The Impact of Artificial Intelligence Use on Enhancing Research Efficacy among Postgraduate Students: The Mediating Role of Research Technical Skills Using PLS-SEM

The Impact of Artificial Intelligence Use on Enhancing Research Efficacy among Postgraduate Students: The Mediating Role of Research Technical Skills Using PLS-SEM | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The Impact of Artificial Intelligence Use on Enhancing Research Efficacy among Postgraduate Students: The Mediating Role of Research Technical Skills Using PLS-SEM Ahmed Al Rantisi, Eiman Koofan This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8566987/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This study examined the impact of Artificial Intelligence (AI) use and Research Technical Skills (RTS) on research writing efficacy among postgraduate students at Dhofar University, Sultanate of Oman. A quantitative cross-sectional survey under a positivist paradigm was conducted. The purposive sample included 30 students from the Departments of Social Sciences and Education, selected for their familiarity with AI tools. Ethical standards were maintained, including voluntary participation, informed consent, confidentiality, and formal approval from the Research Department. The study tested null hypotheses asserting no significant effects of AI and RTS on research writing efficacy. Results revealed significant positive relationships. AI significantly enhanced RTS (β = 0.563; t = 3.700; p < 0.01), which in turn strongly influenced all dimensions of research writing efficacy: Background Writing (β = 0.822), Literature Review (β = 0.564), Methodology (β = 0.723), Analysis (β = 0.693), and Reporting (β = 0.667), all p < 0.01. Predictive relevance was confirmed through the Stone-Geisser Q² parameter. Findings highlight the critical role of AI and RTS in enhancing postgraduate research writing, emphasizing the need to integrate modern technological tools and technical skill development into higher education curricula to strengthen students’ academic performance and research competence. Artificial Intelligence Research Technical Skills Research Efficacy Postgraduate Students PLS-SEM Figures Figure 1 Figure 2 Introduction Environmental concerns have increasingly influenced research efficacy and academic behavior in recent years, particularly with the integration of Artificial Intelligence (AI) tools in research processes) Al-Rantisi et al., 2025(. AI has been shown to enhance various stages of scientific inquiry, such as literature reviews, experimental design, and manuscript preparation, thereby boosting the overall efficiency of research workflows (Lin et al., 2024; Bolaños et al., 2024). For instance, generative AI models can rapidly summarize literature and suggest relevant sources, significantly reducing the time and cognitive load required for academic writing, especially among non-native English speakers (LifeWire, 2021). Research efficacy is vital in the realm of research engagement because conducting research is a challenging and complex task that must follow a structured pattern to achieve reproducible results. Many scholars have noted that research efficacy is fundamental to the development of research skills. Students with high levels of research efficacy are more likely to engage in research activities, persist through challenges, and achieve higher competence in their research endeavours(Bandura, 1997). However, alongside these benefits, several studies raise concerns regarding potential over-reliance on AI. Excessive dependence on AI for writing and idea generation may impair critical thinking, diminish analytical skills, and lead to the degradation of core research competencies (Enciso 2025). Additionally, the phenomenon of AI hallucinations, where models produce fabricated or erroneous references or content, poses serious risks to research integrity and demands cautious oversight (Wikipedia, 2025). Amidst this dual-edged influence of AI, the development of researchers’ technical skills serves as a pivotal intermediary. Effective utilization of AI tools requires not only access to technology but also proficiency in navigating digital platforms, critically evaluating AI-generated outputs, and ethically integrating AI into research workflows. This technical skillset builds a bridge between AI adoption and the preservation of research quality and efficacy. Recent studies have explored the impact of AI on research skills and productivity among students in higher education. AI tools can enhance writing skills, critical thinking, and analysis (Arbab et al., 2024). However, their use raises concerns about academic integrity and potential loss of cognitive skills (Buniel et al., 2025). A training program on generative AI applications showed significant positive effects on lecturers' teaching capabilities in Oman (Al-Saiari et al., 2024). For STEM undergraduate students, AI dependence indirectly influences research productivity through the mediation of research skills, dispositions, and self-efficacy (Buniel et al., 2025). While AI offers potential benefits, students express uncertainties about its use and recognize the need for specialized skills and responsible application (Aguirre-Aguilar et al., 2024). Institutions are encouraged to develop standardized policies and regulations on AI usage in academic settings (Arbab et al., 2024). Building on these insights, the present study explores the relationship between AI use, research technical skills, and research efficacy, with a focus on postgraduate students in Oman. Employing a PLS-SEM approach, the study examines both the direct impact of AI utilization on research efficacy and the mediating role of research technical skills in this relationship. Literature review Studies on the utilization of artificial intelligence in research The integration of technology into research has continuously evolved through ongoing innovations. In recent years, AI which is designed to perform cognitive and problem-solving tasks, has received growing attention (Bonk & Wiley, 2020). Research indicates that AI contributes to a variety of functions within the research process (Ofem, Iyam, et al., 2024; Ofem, Nworgwugwu, et al., 2024). For example, AI applications such as ChatGPT can operate as literature review tools when given suitable prompts (Clark, 2020), assist in monitoring research progress (Swiecki, Ruis, Gautam, Rus, & Williamson Shafer, 2019), support personalized learning experiences (Chiu et al., 2022; Mertala, Fagerlund, & Calderon, 2022), automate the process of data collection (Hefernan & Hefernan, 2014), and conduct data analysis (Vij, Tayal, & Jain, 2020; Yuan, He, Huang, Hou, & Wang, 2020). AI also enables profiling of respondents’ backgrounds, which is essential in identifying disparities among participants (Cohen et al., 2017). In addition, AI tools facilitate academic writing by assisting in editing, generating text, providing language translation, and answering scholarly inquiries (Dwivedi et al., 2023; Kasneci et al., 2023; Lund et al., 2023). Previous research has highlighted the influence of artificial intelligence (AI) on academic research (Adarkwah et al., 2023; Crawford, Cowling, & Allen, 2023; Popenici, Rudolph, Tan, & Tan, 2023). For example, Ofem, Iyam, et al. (2024) reported that students frequently utilize AI in their research activities, with male students showing higher engagement compared to female students. Likewise, Tareq et al. (2023) found that AI adoption in higher education supports students in multiple research-related tasks, although concerns regarding reliability and gaps in skills were also identified. Moreover, Gasaymeh (2018) highlighted that students who have access to information and communication technologies are more capable of utilizing AI in their research compared to those without such access. Similarly, Devadas and Shilpa (2023) found that students’ willingness to adopt ChatGPT depends on multiple influencing factors. In addition, Strzelecki (2023) noted that although students are generally receptive to using ChatGPT in higher education, this acceptance is shaped by elements such as their technological skills and expectations regarding performance. In the Sultanate of Oman, the integration of artificial intelligence (AI) in academic research is still at an early stage, and many postgraduate students have yet to fully recognize its potential in enhancing research outcomes. This study, entitled The Impact of Artificial Intelligence Use on Enhancing Research Efficacy among Postgraduate Students in the Sultanate of Oman: The Mediating Role of Research Technical Skills Using PLS-SEM, was motivated by the need to provide empirical evidence that can inform policy decisions and encourage the adoption of AI tools, ultimately aiming to improve students’ research efficacy through the development of technical research skills. Studies on research efficacy Research efficacy is grounded in Bandura’s (1997) social cognitive theory, which emphasizes research technical skills as a determinant of individuals’ ability to complete challenging tasks successfully. Students with higher levels of research technical skills are more likely to persist, demonstrate resilience, and achieve better performance outcomes. Applied to research, skills encompass cognitive, affective, and psychomotor dimensions, reflected in tasks such as writing research backgrounds, conducting literature reviews, designing methodologies, analyzing data, and producing reports. Efficacy in literature review, for instance, is demonstrated through the ability to critically synthesize studies and identify gaps that guide new inquiries (Machi & McEvoy, 2016). Methodological efficacy involves designing robust and ethical research plans (Creswell & Creswell, 2018), while data analysis efficacy requires competence in statistical tools and interpretation skills to ensure validity. Report writing efficacy, in turn, allows researchers to communicate findings effectively to both academic and non-academic audiences (Field, 2018). Prior research has shown that research opportunities and structured mentoring programs enhance students’ confidence, critical thinking, and problem-solving abilities (Hunter et al., 2007; Lopatto, 2010). Supportive environments that provide guidance, feedback, and mentoring have been found to strengthen research efficacy and foster persistence in academic pursuits (Shadle, Marker, & Earl, 2017; Chemers et al., 2011). However, much of the literature still treats research efficacy as a one-dimensional construct and pays limited attention to how technological advancements, particularly artificial intelligence, may influence its development. Addressing this gap is critical to informing policies that aim to improve students’ research capacity and strengthen innovation in higher education. Conceptual framework This study examines the impact of artificial intelligence (AI) utilization on enhancing research efficacy among postgraduate students in the Sultanate of Oman, with a particular focus on the mediating role of research technical skills (RTS). Research efficacy is conceptualized as a multidimensional construct comprising five components: background writing efficacy (BWE), literature review efficacy (LRE), methodological efficacy (ME), analytical efficacy (AE), and report writing efficacy (RWE). Figure 1 presents the interrelationships among these variables and other relevant factors. Based on the study’s objectives and existing literature, the following main hypotheses were formulated: H01: The use of artificial intelligence has no statistically significant influence on research technical skills among postgraduate students. H02: Research technical skills have no statistically significant influence on background writing efficacy among postgraduate students. H03: Research technical skills have no statistically significant influence on literature review efficacy among postgraduate students. H04: Research technical skills have no statistically significant influence on methodological efficacy among postgraduate students. H05: Research technical skills have no statistically significant influence on analytical efficacy among postgraduate students. H06: Research technical skills have no statistically significant influence on reporting writing efficacy among postgraduate students. H1 H1 H05 Notes : The model illustrates how the use of artificial intelligence influences research efficacy across its five dimensions: background writing efficacy (BWE), literature review efficacy (LRE), methodological efficacy (ME), analytical efficacy (AE), and report writing efficacy (RWE), both directly and indirectly through the mediating role of research technical skills (RTS). Methods The study adopts a positivist research paradigm, relying primarily on quantitative methods. A cross-sectional survey design was employed to collect data from a targeted population at a specific point in time (Ofem et al., 2024). The sample was purposively selected from students enrolled in the Department of Social Sciences and the Department of Education at the College of Arts and Applied Sciences, Dhofar University, Sultanate of Oman. These departments were chosen because they host the largest number of postgraduate students in the college and their curricula ensure that students are generally well-informed about modern technologies, including artificial intelligence tools. A total of 30 students participated in the study, reflecting the proportional representation of these two departments and their relevance to the research context. Ethical considerations were strictly observed throughout the study. Participation was voluntary, and all respondents provided informed consent before taking part in the survey. Respondents were assured of the confidentiality and anonymity of their responses, and they were informed of their right to withdraw from the study at any time without any consequences. The authors obtained formal approval, IRB#DU/AY/2025-26/QUES-002, from the Research Department at Dhofar University to conduct the survey among the students. All data were collected during the Fall semester of 2025–2026. The study adhered to ethical guidelines in accordance with the principles outlined in the Helsinki Declaration and other relevant research ethics standards. Measures/instruments The present study incorporated seven main variables: the use of artificial intelligence (AI Use), research technical skills (RTS) as a mediating variable, background writing efficacy (BWE), literature review efficacy (LRE), methodological efficacy (ME), analytical efficacy (AE), and research writing efficacy (RWE) as response variables. The research instrument was divided into two sections. Section A collected demographic information, including gender, age, academic discipline, academic stage, prior training in artificial intelligence, and frequency of using AI tools in research. This section also included spaces for written informed consent, as well as email addresses and phone numbers for online submission via Google. Forms. Section B contained items measuring the explanatory, mediating, and response variables. The explanatory variable, AI utilization in research, was assessed using a four-item scale adapted from Ofem, Iyam, et al. (2024), which demonstrated robust psychometric properties and was recently used in a similar context. This construct included items reflecting students’ use of AI in research-related tasks, such as supporting background writing, finding appropriate references, rephrasing or refining research texts, and assisting with data organization and analysis. Sample item: “I use AI tools to support the writing of the theoretical background of my research ”. All items were rated on a four-point linear scale. The mediating variable, research technical skills (RTS), was measured with four items capturing students’ competence in applying digital and technological tools to different stages of research. The items reflected their ability to utilize online research tools effectively, assess the reliability of AI-generated outputs, employ statistical or qualitative analysis software, and integrate AI ethically and responsibly into academic work. Sample item: “I can evaluate the reliability of AI-generated outputs for research purposes”. The response variable, research efficacy, was conceptualized as the confidence and belief in one’s ability to successfully conduct research activities, and was examined across five dimensions, each measured with four items. Background Writing Efficacy (BWE) refers to the ability to develop a coherent and well-contextualized research background that establishes the significance of the topic. Sample item: “I can write a coherent research background that highlights the importance of my topic”. Literature Review Efficacy (LRE) reflected the skills needed to critically evaluate and synthesize relevant academic literature, as well as to identify research gaps. Sample item: “I can identify research gaps accurately through reviewing the literature”. Methodological Efficacy (ME) captured students’ competence in selecting research designs, constructing tools, and applying correct procedures. Sample item: “I can design appropriate research instruments such as questionnaires or interviews”. Analytical Efficacy (AE) measured the ability to analyze and interpret data accurately, draw logical conclusions, and compare findings with prior studies. Sample item: “I can