University education involves multiple challenges related to students’ professional development. In addition to providing students with theoretical and practical knowledge on a discipline, universities must promote ethical principles (Mason, 2001; Illingworth, 2004). For this purpose, ethics should be included in professional training to prevent university students from using their skills and knowledge to place their interests above their professional codes of conduct (Moore, 2006; Mion and Bonfanti, 2019). In these contexts, authors such as McCabe and Stephens have highlighted how fundamental it is to preserve academic integrity through the behavioral exercise of honor codes, accompanied by various commitments and responsibilities of different people who fulfill the role of promoters of quality professional training, so that problems of academic dishonesty are prevented (McCabe and Trevino, 1993; McCabe et al., 2006; Stephens et al., 2021). Unfortunately, issues such as plagiarism is one of the prevalent problems which entails a breach of professional ethics.

Plagiarism is the appropriation of words or ideas of other authors without giving them due credit, which also has academic, social, and legal repercussions (Park, 2003; Awasthi, 2019). Recently, research on university education highlights the risk of plagiarism due to its increasing ease of use to “solve” academic tasks, due to quick access to information that can be copied from one document and pasted into another (Kampa et al., 2024; Zhang, 2024), mainly in stressful school situations (Tindall et al., 2021). In addition to this, the ethical debate around plagiarism has been exacerbated by the arrival of artificial intelligence, so it is a phenomenon that must continue to be studied (Eaton, 2023; King, 2023). Plagiarism is a multifactorial phenomenon comprising cognitive, affective, contextual, sociocultural, and institutional variables (Husain et al., 2017; Moss et al., 2018), also found in professional scientific research (Pupovac and Fanelli, 2015). Systematic review (SR) studies on plagiarism have shown a wide range of computer tools for understanding this issue in depth (Moss et al., 2018; Awasthi, 2019). These SR studies have gathered evidence to describe and explain plagiarism, albeit without investigating educational interventions aimed at preventing or reducing plagiarism.

Interventions aimed at avoiding or reducing the incidence of plagiarism by university students primarily consists of conceptually raising awareness of the phenomenon and developing academic writing skills (Marusic et al., 2016). Some studies measured the effectiveness of their interventions in terms of the increase in students’ unfavorable attitudes toward plagiarism, knowledge about plagiarism, and plagiarism detection skills (Curtis et al., 2013; Rathore et al., 2018; Giuliano, 2022). However, these evaluations indirectly measured these variables, which may entail self-report or social desirability biases. Therefore, evidence on plagiarism prevention or reduction must be based on objective measurement criteria (Martin et al., 2009).

Objectively assessing plagiarism should involve detecting coincidences between paragraphs and words in texts prepared by participants and published documents, primarily using software specialized in this task, such as Turnitin (Dahl, 2007; Halgamuge, 2017; Meo and Talha, 2019). Experts should also identify paraphrasing problems in texts, lack of citations or mosaic plagiarism (directly copying and pasting the text and replacing only some words by synonyms to differentiate the text, also known as patchwriting) (Vieyra et al., 2013; Rogerson and McCarthy, 2017; Memon, 2020). These proposals enable us to more easily assess the effectiveness of interventions aimed at reducing plagiarism objectively.

Marusic et al. (2016) performed an SR assessing the effectiveness of interventions aimed at preventing research misconduct and promoting academic integrity in scientific publishing. Among the misconduct topics that they evaluated, plagiarism was found in university students, especially in undergraduate students. Their results showed that interventions based on information defining plagiarism and its consequences, academic integrity modules, feedback from plagiarism detection software, and practical academic writing exercises to promote citation and paraphrasing skills help to mitigate this problem. However, the authors also noted the low quality of evidence. This low quality derives from the lack of homogeneity in intervention techniques and from the use of self-report assessments in multiple studies.

In addition to the lack of objective evaluations and the heterogeneity in interventions to prevent or reduce plagiarism, recent technological advances have had a considerable impact on education in general and university education in particular, in terms of both academic integrity and misconduct (Turnbull et al., 2021). The increase in plagiarism has even been associated with the lack of direct supervision given the conditions of distance education in recent years (Eshet, 2023). Nevertheless, currently available computer tools facilitate efforts to objectively assess the effect of interventions. Accordingly, an SR should be conducted to synthesize knowledge on the effect of educational interventions aimed at preventing or reducing plagiarism by university students. In specific terms, it was also proposed to review how the effectiveness of such interventions has been evaluated objectively and which intervention strategies can be considered the most appropriate.

This review was organized based on the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) criteria (Page et al., 2021), in addition to a protocol registered in International Platform of Registered Systematic Review and Meta-analysis Protocols (Registration number INPLASY202370104).

We performed an SR to synthesize knowledge on the effectiveness of intervention programs aimed at reducing plagiarism by university students. In most studies, the intervention programs decreased the frequency of plagiarism, as shown by objective evaluation. However, the assessment and intervention procedures considerably varied across studies. Thus far, no SR on the effect of programs intended for reducing plagiarism by university students had been conducted specifically using objective evaluation. The closest topics on which similar reviews had been performed were related to academic dishonesty and included behaviors, such as cheating, data fabrication, and facilitation, in addition to plagiarism (Marusic et al., 2016; Chiang et al., 2022). Furthermore, prior studies have also differed in their design and evaluation procedures. In fact, our results corroborate the findings of two reviews showing that interventions can reduce plagiarism, albeit flagging considerable heterogeneity.

