relationship between religious orientation and students' attitudes towards plagiarism, and this is why Abedini and her colleagues recommend that empowering religiosity of the students would play an important role in reducing academic dishonesty.
Hemati Alamdarloo, Shojaee, Salimi, & Arjmandi (2017) compared plagiarism and its risk factors among talented and ordinary students at Shiraz University.Their statistical population included all students at Shiraz University and their sample size was consisted of 156 students (78 talented students and 78 ordinary students).The behavior of plagiarism questionnaire and effective factors on plagiarism questionnaire were used to measure plagiarism and its effective factors.
Using multivariable analysis of variance, they revealed that talented students were far less likely to commit different kinds of plagiarism and academic dishonesty.The y also found that attitude on plagiarism, self-efficiency, credentialism, lack of education on academic dishonesty, and lack of fear of punishment were among the most effective factors which contribute in plagiarism prevalence among students.Data collection took place in various campus locations such as classrooms, lobbies, and food courts.Here, students were informed about the study, and those who consented were given the questionnaire.We emphasized that participation was voluntary and their responses would have no impact on their grades or any other aspect of their academic life.

Method
Considering the sensitive nature of our research topic, we assured the students of complete anonymity, enabling them to respond to our questions candidly and without reservation.
It is a standard requirement for human-centric research to obtain an Institutional Review Board (IRB) certificate, demonstrating adherence to academic ethical standards.However, such a system is not established in Iran.Despite this, we made every effort to maintain the highest ethical standards in our research, as outlined above.

Measure
Research instrument was a self-report 22-item Questionnaire, which included five demographic questions and 17 items directly asking how much the respondents had experienced or witnessed academic dishonesty (hereafter AD) in their immediate academic environment; the respondents were also asked to what extend they themselves participated in AD of any kind.Respondents scored each item either on a 5-point Likert scale, ranging from 1 = completely disagree to 5 = completely agree, or on a 4-point scale ranging from 1 = never to 4 = always.Initial pool of items was gathered by scanning the literature and reported instruments in similar studies.We refined the order, content and response range of items through an interaction with masters and professors of the field to reach the last version.Some items were meaningful and could reflect an important aspect of our interest (such as: I know social problems that are related to my major).While some others were computed to represent a wider significant concept (such as: self-reported AD or exam cheating).

Results
Our study focused on examining students' and professors' perceptions of academic dishonesty (AD).We sought to gauge students' concern about societal issues and understand how they view their specific academic disciplines as tools for addressing these problems.Additionally, demographic data were utilized to explain variations in perceived AD.
Before conducting our primary analyses, we performed an exploratory analysis to identify outliers and assess the normality of our data distributions.We looked for both univariate and multivariate outliers, using criteria such as leverage, Cook's D, and Mahalanobis distance.These outliers were confined to the limits of what is considered the normal range, as outlined by Meyers, Gamst, and Guarino (2016).

Prevalence of AD
A set of frequency distributions is presented in Table 2. Rows represent answers range for each component of AD.Chisquare tests were used to diagnose any nonrandom difference between expected and observed frequencies.

Com.: completely
In the first item, we asked the degree to which students are aware of current problems related to their major.Responses were significantly gathered in 'agree' and 'completely agree' (χ 2 = 410.816,df= 4, p<.001).Many students thought they would have a contribution in their society (χ 2 = 256.301,df= 4, p<.001) and perceived their special knowledge as helpful (χ 2 = 160.600,df= 4, p<.001).For a significant number of students, having a rich CV and publication record was important (χ 2 = 147.800,df= 4, p<.001).
Some students thought that cheating in exams is acceptable, or they had no idea about it (completely agree= 34, agree= 58, and no idea= 62).Although, most answers were cumulated in disagree and completely disagree points (χ 2 = 24.696,df= 4, p<.001), but that number of agreeing or completely agreeing with cheating acceptability is considerable and could be an index of what exists in the society.In the same way, many students had no idea of contextual support for cheating or AD in their faculties (χ 2 = 97.167,df= 4, p<.001).Surprisingly, many others agreed (66 persons) or completely agreed (22 persons) with perceiving the existence of contextual support for cheating.Most students disagreed or completely disagreed they had copy-pasted from Internet (χ 2 = 53.291,df= 4, p<.001) or used others' writings without citation (χ 2 = 118.241,df= 4, p<.001).Nevertheless, the number of agreeing or even completely agreeing students is not ignorable (copy-pasting from Internet= 89 and copying without citation= 63).