analyze research data correctly using the appropriate tools”. Finally, Report Writing Efficacy (RWE) assessed the extent to which students could communicate research findings clearly, structure reports logically, and present results in line with academic standards. Sample item: “I can write a comprehensive research report in line with academic standards”. Content validity The scale items underwent quantitative validation by five experts from different fields, who evaluated the relevance, clarity, and suitability of the items for measuring the study’s constructs. Initially, 30 items were developed based on an extensive review of the literature and adaptations from previous research. Following the initial screening process, two items were excluded. The experts were provided with the questionnaire and an assessment rubric, following the guidelines recommended by specialists (Ofem, Iyam, et al., 2024; Ofem, Nworgwugwu, et al., 2024). Each item was rated on a five-point Likert scale, where 1 indicated “strongly disagree” and 5 indicated “strongly agree,” reflecting the degree to which each item appropriately captured the intended construct. The Item-Content Validity Index (I-CVI) was calculated to determine which items should be retained. I-CVI values range from 0 to 1, with a threshold of 0.78 or higher generally considered acceptable, indicating strong agreement among the expert panel regarding the relevance and clarity of the items. Some researchers, however, consider values as low as 0.70 acceptable under certain conditions, such as smaller panel sizes or strict item development criteria (Haynes, Richard, & Kubany, 1995). Data Analysis A purposeful sample of 30 postgraduate students from the departments of Social Sciences, Education, Mathematics, and Science in the College of Arts and Applied Sciences at Dhofar University, which are the departments that have postgraduate programs. The data were analyzed using SPSS for descriptive statistics and SmartPLS for structural equation modeling (PLS-SEM). The descriptive analysis provided an overview of the participants’ demographic characteristics and responses to the study variables. To test the hypothesized relationships among AI utilization, research technical skills, and research efficacy, the partial least squares structural equation modeling (PLS-SEM) approach was employed. PLS-SEM is a widely used method in social science research for examining theoretically validated additive and linear causal models (Hair et al., 2013). It allows for the simultaneous estimation of both the measurement model, which assesses the validity and reliability of the constructs, and the structural model, which tests the hypothesized relationships among variables. SmartPLS software was used to perform all analyses (Ringle et al., 2015). The evaluation of the measurement model ensured the reliability and validity of the constructs, while the structural model assessment focused on the statistical significance of the hypothesized paths, following the guidelines proposed by Chin (2009) and Hair et al. (2013). This approach guarantees that the study’s findings are both statistically robust and theoretically grounded. Results Descriptive statistics The demographic characteristics of the respondents are summarized as follows: regarding gender, [8] (26.7%) were male and [22] (73.3%) were female; concerning age, [6] (20.0%) were below 25 years, [11] (36.7) were between 25 and 30 years, [8] (26.7%) were between 31 and 35 years and [5] (16.7%) were above 35 years; in terms of academic disciplines, [14] (46.7%) were from Social Sciences Department, [16] (53.3%) from Education Department; regarding academic stage, [22] (73.3%) were in the coursework stage and [8] (26.7%) were in the thesis preparation stage; concerning previous training on artificial intelligence, [18] (60.0%) had received training and [12] (%40.0) had not received any training; in terms of frequency of using artificial intelligence tools in research, [9] (30%) used them occasionally, [15] (50%) used them frequently, and [6] (20%) used them always, every participant provided informed consent. Measurement model evaluation To validate the measurement model, the following steps were taken: (1) calculating the factorial load for reflective constructs (AI, RTS, BWE, LRE, ME, AE, RWE) to determine the item's individual reliability; and (2) determining the construct's validity. To assess whether the questionnaire was sufficiently reliable, many experiments were run. The purpose of the first test was to decide whether to accept a certain item as a component of a reflective construct. The items needed to have a factorial load (𝛌) or simple correlations that were at least 0.707 in order to serve this function (Carmines & Zeller, 1979). Due to weak loads (≤ 0.7), three items (AI-03, RTS-04, and ME-01) were removed. The remaining items, however, were higher than 0.7. Due to the remaining items' contribution to content validity (AVE = 0.5; P C = 0.7), they were kept in place for this investigation. Table 1 displays the outcomes. The Composite Reliability Index ( P C ) and Cronbach's alpha (α) were used in the second test to evaluate the construct's internal consistency. Using the suggestion of Hair et al. (2013), suggesting 0.7 as a reference point for both α and P C . The results of Table 1 show that all constructs were dependable and had satisfactory internal consistency. A PLS instrument that was generated using the AVE was shown to be reliable thanks to the convergent validity test. The AVE coefficient for reflecting structures was higher than 0.5, as shown in Table 1 , which summarizes the results. According to these suggestions, all AVE measures were valid. Table 1 Structural model measurement tool: convergent reliability and validity. Construct Measurement item Loading CA CR AVE AI AI-01 0.804 0.926 0.949 0.822 AI-02 0.794 AI-03 Item deleted a AI-04 0.823 RTS RTS-01 0.702 0.773 0.847 0.584 RTS-02 0.766 RTS-03 0.905 RTS-04 Item deleted b BWE BWE-01 0.809 0.834 0.884 0.656 BWE-02 0.796 BWE03 0.877 BWE-04 0.753 LRE-01 0.816 LRE LRE-02 0.847 0.851 0.898 0.688 LRE-03 0.797 LRE-04 0.856 ME ME-01 ME-02 ME-03 ME-04 Item deleted c 0.724 0.837 0.789 0.737 0.831 0.555 AE AE-01 0.812 0.772 0.842 0.575 AE-02 0.873 AE-03 0.957 AE-04 0.975 RWE RWE-01 0.928 0.937 0.954 0.838 RWE-02 0.935 RWE-03 0.914 RWE-04 0.884 AI-01: I use artificial intelligence tools to support the writing of the theoretical background of my research; AI-02: Artificial intelligence applications help me find suitable references and scientific sources; AI-03: I benefit from artificial intelligence in drafting or rephrasing research texts; AI-04: I use artificial intelligence to analyze or organize data effectively; RTS-01: I am able to use electronic research tools efficiently; RTS-02: I am capable of evaluating the reliability of the outputs generated by artificial intelligence for scientific research purposes; RTS-03: I possess skills in using specialized software for statistical or qualitative analysis; RTS-04: I can ethically and properly integrate digital tools and artificial intelligence into research; BWE-01: I can write a coherent research background that highlights the importance of my topic; BWE-02: I am able to formulate the research problem clearly based on the literature; BWE-03:I can summarize the main points of previous studies in a logical and coherent manner; BWE-04: I am able to present the research background in a way that makes it easy for readers and other researchers to understand; LRE-01: I am able to summarize previous studies and present them critically; LRE-02: I can accurately identify research gaps through a review of the literature; LRE-03: I am able to link the literature to the research problem in a systematic way; LRE-04: I can evaluate the reliability of the academic sources used in the research; ME-01: I am able to choose the appropriate research methodology for my topic; ME-02: I have the ability to design suitable research instruments (questionnaire, interview, etc.); ME-03: I can identify the appropriate statistical or qualitative sample for the study; ME-04: I am able to apply research procedures correctly and systematically; AE-01: I am able to analyze research data correctly using appropriate tools; AE-02: I can interpret research findings in a logical and objective manner; AE-03: I am able to critically compare the results with previous studies; AE-04: I can identify reliable conclusions based on the analyzed data; RWE-01: I am able to write a comprehensive research report in accordance with scientific research standards; RWE-02: I have the ability to present results and discussions in a clear academic manner; RWE-03: I am able to formulate recommendations and conclusions in a clear and results-based way; RWE-04: I can organize the research report logically and sequentially, making it easy for readers to follow; CA: Cronbach’s alpha; CR, composite reliability; AVE: variance extracted. a AI-03, b RTS-04, c ME-01, were deleted due to low factorial loadings. Table 2 Measuring instrument: discriminant validity. AE AI BWE LRE ME RTS RWE AE 0.907 AI 0.611 0.812 BWE 0.410 0.343 0.810 LRE .0734 0.478 0.713 0.829 ME 0.610 0.230 0.587 0.583 0.795 RTS 0.822 0.563 0.564 0.723 0.693 0.796 RWE 0.533 0.303 0.743 0.649 0.713 0.667 0.916 Diagonal: square root of the average variance extracted (AVE). Finally, it is necessary to assess the discriminant validity. For this purpose, the (Fornell & Larcker, 1981) technique, which is founded on the notion that a construct should share more variance with its items than with other constructs in each model, was utilized. Table 2 demonstrates that the square root of the AVE was higher than the correlation between the variables in this regard. This indicates that all constructions had a stronger relationship with their own objects than they did with those of other constructs. Figure 2 : shows the structural model measurement tool: convergent reliability and validity. Structural model evaluation Once the reliability and validity of the constructs had been verified, an analysis of the relationship between the constructs and the predictive capacity of the endogenous variables was carried out. Therefore, we assessed the weight and nature of the relationships (hypothesis) between the different variables. This assessment involved the use of two basic indicators: the explained variance ( R 2 ), which indicates the predictive power of the model, and the standardized path coefficients ( β ), which indicate the strength of the relationships between dependent and independent variables (Johnson, Herrmann et al. 2006) Regarding the predictive capacity of the model, the explained variance ( R² ) of the dependent variables should be equal to or greater than 0.10, since lower values provide little information (Falk & Miller, 1992). According to Chin (1998), R² values of 0.67 or higher are considered substantial, 0.33 to 0.66 are moderate, and 0.19 to 0.32 are weak. The results of this study show that the model demonstrates moderate predictive power for Analytical Efficacy (AE, R² = 0.642), Literature Review Efficacy (LRE, R² = 0.473), Methodological Efficacy (ME, R² = 0.475), and Reporting Writing Efficacy (RWE, R² = 0.401). In contrast, Background Writing Efficacy (BWE, R² = 0.262) and Research Technical Skills (RTS, R² = 0.263) showed weak predictive power, although still above the recommended threshold of 0.10. The cross-validated predictive ability test (CVPAT) has been introduced as an alternative to PLSpredict for assessing the predictive performance of PLS-SEM models. Initially developed by Liengaard et al. (2021) and later extended by Sharma et al. (2023), CVPAT employs an out-of-sample prediction approach that compares the model’s average prediction loss with two benchmarks: indicator averages (IA) and a linear model (LM). Superior predictive capability is demonstrated when the model’s average loss is significantly lower than these benchmarks. In SmartPLS, CVPAT results are integrated into the PLSpredict report, ensuring comparability of results across approaches (Shmueli et al., 2016, 2019; Hair et al., 2022). In the present study, the Q² _predict values obtained through PLSpredict and CVPAT confirm that all dependent variables show some degree of predictive relevance. Specifically, Analytical Efficacy (AE; Q² = 0.146), Research Technical Skills (RTS; Q² = 0.132), and Literature Review Efficacy (LRE; Q² = 0.103) demonstrated the strongest predictive validity, whereas Background Writing Efficacy (BWE; Q² = 0.051), Methodological Efficacy (ME; Q² = 0.020), and Reporting Writing Efficacy (RWE; Q² = 0.012) exhibited weaker yet positive predictive capacity (see Table 3 ). Table 3 Summary of variance explained ( R 2 ). PLSpredict & CVPAT:( Q²predict ). Construct R 2 Q²predict AE 0.642 0.146 BWE 0.262 0.051 LRE 0.473 0.103 ME 0.475 0.020 RTS 0.263 0.132 RWE 0.401 0.012 Predictive Ability of the Model (PLSpredict and CVPAT): The predictive ability of the model was assessed out-of-sample using PLSpredict and CVPAT for all endogenous variables in the study. The results indicate that all variables demonstrated at least some predictive power. For example, Analytical Efficacy (AE), Literature Review Efficacy (LRE), and Research Technical Skills (RTS) exhibited moderate predictive power, with positive Q² _predict values of 0.146, 0.103, and 0.132, respectively, indicating that the model can reasonably predict future values for these variables. Other variables, including Background Writing Efficacy (BWE), Methodological Efficacy (ME), and Reporting Writing Efficacy (RWE), showed weak but positive predictive power, with Q² _predict values of 0.051, 0.020, and 0.012, respectively, indicating a minimal yet acceptable level of predictiveness out-of-sample. Regarding error indicators, the RMSE and MAE values are consistent with the Q² _predict results. Lower RMSE and MAE values for variables with higher predictive power (e.g., RTS and LRE) and slightly higher values for variables with weaker predictive power (e.g., ME and RWE) further support the validity of the model’s predictive assessment. Table 4 Structural model results. Hypothesis Suggested effect Path coefficients t-value (bootstrap) p-value Support H01: ‏+AI → + RTS + 0.563*** 3.700 0.000 Support H1 H02: ‏+RTS → + BWE + 0.822*** 15.948 0.000 Support H1 H03: ‏+RTS → + LRE + 0.564*** 6.241 0.000 Support H1 H04: ‏+RTS → + ME + 0.723*** 13.479 0.000 Support H1 H05: ‏+RTS → + AE + 0.693*** 9.715 0.000 Support H1 H06: ‏+RTS → + RWE + 0.667*** 6.885 0.000 Support H1 In line with the formulated null hypotheses, the study tested whether Artificial Intelligence (AI) and Research Technical Skills (RTS) exerted no statistically significant influence on the dependent constructs. However, the empirical findings revealed otherwise. The relationship between AI and RTS among postgraduate students was found to be positive and significant ( β AI → RTS = 0.563; t = 3.700; p < 0.01), thereby rejecting H01. Similarly, RTS demonstrated a strong and significant positive effect on Background Writing Efficacy (BWE) ( β RTS → BWE = 0.822; t = 15.948; p < 0.01), leading to the rejection of H02. In the same vein, RTS significantly predicted Literature Review Efficacy (LRE) ( β RTS → LRE = 0.564; t = 6.241; p < 0.01), resulting in the rejection of H03. Furthermore, RTS had a substantial positive influence on Methodological Efficacy (ME) ( β RTS → ME = 0.723; t = 13.479; p < 0.01), which rejects H04. The analysis also confirmed a significant positive relationship between RTS and Analytical Efficacy (AE) ( β RTS → AE = 0.693; t = 9.715; p < 0.01), rejecting H05. Finally, RTS was shown to significantly enhance Reporting Writing Efficacy (RWE) ( β RTS → RWE = 0.667; t = 6.885; p < 0.01), thereby rejecting H06. To measure the predictive power of model-dependent constructs, the Stone–Geisser procedure or Q 2 parameter (cross-validated redundancy) was used. This test was run using the blindfolding technique. Parameter Q 2 must be greater than zero to have predictive validity (Chin, 1998) since values above zero determine that the predictability of the model is relevant (Sellin & Versand, 1995). As Table 3 shows, Q 2 value met this condition. Therefore, the predictive relevance of the model in relation to endogenous latent variables was supported. Finally, we calculated the Standardized Root Mean Square Residual (SRMR), which is the average difference between the predicted and observed correlations (variances and covariances) based on the standard error of the residual (Henseler & Sarstedt, 2013). To ensure the fit of the structural model, we calculated the GOF index "match quality index". It is a general indicator defined as follows: \(\:GOF=\sqrt{\left(\stackrel{-}{{R}^{2}}*\stackrel{-}{AVE}\right)}\) The value of \(\:\stackrel{-}{{R}^{2}}=0.878\:\:\:،\:\stackrel{-}{AVE}\:=0.629\) . Therefore, the GOF value is equal to 0.682, which is higher than 0.36 according to Wetzels et al. (2009) which indicates the fit of the proposed structural model. Discussion The present study examined the impact of Artificial Intelligence (AI) on enhancing research efficacy among postgraduate students in the Sultanate of Oman, with research technical skills (RTS) serving as a mediating factor. The results revealed a significant positive relationship between AI use and RTS, indicating that students who engage more with AI tools develop stronger technical research capabilities. This