We identified Turnitin and SafeAssign as the most commonly used software programs in the studies reviewed. Other software programs are available to detect plagiarism, such as Viper, Grammarly, Plagiarisma, and Copygator, several of which recommend a percentage lower than 15 or 20% as the plagiarism threshold. However, finding similarity between texts does not necessarily equate plagiarism. Several classifications and types of plagiarism have been proposed, including copying and pasting, patchwriting, failing to add references, or misattribution, among others (Meo and Talha, 2019; Vrbanec and Meštrović, 2021). Additionally, the frequency of similarity may vary in different sections of a document. For example, a higher number of similarities to other documents are expected in the Introduction and Methods than in the Results and Discussion. While no consensus on the elements of plagiarism has been reached yet, researchers should use objective measures to avoid self-report bias, thereby improving the quality of research (Martin et al., 2009), and describe in detail the type and severity of plagiarism, in addition to identifying the section of the document under analysis (Belyy et al., 2018).

The intervention programs reviewed here similarly specified the definition of plagiarism and used specific and practical plagiarism prevention strategies, such as training in paraphrasing and referencing, providing examples, as reported in previous reviews. Those reviews have indicated that the most effective strategies are based on practical (Marusic et al., 2016), motivational, and environmental (Chiang et al., 2022) elements.

In contrast, intervention programs based on the theory of reasoned action advocate that fostering attitudes and subjective norms associating plagiarism with negative behavior and decreasing self-perceived control over the ease of plagiarism can reduce the intention to plagiarize and plagiarism. Nevertheless, these strategies may be inefficient given the complexity of the phenomenon. Students may plagiarize because they feel anxious, consider other values (for example, time) more relevant than plagiarism, overestimate their plagiarizing abilities, perform activities with tight deadlines, or disregard the usefulness of the academic activity, among other reasons. Moreover, environmental conditions may promote plagiarism, such as extenuating circumstances (for example, a sick family member), cultural factors (for example, deeming paraphrasing the author’s words disrespectful), implied consent, particularly if the conduct is prohibited, and arguing that others plagiarize as well. Plagiarism may even occur unconsciously, for instance, when reading a document and later thinking that your ideas are your own (Moss et al., 2018).

Sorea et al. (2021) summarized five categories of solutions to the problem of plagiarism, namely improving student training, empowering more engaged teachers, using anti-plagiarism software, enforcing clear anti-plagiarism policies, and educating young people on ethics. These solutions may be translated into general elements of the academic field, such as improving learning and teaching strategies, valuing activities promoting personal and professional development, encouraging collaboration and reducing competition, in addition to including specific elements to reduce plagiarism, such as conceptually defining plagiarism, teaching students appropriate referencing and paraphrasing strategies, highlighting the study reviewed to avoid plagiarism and using similarity detection software. Establishing the minimum number of elements that an intervention should include to reduce plagiarism requires conducting further research detailing its procedures and intervention length as well as using objective evaluation measures (Lendrum and Humphrey, 2012; Schultes, 2023).

In terms of limitations, although the six studies were carefully selected, the evidence derived from this SR may need to be complemented with other studies that expand the number of empirical foundations. In this sense, we recommend considering the results of this study and contrasting them with those that arise in future research, so that this knowledge can be strengthened or expanded. Other limitation of the present study lies in including only undergraduate students. We included only this population of students with the intention of providing the most precise evidence possible in terms of a target, but the effects of these interventions on graduate students and professional researchers should be assessed because plagiarism has also been reported in these populations (Pupovac and Fanelli, 2015). We also recommend assessing the effect of universities’ educational policies and sanctions and differentiating between voluntary and involuntary plagiarism (Bruton and Childers, 2016). Additionally, a key factor is the increasing incursion of artificial intelligence (AI) in education in recent years (Mijwil et al., 2023). Using AI, anyone can produce an apparently genuine document, which has not been previously published, albeit with a striking overlap with similar documents also produced using the same AI and for the same purpose (Misra and Chandwar, 2023). Therefore, future studies should also assess the impact of AI on plagiarism.

The results of the present review suggest that university education programs that share information about the characteristics and consequences of plagiarism; include academic integrity modules; promote plagiarism detection, citation, paraphrasing skills; and use similarity detection tools can reduce the frequency of plagiarism from literary sources by undergraduate university students. However, there is little research that evaluates plagiarism objectively and the interventions present a lot of heterogeneity, so it is necessary to carry out more research to reach a conclusion on the effectiveness of interventions to prevent or reduce plagiarism.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

RM-R: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Writing – original draft, Writing – review & editing. JS-N: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing. AR-R: Conceptualization, Methodology, Supervision, Writing – original draft, Writing – review & editing.