Demographics and AD
In this research we included male and female students from different levels and groups (years and faculties).Here are our results for gender, major and level as factors.We used multivariate analysis of variance (MANOVA) to compare self_AD, classmate_AD, professors'_AD, contextual_AD, and plagiarism across these groups.Table 3 shows means separated by gender.Multivariate F was statistically significant (F 5, 274 = 2.455, p=.034) with an effect size of.043.This showed there was at least one mean difference (of the five dependent variable) between male and female.In order to examine difference source, we applied between subject test which runs univariate F-test for each dependent variable.As shown in the table, men (5.622) scored higher in self_AD than women (4.947) did (F= 9.357, p=.002, Eta Squared=.033).Men (7.819) outscored women (7.136) in reported classmate_AD (F= 7.077, p=.008, Eta Squared=.025),too.The Effect size was too small for both differences.There were no significant difference between men and women in terms of professors'_AD (F=.060,p=.807),contextual_AD (F=.142,p=.707),and plagiarism (F= 2.551,p=.111).Further considerations using multiple comparisons for means showed that first year students (4.625) perceived significantly less self_AD than their third (5.532) and fourth (6.180) year peers did.Second year students (5.222) expressed almost the same level of perceived self_AD as first year students.Perceived classmate_AD was also different among students.Students' level of belief in prevalence of AD grew significantly as they went from year one (6.384) to year two (7.444), from year two to year three (8.404), and from year three to year four (8.640).Interestingly, the changes in all three transformations are statistically significant.Students' reported levels of AD among professors grew significantly from the first (7.036) to the second year (7.958).It returned a bit back to a distance which was not significantly different from the first year (third= 7.596 and fourth= 7.600).In the case of contextual_AD which implies the degree to which students perceived their academic environment justified AD, we also have an interesting finding.Student's belief in existence of a somehow support for misconduct grew significantly up from the first (23.268) to the second year (25.167) and stayed almost flat (with a very slow positive slop) through the third and the fourth years.Further we went on to examine how students from different faculties (Social Sciences, Management, and Psychology) differed in terms of self_AD, classmate_AD, professors'_AD, contextual_AD, and plagiarism.Descriptive table (Table 8) summarizes the condition.We could see some differences, but were they big enough to be mentioned as a systematic variance?Let's review MANOVA test to find out.
Qeios, CC-BY 4.0 • Article, January 3, 2024 Qeios ID: O7T4SW • https://doi.org/10.32388/O7T4SW1/19 , Iranians are eager to show off by acquiring academic degrees.Today, about 4.5 million Iranians are studying in Iranian universities while another 100 hundred thousands are studying abroad.It's a big figure for a country of 80 million population with economic difficulties and limited international relations.Anyway, this bubble growth has entailed negative consequences including replacement of quality with quantity (see for example Varij Kazemi and Dehghan Dehnavi (2017) for detail).As a result, this seemingly bright profile has been racked in the last decade by several international debacles.First, DeclanButler (2009) from the famous journal Nature, accused several Iranian officials of plagiarism and academic misconduct, and after that, Iranians were frequently cited by bodies like Plagiarism Watch and others as more instances of plagiarism were identified.It was a great defamation for Iranian academia and many professors and students have been expelled by authorities and university deans in reaction; moreover, some Iranian scholars initiated a website named Professors against Plagiarism to monitor academic publishing work by Iranian scholars and whistle blow plagiarized papers when are published in international journals.
Zamani, Azimi, & Soleymani (2013) investigated factors influencing student plagiarism at Espahan University.They conducted a descriptive-survey study with a sample of 300 university students.Their research tool was a self-developed questionnaire, which showed that credentialism and the desire for better grades were the primary predictors of student plagiarism.Other factors included lack of self-efficacy, inadequate detection and punishment mechanisms for plagiarism, sociocultural influences, insufficient academic writing and ethics education, professors' indifference towards plagiarism, and a lack of fear of punishment.Jamshidi Boroujeni, Saeidi, & Heydari (2014) examined graduate students' awareness of plagiarism and its influencing factors at Shahid Chamran University of Ahvaz.They surveyed 354 students selected through random sampling, using a custom questionnaire.The study found a moderate level of plagiarism awareness among students.The primary causes of plagiarism and academic dishonesty were identified as research incompetence, economic means to hire ghostwriters, credentialism, failure to complete academic assignments, procrastination, and a lack of proper education in academic writing.Abedini, Khezrzadeh, & Zamani (2014) investigated the relationship between students' religious orientation, awareness of the consequences of academic dishonesty and their attitudes toward plagiarism and academic achievement.Their statistical population included students of the Espahan University and Espahan Medical Science

Table 1 .
females and 118 males) from the University of Tehran, aged between 17 and 34 years (mean age 20.55, standard deviation 2.04).These participants came from three different faculties: the Faculty of Psychology and Educational Sciences, the Faculty of Management, and the Faculty of Social Sciences.The criteria for inclusion in the study were: (a) current enrollment as an undergraduate student; (b) absence of any psychological disorder history; and (c) not being classified as a super senior or a student who has been expelled.Demographic details of the participants are provided in Table 1.Demographic characteristics of sampleTo ensure precision and representativeness, we employed proportionate stratified sampling, as described by Levy & Qeios, CC-BY 4.0 • Article, January 3, 2024 Qeios ID: O7T4SW • https://doi.org/10.32388/O7T4SW4/19ParticipantsThestudyinvolved300 undergraduate students(182 Lemeshow (2011).This approach not only prevented biases in our sample but also allowed for detailed analysis of different subgroups.Approximately 85% of the approached sample agreed to participate in our study.The remaining either did not meet our inclusion criteria, declined participation, or submitted incomplete questionnaires.

Table 4 .
Multivariate and between subject tests to examine mean differences between male and female

Table 5 .
Mean separated by education by years

Table 5
represents means of self_AD, classmate_AD, professors'_AD, contextual_AD, and plagiarism for groups with different level of education by year.We wanted to find out if there was a change in students' attitudes on AD across years.

Table 6 .
Multivariate and between subject tests to examine mean differences between students with different education by yearMultivariate F confirmed difference among groups (F 15, 825 = 2.455, p=.034) with a small but statistically significant effect size of.083.That is, students in different levels of education by year reported different levels of perceived AD, at least in one component.We subjected the data to a univariate F-test to see where the difference was.