finding aligns with prior studies emphasizing the growing role of AI in research processes, including literature review, data collection, monitoring research progress, and academic writing (Bonk & Wiley, 2020; Clark, 2020; Swiecki et al., 2019; Dwivedi et al., 2023; Kasneci et al., 2023; Lund et al., 2023). The results also underscore that AI facilitates the development of cognitive and problem-solving skills essential for conducting high-quality research. Furthermore, the study found that RTS had a significant positive effect on all dimensions of research efficacy, including Background Writing Efficacy (BWE), Literature Review Efficacy (LRE), Methodological Efficacy (ME), Analytical Efficacy (AE), and Reporting Writing Efficacy (RWE). This is shown in the results in Fig. 2 . These findings resonate with Bandura’s (1997) social cognitive theory, which posits that technical skills and self-efficacy are crucial determinants of individuals’ ability to complete challenging tasks. The results also align with prior research highlig,hting that structured research opportunities, guidance, and mentoring enhance students’ confidence, critical thinking, and problem-solving abilities (Hunter et al., 2007; Lopatto, 2010; Shadle et al., 2017; Chemers et al., 2011). In relation to specific research tasks, RTS significantly predicted students’ ability to synthesize literature, design methodological frameworks, analyze data, and produce reports. This supports the findings of Machi and McEvoy (2016) regarding literature review efficacy, Creswell and Creswell (2018) regarding methodological planning, and Field (2018) regarding reporting skills. The current results extend these findings by demonstrating that AI use amplifies the benefits of technical skills, enabling students to perform these tasks more effectively. The study further highlights the moderating role of technological readiness and access. Previous research indicated that students with greater access to ICT infrastructure are better able to leverage AI in their research activities (Gasaymeh, 2018; Ofem, Iyam, et al., 2024). Similarly, Tareq et al. (2023) and Strzelecki (2023) emphasized that students’ engagement with AI is influenced by their prior skills, expectations, and confidence in using these tools. The present study confirms that in the Omani context, where AI adoption is still nascent, students’ RTS mediates the effective integration of AI, suggesting that developing technical skills should be a priority in postgraduate programs. Moreover, AI was shown to support multiple facets of the research process, from monitoring progress and automating data collection to enabling personalized learning experiences (Chiu et al., 2022; Mertala et al., 2022; Hefernan & Hefernan, 2014; Vij et al., 2020; Yuan et al., 2020). These functionalities contribute to enhancing research efficacy by providing timely feedback, reducing errors, and increasing productivity, which is consistent with prior literature emphasizing the cognitive and practical benefits of AI in research (Dwivedi et al., 2023; Kasneci et al., 2023; Lund et al., 2023). The findings also address concerns raised by previous studies regarding AI adoption in higher education. For example, Tareq et al. (2023) reported that gaps in technical skills and reliability issues may limit students’ ability to fully benefit from AI. In line with this, the present study demonstrates that while AI has potential, its effectiveness is contingent upon students’ RTS. Thus, universities should focus on capacity-building initiatives, training programs, and infrastructure provision to ensure equitable and effective AI adoption. Overall, this study extends existing knowledge by empirically confirming that AI enhances postgraduate research efficacy through the development of technical research skills. It bridges the gap in the literature regarding the interplay between technological tools and research efficacy, particularly in contexts where AI adoption is emerging, such as in Oman. The results have both theoretical and practical implications, highlighting the importance of integrating AI and skill development into postgraduate curricula to strengthen research competencies, foster innovation, and support academic success. Limitations and Future Studies The results of the present study should be interpreted with caution due to several limitations related to the research design and sample. First, the cross-sectional design limits the ability to establish causal relationships between the variables under investigation. Future studies could employ longitudinal or experimental designs to better examine potential causal links between AI utilization, research technical skills, and research efficacy among graduate students. Second, the sample consisted of only 30 postgraduate students from some departments within the College of Arts and Applied Sciences at Dhofar University. This relatively small and convenience-based sample may limit the generalizability of the findings to the wider population of graduate students at the university or other higher education institutions in Oman. Increasing the sample size and including students from multiple universities would help address this limitation in future research. Third, the use of a self-administered questionnaire may introduce response biases. Although participants provided genuine answers, they might have tended to respond in ways perceived as socially desirable or aligned with academic expectations (Sekaran & Bougie, 2009). Future studies could mitigate this issue by collecting data from multiple sources, including interviews, focus groups, or observational measures. Finally, contextual challenges specific to Dhofar University and Omani higher education may have influenced the responses. These include variations in students’ prior exposure to AI tools, differences in access to digital resources across departments, and varying levels of technical support or guidance provided by faculty. Future research should consider these factors and explore how institutional and technological infrastructure affects students’ AI utilization and research competencies. Conclusions This study demonstrated that Artificial Intelligence (AI) significantly enhances postgraduate students’ Research Technical Skills (RTS), which in turn positively influence multiple dimensions of research efficacy, including background writing, literature review, methodology, analysis, and reporting. The findings confirm that AI facilitates cognitive and practical research tasks, supporting Bandura’s (1997) social cognitive theory on skill development and task performance. The study also highlights that students’ technological readiness and access to AI tools are critical for maximizing these benefits, consistent with previous research (Gasaymeh, 2018; Ofem, Iyam, et al., 2024). Practically, universities should integrate AI applications and provide structured training to strengthen research competencies. Policy interventions should focus on equitable access and guidance for AI adoption. Overall, the study emphasizes that AI, combined with technical skill development, can serve as a transformative tool to enhance research efficacy and academic performance among postgraduate students, particularly in emerging contexts such as the Sultanate of Oman. Declarations Author Contribution Eiman M. Koofan was responsible for developing the theoretical framework, conducting the literature review, and collecting the study data. Ahmed M. Al Rantisi carried out the PLS-SEM data analysis, interpreted the results, and led the drafting and critical revision of the manuscript. References Adarkwah, M. A., Amponsah, S., Micheal, W., Ronghuai, D. D., Ahmed, T. B., Ahmed, H., et al. (2023). Awareness and acceptance of ChatGPT as a generative conversational AI for transforming education by Ghanaian academics: A two-phase study. Journal of Applied Learning & Teaching, 2(6). http://journals.sfu.ca/jalt/index.php/jalt/index . Al-Rantisi, A. M., ElShanti, N. H., & Harb, S. A. (2025). Challenges of using artificial intelligence in social work education. Social Work Education , 1–18. https://doi.org/10.1080/02615479.2025.2483354 Aguirre-Aguilar, G., Esquivel-Gámez, I., & Edel-Navarro, R. (2024). AI in the development of research skills in postgraduate studies. Alteridad: Revista de Educación, 19 (2), 161–176. https://doi.org/10.17163/alt.v19n2.2024.8557 Arbab, A. N., Al-Saiari, M. A., Al-Mughairi, Y. M., Al-Mashaikhi, B. N., & Mudhsh, B. A. (2024). Student's utilization and assistance of AI tools in assessment completion: Perceptions and implications. Qubahan Academic Journal, 4 (3), 315–332. https://doi.org/10.48161/qaj.v4n3a760 Al-Saiari, M. A., Al-Mughairi, Y. M., Al-Mashaikhi, B. N., & Mudhsh, B. A. (2024). Investigating the impact of training program on generative AI applications in improving university teaching. Qubahan Academic Journal, 4 (3), 315–332. https://doi.org/10.48161/qaj.v4n3a760 Bandura, A. (1997). Self-efficacy: The exercise of control. W.H. Freeman and Company. Bonk, C. J., & Wiley, D. A. (2020). Preface: Reflections on the waves of emerging learning technologies. Educational Technology Research & Development, 68 (4), 1595–1612. https://doi.org/10.1007/s11423-020-09809-x Bolaños, F., Salatino, A., Osborne, F. (2024). Artificial intelligence for literature reviews: opportunities and challenges. 57, 259. https://doi.org/10.1007/s10462-024-10902-3 Buniel, J. M., Intano, J., Cuartero, O., & Grustan, K. J. (2025). Modeling the influence of AI dependence on research productivity among STEM undergraduate students: Case of a state university in the Philippines. Frontiers in Education, 10 , 1535466. https://doi.org/10.3389/feduc.2025.1535466 Carmines, E. G., & Zeller, R. A. (1979). Reliability and validity assessment . Sage Publications. Chemers, M. M., Zurbriggen, E. L., Syed, M., Goza, B. K., & Bearman, S. (2011). The role of efficacy and identity in science career commitment among underrepresented minority students. Journal of Social Issues, 67 (3), 469–491. https://doi.org/10.1111/j.1540-4560.2011.01710.x Chin, W. W. (2009). How to write up and report PLS analyses. In Handbook of partial least squares: Concepts, methods and applications, Springer, pp. 655-690. Chin, W. W. (1998). Commentary: Issues and opinion on structural equation modeling, JSTOR : vii-xvi. Chiu, T. K. F., Meng, H., Chai, C. S., King, I., Wong, S., & Yeung, Y. (2022). Creation and evaluation of a pre-tertiary artificial intelligence (AI) curriculum. IEEE Transactions on Education, 65 (1), 30–39. https://doi.org/10.1109/TE.2021.3085878 Clark, D. (2020). Artificial intelligence for learning: How to use AI to support employee development. Kogan Page Publishers. Cohen, I. L., Liu, X., Hudson, M., Gillis, J., Cavalari, R. N., Romanczyk, R. G., … Gardner, J. M. (2017). Level 2 screening with the PDD behavior inventory: Subgroup profiles and implications for differential diagnosis. Canadian Journal of School Psychology, 32 (3–4), 299–315. https://doi.org/10.1177/0829573517721127 Crawford, J., Cowling, M., & Allen, K. A. (2023). Leadership is needed for ethical ChatGPT: Character, assessment, and learning using artificial intelligence (AI). Journal of University Teaching and Learning Practice, 20 (3), 2. https://doi.org/10.53761/1.20.3.02 Creswell, J. W., & Creswell, J. D. (2018). Research design: Qualitative, quantitative, and mixed methods approaches (5th ed.). SAGE Publications. Dwivedi, Y. K., Kshetri, N., Hughes, L., Slade, E. L., Jeyaraj, A., Kar, A. K., … Wright, R. (2023). “So what if ChatGPT wrote it?” Multidisciplinary perspectives on opportunities, challenges and implications of generative conversational AI for research, practice, and policy. International Journal of Information Management, 71 , Article 102642. https://doi.org/10.1016/j.ijinfomgt.2023.102642 Enciso, R. (2025). Artificial Intelligence in Research: Enhancing Efficiency or Compromising Integrity: A Systematic Review. 7, https://doi.org/ 10.13140/RG.2.2.17590.41289 Falk, R. F. and N. B. Miller (1992). A primer for soft modeling, University of Akron Press Field, A. (2018). Discovering statistics using IBM SPSS Statistics (5th ed.). SAGE Publications Fornell,C.G. & Larcker,D.F. (1981). Evaluating structural equation models with unobservable variables and measurement error’. Journal of Marketing Research , 18 (1), 39–50. Gasaymeh, A. M. M. (2018). The impact of digital storytelling on the quality of learning outcomes, motivation, and engagement among university students. Journal of Educational Technology & Society, 21 (4), 92–101. https://doi.org/10.1037/edu0000332 Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2022). A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM) (3 ed.). Thousand Oaks, CA: Sage. Hair, J. F., Ringle, C. M., & Sarstedt, M. J. L. r. p. (2013). Partial least squares structural equation modeling: Rigorous applications, better results and higher acceptance. 46 (1-2), 1-12. https://ssrn.com/abstract=2233795 Heffernan, N. T., & Heffernan, C. L. (2014). The ASSISTments ecosystem: Building a platform that brings scientists and teachers together for minimally invasive research on human learning and teaching. International Journal of Artificial Intelligence in Education, 24 (4), 470–497. https://doi.org/10.1007/s40593-014-0024-x Henseler, J., & Sarstedt, M. (2013). Common beliefs and reality about partial least squares: comments on Rönkkö and Evermann. Organizational Research Methods , 17 (2). https://doi.org/ : 10.1177/1094428114526928 Hunter, A.-B., Laursen, S. L., & Seymour, E. (2007). Becoming a scientist: The role of undergraduate research in students’ cognitive, personal, and professional development. Science Education, 91 (1), 36–74. https://doi.org/10.1002/sce.20173 Johnson, M. D., et al. (2006). "The evolution of loyalty intentions." Journal of marketing 70 (2): 122-132. Kasneci, E., Seßler, K., Küchemann, S., Bannert, M., Dementieva, D., Fischer, F., et al. (2023). ChatGPT for good? On opportunities and challenges of large language models for education. Learning and Individual Differences, 103 , Article 102274. https://doi.org/10.1016/j.lindif.2023.102274 Liengaard, B. D., Sharma, P. N., Hult, G. T. M., Jensen, M. B., Sarstedt, M., Hair, J. F., & Ringle, C. M. (2021). Prediction: Coveted, Yet Forsaken? Introducing a Cross-validated Predictive Ability Test in Partial Least Squares Path Modeling. Decision Sciences, 52 (2), 362-392. LifeWire. (2021). How AI could fake papers and wreck the scientific process. https://www.lifewire.com/how-ai-could-fake-papers-and-wreck-the-scientific-process-6834621 Lin, M. P.-C., Liu, A. L., Poitras, E., Chang, M., & Chang, D. H. (2024). An Exploratory Study on the Efficacy and Inclusivity of AI Technologies in Diverse Learning Environments. Sustainability , 16 (20), 8992. https ://doi.org/10.3390/su16208992 Lopatto, D. (2010). Undergraduate research as a high-impact student experience. Peer Review, 12 (2), 27–30. Lund, B. D., Wang, T., Mannuru, N. R., Nie, B., Shimray, S., & Wang, Z. (2023). ChatGPT and a new academic reality: Artificial Intelligence-written research papers and the ethics of the large language models in scholarly publishing. Journal of the Association for Information Science and Technology, 74 (5), 570–581. https://doi.org/10.1002/asi.24750 Machi, L. A., & McEvoy, B. T. (2016). The literature review: Six steps to success (3rd ed.). Corwin Press. Mertala, P., Fagerlund, J., & Calderon, O. (2022). Finnish 5th and 6th grade students’ pre-instructional conceptions of artificial intelligence (AI) and their implications for AI literacy education. Computers and Education: Artificial Intelligence, 3 , https://doi.org/10.1016/j.caeai.2022.100095 Ofem, U. J., Iyam, M. A., Ovat, S. V., Nworgwugwu, E. C., Anake, P. M., Udeh, M. I., et al. (2024). Artificial intelligence (AI) in academic research: A multi-group analysis of students’ awareness and perceptions using gender and programme type. Journal of Applied Learning & Teaching, 7 (1), 76–105. https://doi.org/10.37074/jalt.2024.7.1.9 Ofem, U. J., Nworgwugwu, E. C., Ovat, S. V., Anake, P. M., Anyin, N. N., Udeh, M. I. (2024). Predicting affective and cognitive learning outcomes: A quantitative analysis using climate change vectors. Eurasian Journal of Science and Environmental Education, 4 (1), 1–12. https://doi.org/10.30935/ejsee/14405 Popenici, S., Rudolph, J., Tan, S., & Tan, S. (2023). A critical perspective on generative AI and learning futures: An interview with Stefan Popenici. Journal of Applied Learning and Teaching, 6 (2), 311–331. https://doi.org/10.37074/jalt.2023.625 Sekaran, U. and Bougie, R. (2009) Research Methods for Business, West Sussex, Wiley. Sellin, N., & Versand, O. (1995). Partial least square modeling in research on educational achievement . Waxmann. Shadle, S. E., Marker, A., & Earl, B. (2017). Faculty drivers and barriers: Laying the groundwork for undergraduate STEM education reform in academic departments. International Journal of STEM Education, 4 (1), 1–14. https://doi.org/10.1186/s40594-017-0062-7 Sharma, P. N., Liengaard, B. D., Hair, J. F., Sarstedt, M., & Ringle, C. M. (2023). Predictive Model Assessment and Selection in Composite-based Modeling Using PLS-SEM: Extensions and Guidelines for Using CVPAT. European Journal of Marketing, 57 (6), 1662-1677. Shmueli, G., Ray, S., Estrada, J. M. V., & Chatla, S. B. (2016). The Elephant in the Room: Predictive Performance of PLS Models. Journal of Business Research, 