Table Of Content
A student using AI tutoring systems to improve critical thinking.

AI Tutoring Systems and Critical Thinking Research Proposal

oleh fadistrik | Mei 5, 2026 | Research Proposal | 0 komentar

Assessing the Impact of Artificial Intelligence-Driven Tutoring Systems on the Development of Critical Thinking Skills in Secondary Education: A Research Proposal

Introduction

Background and Rationale

The integration of Artificial Intelligence (AI) into educational ecosystems has transitioned from a theoretical concept to a tangible paradigm shift over the past decade. AI technologies, ranging from predictive analytics to natural language processing, are increasingly being utilized to personalize learning experiences, optimize administrative workflows, and provide continuous, adaptive student support. Simultaneously, modern pedagogical frameworks emphasize that the primary objective of secondary education is no longer rote memorization, but rather the cultivation of higher-order cognitive skills. Among these, critical thinking—defined as the objective analysis, evaluation, and synthesis of information to form a reasoned judgment—stands out as an indispensable competency for navigating the complexities of the 21st-century knowledge economy. High school curricula are uniquely positioned at a critical developmental juncture where students transition from concrete to abstract reasoning. In this context, AI-driven tutoring systems hold profound potential. Unlike traditional didactic software, contemporary intelligent tutoring systems (ITS) can dynamically adjust scaffolding, pose Socratic questions, and simulate complex problem-solving scenarios, thereby theoretically fostering a cognitive environment highly conducive to the enhancement of critical thinking capabilities (Artificial Intelligence as a Teaching Tool to Promote the Development of Critical Thinking in Primary Education Students, 2024).

Research Objectives

The primary aim of this proposed study is to empirically investigate the efficacy of AI-driven tutoring systems in enhancing the critical thinking skills of high school students compared to traditional instructional methods. To achieve this, the research will address the following specific questions: (1) To what extent does the sustained use of conversational AI tutoring platforms improve students' measurable critical thinking proficiencies in high school science and humanities courses? (2) How do students perceive the pedagogical support provided by AI tutors in facilitating complex problem-solving? (3) What are the distinct variations in critical thinking skill development when AI tutoring is utilized as a primary instructional tool versus a supplementary one? Assessing AI's impact on critical thinking is of paramount significance; as educational institutions increasingly invest in AI infrastructure, it is imperative to establish empirical evidence demonstrating whether these tools genuinely elevate higher-order cognitive functions or merely expedite lower-level knowledge acquisition (Shahzad et al., 2025).

Literature Review

AI in Educational Technology

The extant literature on AI in educational technology highlights a rapid evolution from rudimentary rule-based systems to sophisticated machine learning algorithms capable of nuanced student modeling. Researchers have extensively documented the application of AI in automated grading, early warning systems for at-risk students, and adaptive learning environments. Previous studies specifically investigating AI-driven tutoring systems have generally reported positive outcomes concerning student engagement, content retention, and personalized pacing. For instance, systems utilizing natural language processing to mimic human tutoring dialogue have shown efficacy in guiding students through procedural knowledge tasks. However, a critical review of the current corpus reveals a significant research gap: while the literature is saturated with studies on AI's ability to improve standardized test scores and factual recall, there is a distinct paucity of longitudinal, empirical research examining the nuanced impact of these systems on the cultivation of critical thinking, argumentative logic, and evaluative reasoning (Singh, 2026).

Critical Thinking in Education

Critical thinking is a multifaceted construct. Drawing upon the consensus framework established by the Delphi Report (Facione, 1990) and Bloom’s Revised Taxonomy, it encompasses cognitive skills such as interpretation, analysis, evaluation, inference, explanation, and self-regulation. In secondary education, the role of critical thinking is heavily embedded within both national and international curricula standards, which advocate for pedagogical approaches that encourage inquiry-based learning and cognitive autonomy. Historically, studies examining methods to enhance these skills have championed active learning environments, collaborative problem-based learning, and Socratic questioning facilitated by expert educators. The literature suggests that critical thinking cannot be passively absorbed; it requires deliberate, scaffolded practice where students are challenged to defend their cognitive processes. This established understanding provides a vital theoretical framework for evaluating whether AI systems can successfully replicate or augment the requisite conditions for higher-order cognitive development (Yazidi, 2023).

Methodology

Research Design

This study will employ a mixed-methods explanatory sequential research design, structured to collect and rigorously analyze both quantitative and qualitative data. The quantitative phase will utilize a quasi-experimental approach involving a pre-test/post-test control group design to establish causal inferences regarding the impact of the AI tutoring system on critical thinking metrics. Following the quantitative analysis, qualitative data will be gathered through semi-structured interviews and focus groups. The mixed-methods design is highly justified for this inquiry; while quantitative assessments will provide objective, generalizable data regarding cognitive gains, the qualitative phase is essential for elucidating the underlying mechanisms, user experiences, and pedagogical phenomena that drive those statistical outcomes. The research process will proceed through four distinct phases: baseline assessment, a 16-week intervention utilizing the AI platform, post-intervention assessment, and finally, qualitative exploration (Ahmad et al., 2025).

Participants and Sampling

The target population for this study encompasses high school students (typically aged 15-18) enrolled in diverse, comprehensive secondary educational institutions. A stratified random sampling method will be utilized to select approximately 400 participants across four distinct high schools, ensuring representation across varying socioeconomic backgrounds, academic proficiency levels, and demographic profiles. Participants will be stratified into an experimental group (utilizing the AI tutoring system) and a control group (receiving traditional classroom instruction). Ethical considerations will be paramount throughout participant recruitment and engagement. Prior to data collection, the research protocol will be submitted for Institutional Review Board (IRB) approval. Informed consent will be obtained from all participating educators and the legal guardians of the students, alongside assent from the minors themselves. Strict protocols for data anonymization and privacy will be enforced to protect participant identities and academic records (Cohen et al., 2017).