69 (10), 4552-4564. Shmueli, G., Sarstedt, M., Hair, J. F., Cheah, J. H., Ting, H., Vaithilingam, S., & Ringle, C. M. (2019). Predictive Model Assessment in PLS-SEM: Guidelines for Using PLSpredict. European Journal of Marketing, 53(11), 2322-2347. Strzelecki, A. (2023). Students’ acceptance of ChatGPT in higher education: An extended unified theory of acceptance and use of technology. Innovative Higher Education . https://doi.org/10.1007/s10755-023-09686-1 Swiecki, Z., Ruis, A. R., Gautam, D., Rus, V., & Williamson Shafer, D. (2019). Understanding when students are active-in-thinking through modeling-in-context. British Journal of Educational Technology, 50 (5), 2346–2364. https://doi.org/10.1111/bjet.12869 Tareq, R., Sumesh, N., Diane, K., Mulyadi, R., Fernando de Oliveira, S., Wagner, J., et al. (2023). The role of ChatGPT in higher education: Benefits, challenges, and future research directions. Journal of Applied Learning & Teaching, 6 , 1. https://doi.org/10.37074/jalt.2023.6.1.29 Vij, S., Tayal, D., & Jain, A. (2020). A machine learning approach for automated evaluation of short answers using text similarity based on WordNet graphs. Wireless Personal Communications, 111 (2), 1271–1282. https://doi.org/10.1007/s11277-019-06913-x Wetzels, M., Odekerken-Schröder, G., & Van Oppen, C. (2009). Using PLS path modeling for assessing hierarchical construct models: Guidelines and empirical illustration. MIS quarterly , 177-195. Wikipedia. (2025). Hallucination of artificial intelligence . In Wikipedia . Retrieved August 25, 2025, https://en.wikipedia.org/wiki/Hallucination Yuan, S., He, T., Huang, H., Hou, R., & Wang, M. (2020). Automated Chinese essay scoring based on deep learning. CMC‑Computers, Materials & Continua, 65 (1), 817–833. https://doi.org/10.32604/cmc.2020.010471 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8566987","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":573330641,"identity":"d6033620-ef04-481b-bffb-cf3358b220a4","order_by":0,"name":"Ahmed Al Rantisi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3UlEQVRIiWNgGAWjYFAC5gYQycPAzHxAAshgbCCsBazGAKiFLYE0LSCLDIjTwt9+sO3Dzx1/ZOTbeT7e5mGwkd1wgPnoBnxaJM4kNs/sPWPAY3CYd7M1D0Oa8YYDbGk38FpzILGZgbcNqIWZd5s0D8PhxA0HeMzwapE//7CZ8S9Qi3wzzzOglv+EtRjcSGxmBtnCcJiHDajlAGEthjceNjPLthkD/cJmbDnHINl45mECfpE7n3yY8W2bnL18/+GHN95U2Mn2HW8+ht/7aO4EYmYS1I+CUTAKRsEowA4A1YNHWyNg3AkAAAAASUVORK5CYII=","orcid":"","institution":"Dhofar University","correspondingAuthor":true,"prefix":"","firstName":"Ahmed","middleName":"Al","lastName":"Rantisi","suffix":""},{"id":573330642,"identity":"41aee7ed-481a-4359-8a98-962090888f30","order_by":1,"name":"Eiman Koofan","email":"","orcid":"","institution":"Dhofar University","correspondingAuthor":false,"prefix":"","firstName":"Eiman","middleName":"","lastName":"Koofan","suffix":""}],"badges":[],"createdAt":"2026-01-10 08:53:41","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8566987/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8566987/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":100116808,"identity":"c64a5736-7418-4f30-9a52-be909d7977b8","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":158288,"visible":true,"origin":"","legend":"","description":"","filename":"ManuscriptAnonymous.docx","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/bd1a4822ce2a88e3a59aa457.docx"},{"id":100116804,"identity":"2b611094-cca1-4483-bd68-7c2bd8701071","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4205,"visible":true,"origin":"","legend":"","description":"","filename":"e694823744dd4f308202395133942ea5.json","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/9a2f524fdc301c3e3db4f471.json"},{"id":100116812,"identity":"e1482d8a-edfe-4083-8664-6bd78d76e798","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":96207,"visible":true,"origin":"","legend":"","description":"","filename":"e694823744dd4f308202395133942ea51enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/d85f03f968dd2e014314633c.xml"},{"id":100116806,"identity":"46a51347-8d3d-49c8-836f-ac587302b85c","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"eps","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":397,"visible":true,"origin":"","legend":"","description":"","filename":"drawingimage1.eps","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/af17065b9cf2f49168782e3e.eps"},{"id":100367113,"identity":"1311659e-b0c2-40f3-b649-0f327281d69b","added_by":"auto","created_at":"2026-01-16 07:56:46","extension":"eps","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":380,"visible":true,"origin":"","legend":"","description":"","filename":"drawingimage2.eps","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/1ca171e7ee6c4613fc2cd116.eps"},{"id":100116809,"identity":"eacdee37-b7c5-4eb1-9ca8-8fd290f2fcbd","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"eps","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":583,"visible":true,"origin":"","legend":"","description":"","filename":"drawingimage3.eps","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/59447b20dbd992d2607255b1.eps"},{"id":100116813,"identity":"51e93cea-a79c-4fe6-a52f-1e613fa30a38","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"eps","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54986,"visible":true,"origin":"","legend":"","description":"","filename":"drawingimage4.eps","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/ad801d1b542ea8083cc48396.eps"},{"id":100116810,"identity":"9189f844-56c5-48bc-a290-b7f6cc48a38f","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27540,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/8162b5e0c8f95ab097b828c9.jpeg"},{"id":100367138,"identity":"a433d111-918a-47a1-922a-3a5894c304c2","added_by":"auto","created_at":"2026-01-16 07:56:47","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":63651,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/2e97bfd3024b66de4d536be0.png"},{"id":100116818,"identity":"cf03c40e-5fee-41d1-983c-1d56a02562f8","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":39323,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/58b8fb2f44c13d4df8bfcb48.jpeg"},{"id":100366552,"identity":"d969935a-fdd6-48df-8cc5-0aa0ebc658ac","added_by":"auto","created_at":"2026-01-16 07:56:20","extension":"jpeg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":24063,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/4e48fb5b0944e9e3c7e3913e.jpeg"},{"id":100366882,"identity":"ca04064d-273e-4aec-9ea5-37ec50bdfb60","added_by":"auto","created_at":"2026-01-16 07:56:36","extension":"jpeg","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":60941,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/01556d478bc3c4b3aff23db6.jpeg"},{"id":100367168,"identity":"be116114-62d6-45ef-8d11-d436829bc203","added_by":"auto","created_at":"2026-01-16 07:56:49","extension":"jpeg","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":24497,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/53983cdd57d1dd6bc501edd5.jpeg"},{"id":100365366,"identity":"c7240288-f61a-45e5-b225-66e76c53170e","added_by":"auto","created_at":"2026-01-16 07:55:07","extension":"jpeg","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5031,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/dddf828d5d1352bd23bf617a.jpeg"},{"id":100365461,"identity":"606ab6b0-3977-476c-a67a-a28b8512eab4","added_by":"auto","created_at":"2026-01-16 07:55:13","extension":"jpeg","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16805,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/c69d79368e3dac02e6e8bc1a.jpeg"},{"id":100116831,"identity":"eb65cf90-b222-4f4b-834d-eb9b5371bd7a","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"jpeg","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35798,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/d2f63dc94ff619c4eab229d1.jpeg"},{"id":100116814,"identity":"4ce17350-3ee5-48ab-9dce-b2498ab0326c","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"jpeg","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1074,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/6f37bcd36e1b4e5e374a4b91.jpeg"},{"id":100116820,"identity":"2eb537f6-63b9-453b-bdaa-72d72a754a02","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11202,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/0633e8b75b756d4b547c6681.png"},{"id":100116822,"identity":"90ec429e-81f7-4939-839e-2cc1cd2cbdd8","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23209,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/a9e02e8d3f2bc5b2bcfef117.png"},{"id":100366946,"identity":"a68b41d5-0dd6-4063-83ba-83b94f489d38","added_by":"auto","created_at":"2026-01-16 07:56:41","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11342,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/df4ee79a1993c0a468ce1a8e.png"},{"id":100365553,"identity":"e7c8ad87-3b55-43d3-b179-f125c2034f51","added_by":"auto","created_at":"2026-01-16 07:55:22","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5312,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/ee0e387f7e4fc6ce4d220269.png"},{"id":100366562,"identity":"816fb0c9-0ef2-48b9-9536-fd6db81da73b","added_by":"auto","created_at":"2026-01-16 07:56:21","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18670,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/9d61461e38c498c03a41ec5d.png"},{"id":100116829,"identity":"387adcbe-e48e-4a56-a2b8-eb4cd037087b","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5079,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/46167252979cc36dee720ef7.png"},{"id":100116828,"identity":"53952d2b-1b1a-4de3-95dc-afd8dd6f5ff3","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1824,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/1f3bb496096ef9218ef39f26.png"},{"id":100365923,"identity":"0d65d468-516d-4d2f-b5c2-6f468b74a486","added_by":"auto","created_at":"2026-01-16 07:55:45","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3809,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/7d50ba8e002d16517f94f608.png"},{"id":100116830,"identity":"6e2485d1-44ba-4edb-9b6c-4ae40a885329","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12989,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/a0daaecaca154ed0e555a9f4.png"},{"id":100366280,"identity":"2c2ba3ba-d141-49aa-8991-239452f4887c","added_by":"auto","created_at":"2026-01-16 07:56:11","extension":"png","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":935,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/f615be6adc58388d9c7401aa.png"},{"id":100116833,"identity":"ce942ed5-8d9b-45e8-bbde-674c43b4a3df","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"xml","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":93526,"visible":true,"origin":"","legend":"","description":"","filename":"e694823744dd4f308202395133942ea51structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/5af7d6cef7a94dfa697ae22e.xml"},{"id":100116834,"identity":"ff8fbdbd-0028-4edd-960f-b8bc7ff42180","added_by":"auto","created_at":"2026-01-13 08:07:53","extension":"html","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":105735,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/17ccce8c91054aacfd24626e.html"},{"id":100365212,"identity":"9458e245-8933-4904-8a33-86f452a67849","added_by":"auto","created_at":"2026-01-16 07:54:51","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":62292,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eshows the interrelationships among variables and other relevant factors.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptual model of the impact of artificial intelligence utilization on research technical skills and research efficacy.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNotes:\u003c/strong\u003e The model illustrates how the use of artificial intelligence influences research efficacy across its five dimensions: background writing efficacy (BWE), literature review efficacy (LRE), methodological efficacy (ME), analytical efficacy (AE), and report writing efficacy (RWE), both directly and indirectly through the mediating role of research technical skills (RTS).\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/dfede66cddbce35e4068cfa5.png"},{"id":100365533,"identity":"a40b795f-655d-496f-b77b-0dfd4754cd70","added_by":"auto","created_at":"2026-01-16 07:55:20","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":407190,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eshows the structural model measurement tool: convergent reliability and validity.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStructural model\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/e0b36081d444144c3a4b0d0a.png"},{"id":101397624,"identity":"f044b039-4c6b-4f02-99d1-dabe01d2a056","added_by":"auto","created_at":"2026-01-29 09:33:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1401293,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8566987/v1/2b38f286-014d-4d84-90e0-8b6f85d54c5c.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"The Impact of Artificial Intelligence Use on Enhancing Research Efficacy among Postgraduate Students: The Mediating Role of Research Technical Skills Using PLS-SEM","fulltext":[{"header":"Introduction","content":"\u003cp\u003eEnvironmental concerns have increasingly influenced research efficacy and academic behavior in recent years, particularly with the integration of Artificial Intelligence (AI) tools in research processes) Al-Rantisi et al., 2025(. AI has been shown to enhance various stages of scientific inquiry, such as literature reviews, experimental design, and manuscript preparation, thereby boosting the overall efficiency of research workflows (Lin et al., 2024; Bola\u0026ntilde;os et al., 2024). For instance, generative AI models can rapidly summarize literature and suggest relevant sources, significantly reducing the time and cognitive load required for academic writing, especially among non-native English speakers (LifeWire, 2021). Research efficacy is vital in the realm of research engagement because conducting research is a challenging and complex task that must follow a structured pattern to achieve reproducible results. Many scholars have noted that research efficacy is fundamental to the development of research skills. Students with high levels of research efficacy are more likely to engage in research activities, persist through challenges, and achieve higher competence in their research endeavours(Bandura, 1997).\u003c/p\u003e \u003cp\u003eHowever, alongside these benefits, several studies raise concerns regarding potential over-reliance on AI. Excessive dependence on AI for writing and idea generation may impair critical thinking, diminish analytical skills, and lead to the degradation of core research competencies (Enciso 2025). Additionally, the phenomenon of AI hallucinations, where models produce fabricated or erroneous references or content, poses serious risks to research integrity and demands cautious oversight (Wikipedia, 2025). Amidst this dual-edged influence of AI, the development of researchers\u0026rsquo; technical skills serves as a pivotal intermediary. Effective utilization of AI tools requires not only access to technology but also proficiency in navigating digital platforms, critically evaluating AI-generated outputs, and ethically integrating AI into research workflows. This technical skillset builds a bridge between AI adoption and the preservation of research quality and efficacy.\u003c/p\u003e \u003cp\u003eRecent studies have explored the impact of AI on research skills and productivity among students in higher education. AI tools can enhance writing skills, critical thinking, and analysis (Arbab et al., 2024). However, their use raises concerns about academic integrity and potential loss of cognitive skills (Buniel et al., 2025). A training program on generative AI applications showed significant positive effects on lecturers' teaching capabilities in Oman (Al-Saiari et al., 2024). For STEM undergraduate students, AI dependence indirectly influences research productivity through the mediation of research skills, dispositions, and self-efficacy (Buniel et al., 2025). While AI offers potential benefits, students express uncertainties about its use and recognize the need for specialized skills and responsible application (Aguirre-Aguilar et al., 2024). Institutions are encouraged to develop standardized policies and regulations on AI usage in academic settings (Arbab et al., 2024).\u003c/p\u003e \u003cp\u003eBuilding on these insights, the present study explores the relationship between AI use, research technical skills, and research efficacy, with a focus on postgraduate students in Oman. Employing a PLS-SEM approach, the study examines both the direct impact of AI utilization on research efficacy and the mediating role of research technical skills in this relationship.\u003c/p\u003e"},{"header":"Literature review","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudies on the utilization of artificial intelligence in research\u003c/h2\u003e \u003cp\u003eThe integration of technology into research has continuously evolved through ongoing innovations. In recent years, AI which is designed to perform cognitive and problem-solving tasks, has received growing attention (Bonk \u0026amp; Wiley, 2020). Research indicates that AI contributes to a variety of functions within the research process (Ofem, Iyam, et al., 2024; Ofem, Nworgwugwu, et al., 2024). For example, AI applications such as ChatGPT can operate as literature review tools when given suitable prompts (Clark, 2020), assist in monitoring research progress (Swiecki, Ruis, Gautam, Rus, \u0026amp; Williamson Shafer, 2019), support personalized learning experiences (Chiu et al., 2022; Mertala, Fagerlund, \u0026amp; Calderon, 2022), automate the process of data collection (Hefernan \u0026amp; Hefernan, 2014), and conduct data analysis (Vij, Tayal, \u0026amp; Jain, 2020; Yuan, He, Huang, Hou, \u0026amp; Wang, 2020). AI also enables profiling of respondents\u0026rsquo; backgrounds, which is essential in identifying disparities among participants (Cohen et al., 2017). In addition, AI tools facilitate academic writing by assisting in editing, generating text, providing language translation, and answering scholarly inquiries (Dwivedi et al., 2023; Kasneci et al., 2023; Lund et al., 2023).\u003c/p\u003e \u003cp\u003ePrevious research has highlighted the influence of artificial intelligence (AI) on academic research (Adarkwah et al., 2023; Crawford, Cowling, \u0026amp; Allen, 2023; Popenici, Rudolph, Tan, \u0026amp; Tan, 2023). For example, Ofem, Iyam, et al. (2024) reported that students frequently utilize AI in their research activities, with male students showing higher engagement compared to female students. Likewise, Tareq et al. (2023) found that AI adoption in higher education supports students in multiple research-related tasks, although concerns regarding reliability and gaps in skills were also identified. Moreover, Gasaymeh (2018) highlighted that students who have access to information and communication technologies are more capable of utilizing AI in their research compared to those without such access. Similarly, Devadas and Shilpa (2023) found that students\u0026rsquo; willingness to adopt ChatGPT depends on multiple influencing factors. In addition, Strzelecki (2023) noted that although students are generally receptive to using ChatGPT in higher education, this acceptance is shaped by elements such as their technological skills and expectations regarding performance. In the Sultanate of Oman, the integration of artificial intelligence (AI) in academic research is still at an early stage, and many postgraduate students have yet to fully recognize its potential in enhancing research outcomes. This study, entitled The Impact of Artificial Intelligence Use on Enhancing Research Efficacy among Postgraduate Students in the Sultanate of Oman: The Mediating Role of Research Technical Skills Using PLS-SEM, was motivated by the need to provide empirical evidence that can inform policy decisions and encourage the adoption of AI tools, ultimately aiming to improve students\u0026rsquo; research efficacy through the development of technical research skills.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eStudies on research efficacy\u003c/h3\u003e\n\u003cp\u003eResearch efficacy is grounded in Bandura\u0026rsquo;s (1997) social cognitive theory, which emphasizes research technical skills as a determinant of individuals\u0026rsquo; ability to complete challenging tasks successfully. Students with higher levels of research technical skills are more likely to persist, demonstrate resilience, and achieve better performance outcomes. Applied to research, skills encompass cognitive, affective, and psychomotor dimensions, reflected in tasks such as writing research backgrounds, conducting literature reviews, designing methodologies, analyzing data, and producing reports.\u003c/p\u003e \u003cp\u003eEfficacy in literature review, for instance, is demonstrated through the ability to critically synthesize studies and identify gaps that guide new inquiries (Machi \u0026amp; McEvoy, 2016). Methodological efficacy involves designing robust and ethical research plans (Creswell \u0026amp; Creswell, 2018), while data analysis efficacy requires competence in statistical tools and interpretation skills to ensure validity. Report writing efficacy, in turn, allows researchers to communicate findings effectively to both academic and non-academic audiences (Field, 2018).\u003c/p\u003e \u003cp\u003ePrior research has shown that research opportunities and structured mentoring programs enhance students\u0026rsquo; confidence, critical thinking, and problem-solving abilities (Hunter et al., 2007; Lopatto, 2010). Supportive environments that provide guidance, feedback, and mentoring have been found to strengthen research efficacy and foster persistence in academic pursuits (Shadle, Marker, \u0026amp; Earl, 2017; Chemers et al., 2011). However, much of the literature still treats research efficacy as a one-dimensional construct and pays limited attention to how technological advancements, particularly artificial intelligence, may influence its development. Addressing this gap is critical to informing policies that aim to improve students\u0026rsquo; research capacity and strengthen innovation in higher education.\u003c/p\u003e\n\u003ch3\u003eConceptual framework\u003c/h3\u003e\n\u003cp\u003eThis study examines the impact of artificial intelligence (AI) utilization on enhancing research efficacy among postgraduate students in the Sultanate of Oman, with a particular focus on the mediating role of research technical skills (RTS). Research efficacy is conceptualized as a multidimensional construct comprising five components: background writing efficacy (BWE), literature review efficacy (LRE), methodological efficacy (ME), analytical efficacy (AE), and report writing efficacy (RWE). Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents the interrelationships among these variables and other relevant factors.\u003c/p\u003e \u003cp\u003eBased on the study\u0026rsquo;s objectives and existing literature, the following main hypotheses were formulated:\u003c/p\u003e \u003cp\u003eH01: The use of artificial intelligence has no statistically significant influence on research technical skills among postgraduate students.\u003c/p\u003e \u003cp\u003eH02: Research technical skills have no statistically significant influence on background writing efficacy among postgraduate students.\u003c/p\u003e \u003cp\u003eH03: Research technical skills have no statistically significant influence on literature review efficacy among postgraduate students.\u003c/p\u003e \u003cp\u003eH04: Research technical skills have no statistically significant influence on methodological efficacy among postgraduate students.\u003c/p\u003e \u003cp\u003eH05: Research technical skills have no statistically significant influence on analytical efficacy among postgraduate students.\u003c/p\u003e \u003cp\u003eH06: Research technical skills have no statistically significant influence on reporting writing efficacy among postgraduate students.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eH1\u003c/h3\u003e\n\u003cdiv class=\"Heading\"\u003eH1\u003c/div\u003e \u003cp\u003e \u003cem\u003eH05\u003c/em\u003e\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eNotes\u003c/b\u003e: The model illustrates how the use of artificial intelligence influences research efficacy across its five dimensions: background writing efficacy (BWE), literature review efficacy (LRE), methodological efficacy (ME), analytical efficacy (AE), and report writing efficacy (RWE), both directly and indirectly through the mediating role of research technical skills (RTS).\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eThe study adopts a positivist research paradigm, relying primarily on quantitative methods. A cross-sectional survey design was employed to collect data from a targeted population at a specific point in time (Ofem et al., 2024). The sample was purposively selected from students enrolled in the Department of Social Sciences and the Department of Education at the College of Arts and Applied Sciences, Dhofar University, Sultanate of Oman. These departments were chosen because they host the largest number of postgraduate students in the college and their curricula ensure that students are generally well-informed about modern technologies, including artificial intelligence tools. A total of 30 students participated in the study, reflecting the proportional representation of these two departments and their relevance to the research context.\u003c/p\u003e \u003cp\u003eEthical considerations were strictly observed throughout the study. Participation was voluntary, and all respondents provided informed consent before taking part in the survey. Respondents were assured of the confidentiality and anonymity of their responses, and they were informed of their right to withdraw from the study at any time without any consequences. The authors obtained formal approval, IRB#DU/AY/2025-26/QUES-002, from the Research Department at Dhofar University to conduct the survey among the students. All data were collected during the Fall semester of 2025\u0026ndash;2026. The study adhered to ethical guidelines in accordance with the principles outlined in the Helsinki Declaration and other relevant research ethics standards.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eMeasures/instruments\u003c/h2\u003e \u003cp\u003eThe present study incorporated seven main variables: the use of artificial intelligence (AI Use), research technical skills (RTS) as a mediating variable, background writing efficacy (BWE), literature review efficacy (LRE), methodological efficacy (ME), analytical efficacy (AE), and research writing efficacy (RWE) as response variables. The research instrument was divided into two sections. Section A collected demographic information, including gender, age, academic discipline, academic stage, prior training in artificial intelligence, and frequency of using AI tools in research. This section also included spaces for written informed consent, as well as email addresses and phone numbers for online submission via Google. Forms.\u003c/p\u003e \u003cp\u003eSection B contained items measuring the explanatory, mediating, and response variables. The explanatory variable, AI utilization in research, was assessed using a four-item scale adapted from Ofem, Iyam, et al. (2024), which demonstrated robust psychometric properties and was recently used in a similar context. This construct included items reflecting students\u0026rsquo; use of AI in research-related tasks, such as supporting background writing, finding appropriate references, rephrasing or refining research texts, and assisting with data organization and analysis. Sample item: \u003cem\u003e\u0026ldquo;I use AI tools to support the writing of the theoretical background of my research\u003c/em\u003e\u0026rdquo;. All items were rated on a four-point linear scale.\u003c/p\u003e \u003cp\u003eThe mediating variable, research technical skills (RTS), was measured with four items capturing students\u0026rsquo; competence in applying digital and technological tools to different stages of research. The items reflected their ability to utilize online research tools effectively, assess the reliability of AI-generated outputs, employ statistical or qualitative analysis software, and integrate AI ethically and responsibly into academic work. Sample item: \u003cem\u003e\u0026ldquo;I can evaluate the reliability of AI-generated outputs for research purposes\u0026rdquo;.\u003c/em\u003e\u003c/p\u003e \u003cp\u003eThe response variable, research efficacy, was conceptualized as the confidence and belief in one\u0026rsquo;s ability to successfully conduct research activities, and was examined across five dimensions, each measured with four items. Background Writing Efficacy (BWE) refers to the ability to develop a coherent and well-contextualized research background that establishes the significance of the topic. Sample item: \u003cem\u003e\u0026ldquo;I can write a coherent research background that highlights the importance of my topic\u0026rdquo;.\u003c/em\u003e Literature Review Efficacy (LRE) reflected the skills needed to critically evaluate and synthesize relevant academic literature, as well as to identify research gaps. Sample item: \u003cem\u003e\u0026ldquo;I can identify research gaps accurately through reviewing the literature\u0026rdquo;.\u003c/em\u003e Methodological Efficacy (ME) captured students\u0026rsquo; competence in selecting research designs, constructing tools, and applying correct procedures. Sample item: \u003cem\u003e\u0026ldquo;I can design appropriate research instruments such as questionnaires or interviews\u0026rdquo;.\u003c/em\u003e Analytical Efficacy (AE) measured the ability to analyze and interpret data accurately, draw logical conclusions, and compare findings with prior studies. Sample item: \u003cem\u003e\u0026ldquo;I can analyze research data correctly using the appropriate tools\u0026rdquo;.\u003c/em\u003e Finally, Report Writing Efficacy (RWE) assessed the extent to which students could communicate research findings clearly, structure reports logically, and present results in line with academic standards. Sample item: \u003cem\u003e\u0026ldquo;I can write a comprehensive research report in line with academic standards\u0026rdquo;.\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eContent validity\u003c/h3\u003e\n\u003cp\u003eThe scale items underwent quantitative validation by five experts from different fields, who evaluated the relevance, clarity, and suitability of the items for measuring the study\u0026rsquo;s constructs. Initially, 30 items were developed based on an extensive review of the literature and adaptations from previous research. Following the initial screening process, two items were excluded. The experts were provided with the questionnaire and an assessment rubric, following the guidelines recommended by specialists (Ofem, Iyam, et al., 2024; Ofem, Nworgwugwu, et al., 2024).\u003c/p\u003e \u003cp\u003eEach item was rated on a five-point Likert scale, where 1 indicated \u0026ldquo;strongly disagree\u0026rdquo; and 5 indicated \u0026ldquo;strongly agree,\u0026rdquo; reflecting the degree to which each item appropriately captured the intended construct. The Item-Content Validity Index (I-CVI) was calculated to determine which items should be retained. I-CVI values range from 0 to 1, with a threshold of 0.78 or higher generally considered acceptable, indicating strong agreement among the expert panel regarding the relevance and clarity of the items. Some researchers, however, consider values as low as 0.70 acceptable under certain conditions, such as smaller panel sizes or strict item development criteria (Haynes, Richard, \u0026amp; Kubany, 1995).\u003c/p\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eData Analysis\u003c/h2\u003e \u003cp\u003eA purposeful sample of 30 postgraduate students from the departments of Social Sciences, Education, Mathematics, and Science in the College of Arts and Applied Sciences at Dhofar University, which are the departments that have postgraduate programs. The data were analyzed using SPSS for descriptive statistics and SmartPLS for structural equation modeling (PLS-SEM). The descriptive analysis provided an overview of the participants\u0026rsquo; demographic characteristics and responses to the study variables.\u003c/p\u003e \u003cp\u003eTo test the hypothesized relationships among AI utilization, research technical skills, and research efficacy, the partial least squares structural equation modeling (PLS-SEM) approach was employed. PLS-SEM is a widely used method in social science research for examining theoretically validated additive and linear causal models (Hair et al., 2013). It allows for the simultaneous estimation of both the measurement model, which assesses the validity and reliability of the constructs, and the structural model, which tests the hypothesized relationships among variables. SmartPLS software was used to perform all analyses (Ringle et al., 2015). The evaluation of the measurement model ensured the reliability and validity of the constructs, while the structural model assessment focused on the statistical significance of the hypothesized paths, following the guidelines proposed by Chin (2009) and Hair et al. (2013). This approach guarantees that the study\u0026rsquo;s findings are both statistically robust and theoretically grounded.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eDescriptive statistics\u003c/h2\u003e \u003cp\u003eThe demographic characteristics of the respondents are summarized as follows: regarding gender, [8] (26.7%) were male and [22] (73.3%) were female; concerning age, [6] (20.0%) were below 25 years, [11] (36.7) were between 25 and 30 years, [8] (26.7%) were between 31 and 35 years and [5] (16.7%) were above 35 years; in terms of academic disciplines, [14] (46.7%) were from Social Sciences Department, [16] (53.3%) from Education Department; regarding academic stage, [22] (73.3%) were in the coursework stage and [8] (26.7%) were in the thesis preparation stage; concerning previous training on artificial intelligence, [18] (60.0%) had received training and [12] (%40.0) had not received any training; in terms of frequency of using artificial intelligence tools in research, [9] (30%) used them occasionally, [15] (50%) used them frequently, and [6] (20%) used them always, every participant provided informed consent.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eMeasurement model evaluation\u003c/h2\u003e \u003cp\u003eTo validate the measurement model, the following steps were taken: (1) calculating the factorial load for reflective constructs (AI, RTS, BWE, LRE, ME, AE, RWE) to determine the item's individual reliability; and (2) determining the construct's validity.\u003c/p\u003e \u003cp\u003eTo assess whether the questionnaire was sufficiently reliable, many experiments were run. The purpose of the first test was to decide whether to accept a certain item as a component of a reflective construct. The items needed to have a factorial load (\u0026#120524;) or simple correlations that were at least 0.707 in order to serve this function (Carmines \u0026amp; Zeller, 1979). Due to weak loads (\u0026le;\u0026thinsp;0.7), three items (AI-03, RTS-04, and ME-01) were removed. The remaining items, however, were higher than 0.7. Due to the remaining items' contribution to content validity (AVE\u0026thinsp;=\u0026thinsp;0.5; \u003cem\u003eP\u003c/em\u003e\u003csub\u003e\u003cem\u003eC\u003c/em\u003e\u003c/sub\u003e = 0.7), they were kept in place for this investigation. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e displays the outcomes.\u003c/p\u003e \u003cp\u003eThe Composite Reliability Index (\u003cem\u003eP\u003c/em\u003e\u003csub\u003e\u003cem\u003eC\u003c/em\u003e\u003c/sub\u003e) and Cronbach's alpha (α) were used in the second test to evaluate the construct's internal consistency. Using the suggestion of Hair et al. (2013), suggesting 0.7 as a reference point for both α and \u003cem\u003eP\u003c/em\u003e\u003csub\u003e\u003cem\u003eC\u003c/em\u003e\u003c/sub\u003e. The results of Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e show that all constructs were dependable and had satisfactory internal consistency.\u003c/p\u003e \u003cp\u003eA PLS instrument that was generated using the AVE was shown to be reliable thanks to the convergent validity test. The AVE coefficient for reflecting structures was higher than 0.5, as shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, which summarizes the results. According to these suggestions, all AVE measures were valid.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eStructural model measurement tool: convergent reliability and validity.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstruct\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeasurement item\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLoading\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eCR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAVE\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eAI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAI-01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.804\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003e0.926\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003e0.949\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003e0.822\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAI-02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.794\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAI-03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eItem deleted\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAI-04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.823\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eRTS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRTS-01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.702\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003e0.773\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003e0.847\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003e0.584\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRTS-02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.766\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRTS-03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.905\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRTS-04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eItem deleted\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eBWE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBWE-01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.809\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003e0.834\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003e0.884\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003e0.656\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBWE-02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.796\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBWE03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.877\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBWE-04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.753\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLRE-01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.816\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLRE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLRE-02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.847\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.851\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.898\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.688\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLRE-03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.797\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLRE-04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.856\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eME\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eME-01\u003c/p\u003e \u003cp\u003eME-02\u003c/p\u003e \u003cp\u003eME-03\u003c/p\u003e \u003cp\u003eME-04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eItem deleted\u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e \u003cp\u003e0.724\u003c/p\u003e \u003cp\u003e0.837\u003c/p\u003e \u003cp\u003e0.789\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.737\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.831\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.555\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAE-01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.812\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.772\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.842\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.575\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAE-02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.873\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAE-03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.957\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAE-04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.975\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRWE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRWE-01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.928\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.937\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.954\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.838\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRWE-02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.935\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRWE-03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.914\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRWE-04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.884\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eAI-01: I use artificial intelligence tools to support the writing of the theoretical background of my research; AI-02: Artificial intelligence applications help me find suitable references and scientific sources; AI-03: I benefit from artificial intelligence in drafting or rephrasing research texts; AI-04: I use artificial intelligence to analyze or organize data effectively; RTS-01: I am able to use electronic research tools efficiently; RTS-02: I am capable of evaluating the reliability of the outputs generated by artificial intelligence for scientific research purposes; RTS-03: I possess skills in using specialized software for statistical or qualitative analysis; RTS-04: I can ethically and properly integrate digital tools and artificial intelligence into research; BWE-01: I can write a coherent research background that highlights the importance of my topic; BWE-02: I am able to formulate the research problem clearly based on the literature; BWE-03:I can summarize the main points of previous studies in a logical and coherent manner; BWE-04: I am able to present the research background in a way that makes it easy for readers and other researchers to understand; LRE-01: I am able to summarize previous studies and present them critically; LRE-02: I can accurately identify research gaps through a review of the literature; LRE-03: I am able to link the literature to the research problem in a systematic way; LRE-04: I can evaluate the reliability of the academic sources used in the research; ME-01: I am able to choose the appropriate research methodology for my topic; ME-02: I have the ability to design suitable research instruments (questionnaire, interview, etc.); ME-03: I can identify the appropriate statistical or qualitative sample for the study; ME-04: I am able to apply research procedures correctly and systematically; AE-01: I am able to analyze research data correctly using appropriate tools; AE-02: I can interpret research findings in a logical and objective manner; AE-03: I am able to critically compare the results with previous studies; AE-04: I can identify reliable conclusions based on the analyzed data; RWE-01: I am able to write a comprehensive research report in accordance with scientific research standards; RWE-02: I have the ability to present results and discussions in a clear academic manner; RWE-03: I am able to formulate recommendations and conclusions in a clear and results-based way; RWE-04: I can organize the research report logically and sequentially, making it easy for readers to follow; CA: Cronbach\u0026rsquo;s alpha; CR, composite reliability; AVE: variance extracted.\u003c/p\u003e \u003cp\u003e \u003csup\u003ea\u003c/sup\u003eAI-03, \u003csup\u003eb\u003c/sup\u003eRTS-04, \u003csup\u003ec\u003c/sup\u003eME-01, were deleted due to low factorial loadings.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eMeasuring instrument: discriminant validity.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBWE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLRE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eME\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eRTS\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eRWE\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.907\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.611\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.812\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBWE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.410\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.343\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.810\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLRE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.0734\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.478\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.713\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.829\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eME\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.610\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.230\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.587\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.583\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.795\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRTS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.822\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.563\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.564\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.723\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.693\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.796\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRWE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.533\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.303\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.743\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.649\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.713\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.667\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.916\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eDiagonal: square root of the average variance extracted (AVE).\u003c/p\u003e \u003cp\u003eFinally, it is necessary to assess the discriminant validity. For this purpose, the (Fornell \u0026amp; Larcker, 1981) technique, which is founded on the notion that a construct should share more variance with its items than with other constructs in each model, was utilized. Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e demonstrates that the square root of the AVE was higher than the correlation between the variables in this regard. This indicates that all constructions had a stronger relationship with their own objects than they did with those of other constructs.\u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003e: \u003cb\u003eshows the structural model measurement tool: convergent reliability and validity.\u003c/b\u003e\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eStructural model evaluation\u003c/h2\u003e \u003cp\u003eOnce the reliability and validity of the constructs had been verified, an analysis of the relationship between the constructs and the predictive capacity of the endogenous variables was carried out. Therefore, we assessed the weight and nature of the relationships (hypothesis) between the different variables. This assessment involved the use of two basic indicators: the explained variance (\u003cem\u003eR\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e), which indicates the predictive power of the model, and the standardized path coefficients (\u003cem\u003eβ\u003c/em\u003e), which indicate the strength of the relationships between dependent and independent variables (Johnson, Herrmann et al. 2006)\u003c/p\u003e \u003cp\u003eRegarding the predictive capacity of the model, the explained variance (\u003cem\u003eR\u0026sup2;\u003c/em\u003e) of the dependent variables should be equal to or greater than 0.10, since lower values provide little information (Falk \u0026amp; Miller, 1992). According to Chin (1998), R\u0026sup2; values of 0.67 or higher are considered substantial, 0.33 to 0.66 are moderate, and 0.19 to 0.32 are weak. The results of this study show that the model demonstrates moderate predictive power for Analytical Efficacy (AE, \u003cem\u003eR\u0026sup2;\u003c/em\u003e = 0.642), Literature Review Efficacy (LRE, \u003cem\u003eR\u0026sup2;\u003c/em\u003e = 0.473), Methodological Efficacy (ME, \u003cem\u003eR\u0026sup2;\u003c/em\u003e = 0.475), and Reporting Writing Efficacy (RWE, \u003cem\u003eR\u0026sup2;\u003c/em\u003e = 0.401). In contrast, Background Writing Efficacy (BWE, \u003cem\u003eR\u0026sup2;\u003c/em\u003e = 0.262) and Research Technical Skills (RTS, \u003cem\u003eR\u0026sup2;\u003c/em\u003e = 0.263) showed weak predictive power, although still above the recommended threshold of 0.10.\u003c/p\u003e \u003cp\u003eThe cross-validated predictive ability test (CVPAT) has been introduced as an alternative to PLSpredict for assessing the predictive performance of PLS-SEM models. Initially developed by Liengaard et al. (2021) and later extended by Sharma et al. (2023), CVPAT employs an out-of-sample prediction approach that compares the model\u0026rsquo;s average prediction loss with two benchmarks: indicator averages (IA) and a linear model (LM). Superior predictive capability is demonstrated when the model\u0026rsquo;s average loss is significantly lower than these benchmarks. In SmartPLS, CVPAT results are integrated into the PLSpredict report, ensuring comparability of results across approaches (Shmueli et al., 2016, 2019; Hair et al., 2022).\u003c/p\u003e \u003cp\u003eIn the present study, the \u003cem\u003eQ\u0026sup2;\u003c/em\u003e_predict values obtained through PLSpredict and CVPAT confirm that all dependent variables show some degree of predictive relevance. Specifically, Analytical Efficacy (AE; \u003cem\u003eQ\u0026sup2;\u003c/em\u003e = 0.146), Research Technical Skills (RTS; \u003cem\u003eQ\u0026sup2;\u003c/em\u003e = 0.132), and Literature Review Efficacy (LRE; \u003cem\u003eQ\u0026sup2;\u003c/em\u003e = 0.103) demonstrated the strongest predictive validity, whereas Background Writing Efficacy (BWE; \u003cem\u003eQ\u0026sup2;\u003c/em\u003e = 0.051), Methodological Efficacy (ME; \u003cem\u003eQ\u0026sup2;\u003c/em\u003e = 0.020), and Reporting Writing Efficacy (RWE; \u003cem\u003eQ\u0026sup2;\u003c/em\u003e = 0.012) exhibited weaker yet positive predictive capacity (see Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSummary of variance explained (\u003cem\u003eR\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e). PLSpredict \u0026amp; CVPAT:(\u003cem\u003eQ\u0026sup2;predict\u003c/em\u003e).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstruct\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eQ\u0026sup2;predict\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.642\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.146\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBWE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.262\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.051\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLRE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.473\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.103\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eME\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.475\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.020\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRTS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.263\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.132\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRWE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.401\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.012\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003ePredictive Ability of the Model (PLSpredict and CVPAT):\u003c/h2\u003e \u003cp\u003eThe predictive ability of the model was assessed out-of-sample using PLSpredict and CVPAT for all endogenous variables in the study. The results indicate that all variables demonstrated at least some predictive power. For example, Analytical Efficacy (AE), Literature Review Efficacy (LRE), and Research Technical Skills (RTS) exhibited moderate predictive power, with positive \u003cem\u003eQ\u0026sup2;\u003c/em\u003e_predict values of 0.146, 0.103, and 0.132, respectively, indicating that the model can reasonably predict future values for these variables.\u003c/p\u003e \u003cp\u003eOther variables, including Background Writing Efficacy (BWE), Methodological Efficacy (ME), and Reporting Writing Efficacy (RWE), showed weak but positive predictive power, with \u003cem\u003eQ\u0026sup2;\u003c/em\u003e_predict values of 0.051, 0.020, and 0.012, respectively, indicating a minimal yet acceptable level of predictiveness out-of-sample.\u003c/p\u003e \u003cp\u003eRegarding error indicators, the RMSE and MAE values are consistent with the \u003cem\u003eQ\u0026sup2;\u003c/em\u003e_predict results. Lower RMSE and MAE values for variables with higher predictive power (e.g., RTS and LRE) and slightly higher values for variables with weaker predictive power (e.g., ME and RWE) further support the validity of the model\u0026rsquo;s predictive assessment.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eStructural model results.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypothesis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSuggested\u003c/p\u003e \u003cp\u003eeffect\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePath\u003c/p\u003e \u003cp\u003ecoefficients\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003et-value\u003c/p\u003e \u003cp\u003e(bootstrap)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSupport\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH01: \u0026rlm;+AI \u0026rarr; + RTS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.563***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.700\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSupport H1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH02: \u0026rlm;+RTS \u0026rarr; + BWE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.822***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e15.948\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSupport H1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH03: \u0026rlm;+RTS \u0026rarr; + LRE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.564***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.241\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSupport H1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH04: \u0026rlm;+RTS \u0026rarr; + ME\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.723***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e13.479\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSupport H1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH05: \u0026rlm;+RTS \u0026rarr; + AE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.693***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9.715\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSupport H1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH06: \u0026rlm;+RTS \u0026rarr; + RWE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.667***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.885\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSupport H1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn line with the formulated null hypotheses, the study tested whether Artificial Intelligence (AI) and Research Technical Skills (RTS) exerted no statistically significant influence on the dependent constructs. However, the empirical findings revealed otherwise. The relationship between AI and RTS among postgraduate students was found to be positive and significant (\u003cem\u003eβ\u003c/em\u003eAI \u0026rarr; RTS\u0026thinsp;=\u0026thinsp;0.563; t\u0026thinsp;=\u0026thinsp;3.700; p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), thereby rejecting H01. Similarly, RTS demonstrated a strong and significant positive effect on Background Writing Efficacy (BWE) (\u003cem\u003eβ\u003c/em\u003eRTS \u0026rarr; BWE\u0026thinsp;=\u0026thinsp;0.822; t\u0026thinsp;=\u0026thinsp;15.948; p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), leading to the rejection of H02. In the same vein, RTS significantly predicted Literature Review Efficacy (LRE) (\u003cem\u003eβ\u003c/em\u003eRTS \u0026rarr; LRE\u0026thinsp;=\u0026thinsp;0.564; t\u0026thinsp;=\u0026thinsp;6.241; p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), resulting in the rejection of H03. Furthermore, RTS had a substantial positive influence on Methodological Efficacy (ME) (\u003cem\u003eβ\u003c/em\u003eRTS \u0026rarr; ME\u0026thinsp;=\u0026thinsp;0.723; t\u0026thinsp;=\u0026thinsp;13.479; p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), which rejects H04. The analysis also confirmed a significant positive relationship between RTS and Analytical Efficacy (AE) (\u003cem\u003eβ\u003c/em\u003eRTS \u0026rarr; AE\u0026thinsp;=\u0026thinsp;0.693; t\u0026thinsp;=\u0026thinsp;9.715; p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), rejecting H05. Finally, RTS was shown to significantly enhance Reporting Writing Efficacy (RWE) (\u003cem\u003eβ\u003c/em\u003eRTS \u0026rarr; RWE\u0026thinsp;=\u0026thinsp;0.667; t\u0026thinsp;=\u0026thinsp;6.885; p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), thereby rejecting H06.\u003c/p\u003e \u003cp\u003eTo measure the predictive power of model-dependent constructs, the Stone\u0026ndash;Geisser procedure or \u003cem\u003eQ\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e parameter (cross-validated redundancy) was used. This test was run using the blindfolding technique. Parameter \u003cem\u003eQ\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e must be greater than zero to have predictive validity (Chin, 1998) since values above zero determine that the predictability of the model is relevant (Sellin \u0026amp; Versand, 1995). As Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e shows, \u003cem\u003eQ\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e value met this condition. Therefore, the predictive relevance of the model in relation to endogenous latent variables was supported.\u003c/p\u003e \u003cp\u003eFinally, we calculated the Standardized Root Mean Square Residual (SRMR), which is the average difference between the predicted and observed correlations (variances and covariances) based on the standard error of the residual (Henseler \u0026amp; Sarstedt, 2013). To ensure the fit of the structural model, we calculated the GOF index \"match quality index\". It is a general indicator defined as follows: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:GOF=\\sqrt{\\left(\\stackrel{-}{{R}^{2}}*\\stackrel{-}{AVE}\\right)}\\)\u003c/span\u003e\u003c/span\u003e The value of \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\stackrel{-}{{R}^{2}}=0.878\\:\\:\\:،\\:\\stackrel{-}{AVE}\\:=0.629\\)\u003c/span\u003e\u003c/span\u003e. Therefore, the GOF value is equal to 0.682, which is higher than 0.36 according to Wetzels et al. (2009) which indicates the fit of the proposed structural model.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe present study examined the impact of Artificial Intelligence (AI) on enhancing research efficacy among postgraduate students in the Sultanate of Oman, with research technical skills (RTS) serving as a mediating factor. The results revealed a significant positive relationship between AI use and RTS, indicating that students who engage more with AI tools develop stronger technical research capabilities. This finding aligns with prior studies emphasizing the growing role of AI in research processes, including literature review, data collection, monitoring research progress, and academic writing (Bonk \u0026amp; Wiley, 2020; Clark, 2020; Swiecki et al., 2019; Dwivedi et al., 2023; Kasneci et al., 2023; Lund et al., 2023). The results also underscore that AI facilitates the development of cognitive and problem-solving skills essential for conducting high-quality research. Furthermore, the study found that RTS had a significant positive effect on all dimensions of research efficacy, including Background Writing Efficacy (BWE), Literature Review Efficacy (LRE), Methodological Efficacy (ME), Analytical Efficacy (AE), and Reporting Writing Efficacy (RWE). This is shown in the results in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003e. These findings resonate with Bandura\u0026rsquo;s (1997) social cognitive theory, which posits that technical skills and self-efficacy are crucial determinants of individuals\u0026rsquo; ability to complete challenging tasks. The results also align with prior research highlig,hting that structured research opportunities, guidance, and mentoring enhance students\u0026rsquo; confidence, critical thinking, and problem-solving abilities (Hunter et al., 2007; Lopatto, 2010; Shadle et al., 2017; Chemers et al., 2011). In relation to specific research tasks, RTS significantly predicted students\u0026rsquo; ability to synthesize literature, design methodological frameworks, analyze data, and produce reports. This supports the findings of Machi and McEvoy (2016) regarding literature review efficacy, Creswell and Creswell (2018) regarding methodological planning, and Field (2018) regarding reporting skills. The current results extend these findings by demonstrating that AI use amplifies the benefits of technical skills, enabling students to perform these tasks more effectively.\u003c/p\u003e \u003cp\u003eThe study further highlights the moderating role of technological readiness and access. Previous research indicated that students with greater access to ICT infrastructure are better able to leverage AI in their research activities (Gasaymeh, 2018; Ofem, Iyam, et al., 2024). Similarly, Tareq et al. (2023) and Strzelecki (2023) emphasized that students\u0026rsquo; engagement with AI is influenced by their prior skills, expectations, and confidence in using these tools. The present study confirms that in the Omani context, where AI adoption is still nascent, students\u0026rsquo; RTS mediates the effective integration of AI, suggesting that developing technical skills should be a priority in postgraduate programs. Moreover, AI was shown to support multiple facets of the research process, from monitoring progress and automating data collection to enabling personalized learning experiences (Chiu et al., 2022; Mertala et al., 2022; Hefernan \u0026amp; Hefernan, 2014; Vij et al., 2020; Yuan et al., 2020). These functionalities contribute to enhancing research efficacy by providing timely feedback, reducing errors, and increasing productivity, which is consistent with prior literature emphasizing the cognitive and practical benefits of AI in research (Dwivedi et al., 2023; Kasneci et al., 2023; Lund et al., 2023).\u003c/p\u003e \u003cp\u003eThe findings also address concerns raised by previous studies regarding AI adoption in higher education. For example, Tareq et al. (2023) reported that gaps in technical skills and reliability issues may limit students\u0026rsquo; ability to fully benefit from AI. In line with this, the present study demonstrates that while AI has potential, its effectiveness is contingent upon students\u0026rsquo; RTS. Thus, universities should focus on capacity-building initiatives, training programs, and infrastructure provision to ensure equitable and effective AI adoption. Overall, this study extends existing knowledge by empirically confirming that AI enhances postgraduate research efficacy through the development of technical research skills. It bridges the gap in the literature regarding the interplay between technological tools and research efficacy, particularly in contexts where AI adoption is emerging, such as in Oman. The results have both theoretical and practical implications, highlighting the importance of integrating AI and skill development into postgraduate curricula to strengthen research competencies, foster innovation, and support academic success.\u003c/p\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eLimitations and Future Studies\u003c/h2\u003e \u003cp\u003eThe results of the present study should be interpreted with caution due to several limitations related to the research design and sample. First, the cross-sectional design limits the ability to establish causal relationships between the variables under investigation. Future studies could employ longitudinal or experimental designs to better examine potential causal links between AI utilization, research technical skills, and research efficacy among graduate students.\u003c/p\u003e \u003cp\u003eSecond, the sample consisted of only 30 postgraduate students from some departments within the College of Arts and Applied Sciences at Dhofar University. This relatively small and convenience-based sample may limit the generalizability of the findings to the wider population of graduate students at the university or other higher education institutions in Oman. Increasing the sample size and including students from multiple universities would help address this limitation in future research.\u003c/p\u003e \u003cp\u003eThird, the use of a self-administered questionnaire may introduce response biases. Although participants provided genuine answers, they might have tended to respond in ways perceived as socially desirable or aligned with academic expectations (Sekaran \u0026amp; Bougie, 2009). Future studies could mitigate this issue by collecting data from multiple sources, including interviews, focus groups, or observational measures.\u003c/p\u003e \u003cp\u003eFinally, contextual challenges specific to Dhofar University and Omani higher education may have influenced the responses. These include variations in students\u0026rsquo; prior exposure to AI tools, differences in access to digital resources across departments, and varying levels of technical support or guidance provided by faculty. Future research should consider these factors and explore how institutional and technological infrastructure affects students\u0026rsquo; AI utilization and research competencies.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusions","content":"\u003cp\u003eThis study demonstrated that Artificial Intelligence (AI) significantly enhances postgraduate students\u0026rsquo; Research Technical Skills (RTS), which in turn positively influence multiple dimensions of research efficacy, including background writing, literature review, methodology, analysis, and reporting. The findings confirm that AI facilitates cognitive and practical research tasks, supporting Bandura\u0026rsquo;s (1997) social cognitive theory on skill development and task performance. The study also highlights that students\u0026rsquo; technological readiness and access to AI tools are critical for maximizing these benefits, consistent with previous research (Gasaymeh, 2018; Ofem, Iyam, et al., 2024). Practically, universities should integrate AI applications and provide structured training to strengthen research competencies. Policy interventions should focus on equitable access and guidance for AI adoption. Overall, the study emphasizes that AI, combined with technical skill development, can serve as a transformative tool to enhance research efficacy and academic performance among postgraduate students, particularly in emerging contexts such as the Sultanate of Oman.\u003c/p\u003e "},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eEiman M. Koofan was responsible for developing the theoretical framework, conducting the literature review, and collecting the study data. Ahmed M. Al Rantisi carried out the PLS-SEM data analysis, interpreted the results, and led the drafting and critical revision of the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAdarkwah, M. A., Amponsah, S., Micheal, W., Ronghuai, D. D., Ahmed, T. B., Ahmed, H., et al. (2023). Awareness and acceptance of ChatGPT as a generative conversational AI for transforming education by Ghanaian academics: A two-phase study. Journal of Applied Learning \u0026amp; Teaching, 2(6). http://journals.sfu.ca/jalt/index.php/jalt/index\u003cu\u003e.\u003c/u\u003e\u003c/li\u003e\n\u003cli\u003eAl-Rantisi, A. M., ElShanti, N. H., \u0026amp; Harb, S. A. (2025). Challenges of using artificial intelligence in social work education. \u003cem\u003eSocial Work Education\u003c/em\u003e, 1\u0026ndash;18. https://doi.org/10.1080/02615479.2025.2483354\u003c/li\u003e\n\u003cli\u003eAguirre-Aguilar, G., Esquivel-G\u0026aacute;mez, I., \u0026amp; Edel-Navarro, R. (2024). AI in the development of research skills in postgraduate studies. \u003cem\u003eAlteridad: Revista de Educaci\u0026oacute;n, 19\u003c/em\u003e(2), 161\u0026ndash;176. https://doi.org/10.17163/alt.v19n2.2024.8557\u003c/li\u003e\n\u003cli\u003eArbab, A. N., Al-Saiari, M. A., Al-Mughairi, Y. M., Al-Mashaikhi, B. N., \u0026amp; Mudhsh, B. A. (2024). Student\u0026apos;s utilization and assistance of AI tools in assessment completion: Perceptions and implications. \u003cem\u003eQubahan Academic Journal, 4\u003c/em\u003e(3), 315\u0026ndash;332. https://doi.org/10.48161/qaj.v4n3a760\u003c/li\u003e\n\u003cli\u003eAl-Saiari, M. A., Al-Mughairi, Y. M., Al-Mashaikhi, B. N., \u0026amp; Mudhsh, B. A. (2024). Investigating the impact of training program on generative AI applications in improving university teaching. \u003cem\u003eQubahan Academic Journal, 4\u003c/em\u003e(3), 315\u0026ndash;332. https://doi.org/10.48161/qaj.v4n3a760\u003c/li\u003e\n\u003cli\u003eBandura, A. (1997). Self-efficacy: The exercise of control. W.H. Freeman and Company.\u003c/li\u003e\n\u003cli\u003eBonk, C. J., \u0026amp; Wiley, D. A. (2020). Preface: Reflections on the waves of emerging learning technologies. \u003cem\u003eEducational Technology Research \u0026amp; Development, 68\u003c/em\u003e(4), 1595\u0026ndash;1612. https://doi.org/10.1007/s11423-020-09809-x\u003c/li\u003e\n\u003cli\u003eBola\u0026ntilde;os, F., Salatino, A., Osborne, F. (2024). Artificial intelligence for literature reviews: opportunities and challenges. 57, 259. https://doi.org/10.1007/s10462-024-10902-3\u003c/li\u003e\n\u003cli\u003e\u003cspan dir=\"RTL\"\u003e Buniel, J. M., Intano, J., Cuartero, O., \u0026amp; Grustan, K. J. (2025). Modeling the influence of AI dependence on research productivity among STEM undergraduate students: Case of a state university in the Philippines. \u003cem\u003eFrontiers in Education, 10\u003c/em\u003e, 1535466. https://doi.org/10.3389/feduc.2025.1535466\u003c/span\u003e\u003c/li\u003e\n\u003cli\u003eCarmines, E. G., \u0026amp; Zeller, R. A. (1979). \u003cem\u003eReliability and validity assessment\u003c/em\u003e. Sage Publications. \u003c/li\u003e\n\u003cli\u003eChemers, M. M., Zurbriggen, E. L., Syed, M., Goza, B. K., \u0026amp; Bearman, S. (2011). The role of efficacy and identity in science career commitment among underrepresented minority students. \u003cem\u003eJournal of Social Issues, 67\u003c/em\u003e(3), 469\u0026ndash;491. \u003cu\u003ehttps://doi.org/10.1111/j.1540-4560.2011.01710.x \u003c/u\u003e\u003c/li\u003e\n\u003cli\u003eChin, W. W. (2009). How to write up and report PLS analyses. In Handbook of partial least squares: Concepts, methods and applications, Springer, pp. 655-690. \u003c/li\u003e\n\u003cli\u003eChin, W. W. (1998). Commentary: Issues and opinion on structural equation modeling, JSTOR\u003cstrong\u003e: \u003c/strong\u003evii-xvi.\u003c/li\u003e\n\u003cli\u003eChiu, T. K. F., Meng, H., Chai, C. S., King, I., Wong, S., \u0026amp; Yeung, Y. (2022). Creation and evaluation of a pre-tertiary artificial intelligence (AI) curriculum. \u003cem\u003eIEEE Transactions on Education, 65\u003c/em\u003e(1), 30\u0026ndash;39. https://doi.org/10.1109/TE.2021.3085878\u003c/li\u003e\n\u003cli\u003eClark, D. (2020). \u003cem\u003eArtificial intelligence for learning: How to use AI to support employee development.\u003c/em\u003e Kogan Page Publishers.\u003c/li\u003e\n\u003cli\u003eCohen, I. L., Liu, X., Hudson, M., Gillis, J., Cavalari, R. N., Romanczyk, R. G., \u0026hellip; Gardner, J. M. (2017). Level 2 screening with the PDD behavior inventory: Subgroup profiles and implications for differential diagnosis. \u003cem\u003eCanadian Journal of School Psychology, 32\u003c/em\u003e(3\u0026ndash;4), 299\u0026ndash;315. https://doi.org/10.1177/0829573517721127\u003c/li\u003e\n\u003cli\u003eCrawford, J., Cowling, M., \u0026amp; Allen, K. A. (2023). Leadership is needed for ethical ChatGPT: Character, assessment, and learning using artificial intelligence (AI). \u003cem\u003eJournal of University Teaching and Learning Practice, 20\u003c/em\u003e(3), 2. https://doi.org/10.53761/1.20.3.02\u003c/li\u003e\n\u003cli\u003eCreswell, J. W., \u0026amp; Creswell, J. D. (2018). \u003cem\u003eResearch design: Qualitative, quantitative, and mixed methods approaches\u003c/em\u003e (5th ed.). SAGE Publications. \u003c/li\u003e\n\u003cli\u003eDwivedi, Y. K., Kshetri, N., Hughes, L., Slade, E. L., Jeyaraj, A., Kar, A. K., \u0026hellip; Wright, R. (2023). \u0026ldquo;So what if ChatGPT wrote it?\u0026rdquo; Multidisciplinary perspectives on opportunities, challenges and implications of generative conversational AI for research, practice, and policy. \u003cem\u003eInternational Journal of Information Management, 71\u003c/em\u003e, Article 102642. https://doi.org/10.1016/j.ijinfomgt.2023.102642\u003c/li\u003e\n\u003cli\u003eEnciso, R. (2025). Artificial Intelligence in Research: Enhancing Efficiency or Compromising Integrity: A Systematic Review. 7, https://doi.org/ 10.13140/RG.2.2.17590.41289\u003c/li\u003e\n\u003cli\u003eFalk, R. F. and N. B. Miller (1992). A primer for soft modeling, University of Akron Press\u003c/li\u003e\n\u003cli\u003eField, A. (2018). \u003cem\u003eDiscovering statistics using IBM SPSS Statistics\u003c/em\u003e (5th ed.). SAGE Publications\u003c/li\u003e\n\u003cli\u003eFornell,C.G. \u0026amp; Larcker,D.F. (1981). Evaluating structural equation models with unobservable variables and measurement error\u0026rsquo;.\u003cem\u003e Journal of Marketing Research\u003c/em\u003e,\u003cem\u003e 18\u003c/em\u003e(1), 39\u0026ndash;50.\u003c/li\u003e\n\u003cli\u003eGasaymeh, A. M. M. (2018). The impact of digital storytelling on the quality of learning outcomes, motivation, and engagement among university students. \u003cem\u003eJournal of Educational Technology \u0026amp; Society, 21\u003c/em\u003e(4), 92\u0026ndash;101. https://doi.org/10.1037/edu0000332\u003c/li\u003e\n\u003cli\u003eHair, J. F., Hult, G. T. M., Ringle, C. M., \u0026amp; Sarstedt, M. (2022). \u003cem\u003eA Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM)\u003c/em\u003e (3 ed.). Thousand Oaks, CA: Sage.\u003c/li\u003e\n\u003cli\u003eHair, J. F., Ringle, C. M., \u0026amp; Sarstedt, M. J. L. r. p. (2013). Partial least squares structural equation modeling: Rigorous applications, better results and higher acceptance.\u003cem\u003e 46\u003c/em\u003e(1-2), 1-12. https://ssrn.com/abstract=2233795 \u003c/li\u003e\n\u003cli\u003eHeffernan, N. T., \u0026amp; Heffernan, C. L. (2014). The ASSISTments ecosystem: Building a platform that brings scientists and teachers together for minimally invasive research on human learning and teaching. \u003cem\u003eInternational Journal of Artificial Intelligence in Education, 24\u003c/em\u003e(4), 470\u0026ndash;497. https://doi.org/10.1007/s40593-014-0024-x\u003c/li\u003e\n\u003cli\u003eHenseler, J., \u0026amp; Sarstedt, M. (2013). Common beliefs and reality about partial least squares: comments on R\u0026ouml;nkk\u0026ouml; and Evermann. \u003cem\u003eOrganizational Research Methods\u003c/em\u003e,\u003cem\u003e 17\u003c/em\u003e(2). https://doi.org/\u003cu\u003e: 10.1177/1094428114526928 \u003c/u\u003e\u003c/li\u003e\n\u003cli\u003eHunter, A.-B., Laursen, S. L., \u0026amp; Seymour, E. (2007). Becoming a scientist: The role of undergraduate research in students\u0026rsquo; cognitive, personal, and professional development. \u003cem\u003eScience Education, 91\u003c/em\u003e(1), 36\u0026ndash;74. https://doi.org/10.1002/sce.20173 \u003c/li\u003e\n\u003cli\u003eJohnson, M. D., et al. (2006). \u0026quot;The evolution of loyalty intentions.\u0026quot; \u003cu\u003eJournal of marketing\u003c/u\u003e\u003cstrong\u003e70\u003c/strong\u003e(2): 122-132.\u003c/li\u003e\n\u003cli\u003eKasneci, E., Se\u0026szlig;ler, K., K\u0026uuml;chemann, S., Bannert, M., Dementieva, D., Fischer, F., et al. (2023). ChatGPT for good? On opportunities and challenges of large language models for education. \u003cem\u003eLearning and Individual Differences, 103\u003c/em\u003e, Article 102274. https://doi.org/10.1016/j.lindif.2023.102274\u003c/li\u003e\n\u003cli\u003eLiengaard, B. D., Sharma, P. N., Hult, G. T. M., Jensen, M. B., Sarstedt, M., Hair, J. F., \u0026amp; Ringle, C. M. (2021). Prediction: Coveted, Yet Forsaken? Introducing a Cross-validated Predictive Ability Test in Partial Least Squares Path Modeling. \u003cem\u003eDecision Sciences, 52\u003c/em\u003e(2), 362-392.\u003c/li\u003e\n\u003cli\u003eLifeWire. (2021). How AI could fake papers and wreck the scientific process. https://www.lifewire.com/how-ai-could-fake-papers-and-wreck-the-scientific-process-6834621\u003c/li\u003e\n\u003cli\u003eLin, M. P.-C., Liu, A. L., Poitras, E., Chang, M., \u0026amp; Chang, D. H. (2024). An Exploratory Study on the Efficacy and Inclusivity of AI Technologies in Diverse Learning Environments. \u003cem\u003eSustainability\u003c/em\u003e, \u003cem\u003e16\u003c/em\u003e(20), 8992. \u003cstrong\u003ehttps\u003c/strong\u003e://doi.org/10.3390/su16208992\u003c/li\u003e\n\u003cli\u003eLopatto, D. (2010). Undergraduate research as a high-impact student experience. \u003cem\u003ePeer Review, 12\u003c/em\u003e(2), 27\u0026ndash;30.\u003c/li\u003e\n\u003cli\u003eLund, B. D., Wang, T., Mannuru, N. R., Nie, B., Shimray, S., \u0026amp; Wang, Z. (2023). ChatGPT and a new academic reality: Artificial Intelligence-written research papers and the ethics of the large language models in scholarly publishing. \u003cem\u003eJournal of the Association for Information Science and Technology, 74\u003c/em\u003e(5), 570\u0026ndash;581. https://doi.org/10.1002/asi.24750\u003c/li\u003e\n\u003cli\u003eMachi, L. A., \u0026amp; McEvoy, B. T. (2016). \u003cem\u003eThe literature review: Six steps to success\u003c/em\u003e (3rd ed.). Corwin Press.\u003c/li\u003e\n\u003cli\u003eMertala, P., Fagerlund, J., \u0026amp; Calderon, O. (2022). Finnish 5th and 6th grade students\u0026rsquo; pre-instructional conceptions of artificial intelligence (AI) and their implications for AI literacy education. \u003cem\u003eComputers and Education: Artificial Intelligence, 3\u003c/em\u003e, https://doi.org/10.1016/j.caeai.2022.100095\u003c/li\u003e\n\u003cli\u003eOfem, U. J., Iyam, M. A., Ovat, S. V., Nworgwugwu, E. C., Anake, P. M., Udeh, M. I., et al. (2024). \u003cem\u003eArtificial intelligence (AI) in academic research: A multi-group analysis of students\u0026rsquo; awareness and perceptions using gender and programme type.\u003c/em\u003e\u003cstrong\u003e\u003cem\u003eJournal of Applied Learning \u0026amp; Teaching, 7\u003c/em\u003e\u003c/strong\u003e(1), 76\u0026ndash;105. https://doi.org/10.37074/jalt.2024.7.1.9\u003c/li\u003e\n\u003cli\u003eOfem, U. J., Nworgwugwu, E. C., Ovat, S. V., Anake, P. M., Anyin, N. N., Udeh, M. I. (2024). Predicting affective and cognitive learning outcomes: A quantitative analysis using climate change vectors. \u003cem\u003eEurasian Journal of Science and Environmental Education, 4\u003c/em\u003e(1), 1\u0026ndash;12. https://doi.org/10.30935/ejsee/14405\u003c/li\u003e\n\u003cli\u003ePopenici, S., Rudolph, J., Tan, S., \u0026amp; Tan, S. (2023). A critical perspective on generative AI and learning futures: An interview with Stefan Popenici. \u003cem\u003eJournal of Applied Learning and Teaching, 6\u003c/em\u003e(2), 311\u0026ndash;331. https://doi.org/10.37074/jalt.2023.625\u003c/li\u003e\n\u003cli\u003eSekaran, U. and Bougie, R. (2009) Research Methods for Business, West Sussex, Wiley.\u003c/li\u003e\n\u003cli\u003eSellin, N., \u0026amp; Versand, O. (1995). \u003cem\u003ePartial least square modeling in research on educational achievement\u003c/em\u003e. Waxmann. \u003c/li\u003e\n\u003cli\u003eShadle, S. E., Marker, A., \u0026amp; Earl, B. (2017). Faculty drivers and barriers: Laying the groundwork for undergraduate STEM education reform in academic departments. \u003cem\u003eInternational Journal of STEM Education, 4\u003c/em\u003e(1), 1\u0026ndash;14. \u003cu\u003ehttps://doi.org/10.1186/s40594-017-0062-7 \u003c/u\u003e\u003c/li\u003e\n\u003cli\u003eSharma, P. N., Liengaard, B. D., Hair, J. F., Sarstedt, M., \u0026amp; Ringle, C. M. (2023). Predictive Model Assessment and Selection in Composite-based Modeling Using PLS-SEM: Extensions and Guidelines for Using CVPAT. \u003cem\u003eEuropean Journal of Marketing, 57\u003c/em\u003e(6), 1662-1677.\u003c/li\u003e\n\u003cli\u003eShmueli, G., Ray, S., Estrada, J. M. V., \u0026amp; Chatla, S. B. (2016). The Elephant in the Room: Predictive Performance of PLS Models. \u003cem\u003eJournal of Business Research, 69\u003c/em\u003e(10), 4552-4564.\u003c/li\u003e\n\u003cli\u003eShmueli, G., Sarstedt, M., Hair, J. F., Cheah, J. H., Ting, H., Vaithilingam, S., \u0026amp; Ringle, C. M. (2019). Predictive Model Assessment in PLS-SEM: Guidelines for Using PLSpredict. European Journal of Marketing, 53(11), 2322-2347.\u003c/li\u003e\n\u003cli\u003eStrzelecki, A. (2023). Students\u0026rsquo; acceptance of ChatGPT in higher education: An extended unified theory of acceptance and use of technology. \u003cem\u003eInnovative Higher Education\u003c/em\u003e. https://doi.org/10.1007/s10755-023-09686-1\u003c/li\u003e\n\u003cli\u003eSwiecki, Z., Ruis, A. R., Gautam, D., Rus, V., \u0026amp; Williamson Shafer, D. (2019). Understanding when students are active-in-thinking through modeling-in-context. \u003cem\u003eBritish Journal of Educational Technology, 50\u003c/em\u003e(5), 2346\u0026ndash;2364. https://doi.org/10.1111/bjet.12869\u003c/li\u003e\n\u003cli\u003eTareq, R., Sumesh, N., Diane, K., Mulyadi, R., Fernando de Oliveira, S., Wagner, J., et al. (2023). The role of ChatGPT in higher education: Benefits, challenges, and future research directions. \u003cem\u003eJournal of Applied Learning \u0026amp; Teaching, 6\u003c/em\u003e, 1. https://doi.org/10.37074/jalt.2023.6.1.29\u003c/li\u003e\n\u003cli\u003eVij, S., Tayal, D., \u0026amp; Jain, A. (2020). A machine learning approach for automated evaluation of short answers using text similarity based on WordNet graphs. \u003cem\u003eWireless Personal Communications, 111\u003c/em\u003e(2), 1271\u0026ndash;1282. https://doi.org/10.1007/s11277-019-06913-x\u003c/li\u003e\n\u003cli\u003eWetzels, M., Odekerken-Schr\u0026ouml;der, G., \u0026amp; Van Oppen, C. (2009). Using PLS path modeling for assessing hierarchical construct models: Guidelines and empirical illustration. \u003cem\u003eMIS quarterly\u003c/em\u003e, 177-195.\u003c/li\u003e\n\u003cli\u003eWikipedia. (2025). \u003cem\u003eHallucination of artificial intelligence\u003c/em\u003e. In \u003cem\u003eWikipedia\u003c/em\u003e. Retrieved August 25, 2025, https://en.wikipedia.org/wiki/Hallucination\u003c/li\u003e\n\u003cli\u003eYuan, S., He, T., Huang, H., Hou, R., \u0026amp; Wang, M. (2020). Automated Chinese essay scoring based on deep learning. \u003cem\u003eCMC‑Computers, Materials \u0026amp; Continua, 65\u003c/em\u003e(1), 817\u0026ndash;833. https://doi.org/10.32604/cmc.2020.010471\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Artificial Intelligence, Research Technical Skills, Research Efficacy, Postgraduate Students, PLS-SEM","lastPublishedDoi":"10.21203/rs.3.rs-8566987/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8566987/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study examined the impact of Artificial Intelligence (AI) use and Research Technical Skills (RTS) on research writing efficacy among postgraduate students at Dhofar University, Sultanate of Oman. A quantitative cross-sectional survey under a positivist paradigm was conducted. The purposive sample included 30 students from the Departments of Social Sciences and Education, selected for their familiarity with AI tools. Ethical standards were maintained, including voluntary participation, informed consent, confidentiality, and formal approval from the Research Department. The study tested null hypotheses asserting no significant effects of AI and RTS on research writing efficacy. Results revealed significant positive relationships. AI significantly enhanced RTS (β\u0026thinsp;=\u0026thinsp;0.563; t\u0026thinsp;=\u0026thinsp;3.700; p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), which in turn strongly influenced all dimensions of research writing efficacy: Background Writing (β\u0026thinsp;=\u0026thinsp;0.822), Literature Review (β\u0026thinsp;=\u0026thinsp;0.564), Methodology (β\u0026thinsp;=\u0026thinsp;0.723), Analysis (β\u0026thinsp;=\u0026thinsp;0.693), and Reporting (β\u0026thinsp;=\u0026thinsp;0.667), all p\u0026thinsp;\u0026lt;\u0026thinsp;0.01. Predictive relevance was confirmed through the Stone-Geisser Q\u0026sup2; parameter. Findings highlight the critical role of AI and RTS in enhancing postgraduate research writing, emphasizing the need to integrate modern technological tools and technical skill development into higher education curricula to strengthen students\u0026rsquo; academic performance and research competence.\u003c/p\u003e","manuscriptTitle":"The Impact of Artificial Intelligence Use on Enhancing Research Efficacy among Postgraduate Students: The Mediating Role of Research Technical Skills Using PLS-SEM","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-13 08:07:48","doi":"10.21203/rs.3.rs-8566987/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"95666117-0ac2-484b-a771-02ee939ce04f","owner":[],"postedDate":"January 13th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-01-23T08:17:00+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-13 08:07:48","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8566987","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8566987","identity":"rs-8566987","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}