Data Collection Methods

Data collection will leverage a combination of standardized psychometric instruments, system-generated analytics, and qualitative protocols. The primary quantitative tool will be the Cornell Critical Thinking Test (Level X), administered as both a pre-test and post-test to objectively measure shifts in deductive reasoning, inductive reasoning, and evaluative skills. The AI-driven tutoring system will be fully integrated into the experimental groups' curriculum for a duration of one academic semester (approximately 16 weeks), specifically targeting complex subjects such as Advanced Biology and World History. During this period, the AI platform will continuously capture telemetry data, including interaction frequency, questioning pathways, and time-on-task. Qualitative data will be collected post-intervention via semi-structured interviews with a purposive subsample of 40 students and their respective teachers, focusing on their subjective experiences with the AI’s Socratic prompting (Ahmad et al., 2025).

Data Analysis

Quantitative data analysis will be conducted utilizing the Statistical Package for the Social Sciences (SPSS) or R. To evaluate the impact of the intervention, an Analysis of Covariance (ANCOVA) will be performed, using pre-test scores as the covariate to control for initial differences in critical thinking abilities between the experimental and control groups. Further descriptive and inferential statistics will explore correlations between AI platform usage metrics and cognitive gains. Qualitative data derived from interview transcripts will be imported into NVivo qualitative data analysis software. A deductive and inductive thematic analysis approach will be applied to identify recurring patterns, pedagogical friction points, and emergent themes related to the students' cognitive engagement with the AI tutor. The integration of both data sets will occur during the interpretative phase to construct a comprehensive narrative of the intervention's efficacy (Arikan, 2025).

Expected Outcomes

Anticipated Findings

It is anticipated that the study will reveal a statistically significant improvement in the overall critical thinking scores of the experimental group compared to the control group, particularly in the domains of inference and explanation. Furthermore, qualitative findings are expected to demonstrate that students perceive conversational AI systems as non-judgmental environments that encourage intellectual risk-taking and deeper inquiry. These findings could represent a profound contribution to the field of educational technology by shifting the discourse from AI as a mere knowledge-delivery mechanism to AI as a viable cognitive scaffolding tool. The implications for high school curricula are substantial; empirical evidence supporting AI's role in higher-order thinking could drive the adoption of blended learning models where human educators are liberated from routine remedial tasks to focus on complex, dialectical classroom facilitation (Fasahat et al., 2025).

Limitations and Delimitations

Despite rigorous methodological planning, several potential limitations must be acknowledged. The "novelty effect" poses a risk, wherein initial increases in student engagement and cognitive effort may be attributed to the newness of the AI technology rather than its enduring pedagogical value. Additionally, unforeseen technical issues, such as variable internet bandwidth across different school districts or algorithmic hallucinations within the AI system, might disrupt the continuity of the intervention and skew the interaction data. To manage the scope, this study contains specific delimitations: the research is strictly confined to secondary education (grades 10-12) and will only investigate the utilization of one specific architectural framework of conversational AI tutoring. Therefore, the findings may not be inherently generalizable to primary education settings or to other models of artificial intelligence, such as strictly predictive analytics software (Joyner, 2024).

Conclusion

Summary of Proposal

This research proposal outlines a comprehensive, mixed-methods investigation designed to rigorously assess the impact of AI-driven tutoring systems on the development of critical thinking skills among high school students. By integrating standardized cognitive assessments with deep qualitative inquiry over a 16-week intervention period, the study aims to fill a critical void in current educational technology literature. If the anticipated findings hold true, this research will provide foundational empirical evidence that AI can actively foster higher-order cognitive competencies, fundamentally transforming how educational practices are structured and delivered in the modern classroom. It emphasizes that as AI becomes ubiquitous, educational research must pivot toward understanding its capacity to cultivate the analytical minds required for the future.

Future Research Directions

Building upon the anticipated findings of this study, several vital avenues for future research emerge. A priority should be the execution of longitudinal studies spanning multiple academic years to determine whether the critical thinking gains facilitated by AI tutors are sustained over time and successfully transferred to novel, non-academic problem-solving scenarios. Furthermore, future research should explore expanding this inquiry to diverse educational levels, including primary and post-secondary environments, to establish a developmental continuum of AI-assisted cognitive growth. Ultimately, researchers must continuously investigate the long-term sociological and pedagogical implications of AI in education, ensuring that as machine intelligence evolves, it remains rigorously aligned with the humanistic goal of cultivating independent, critical, and ethical thinkers.

References

Ahmad, I., Maqbool, D., Kanwal, S., Taj, S., & Shahid, J. (2025). Exploring AI-Based Interventions for Developing Critical Thinking Among University Students. The Critical Review of Social Sciences Studies, 3(4), 563–571. https://doi.org/10.59075/0r2gvf74
Arikan, A. (2025). The Effect of Artificial Intelligence-Supported Activities on Academic Achievement, Critical Thinking and Innovation Skills in Social Studies Classes. Anadolu Kültürel Araştırmalar Dergisi, 9(4), 917–950. https://doi.org/10.63556/ankad.v9i4.269
Artificial Intelligence as a Teaching Tool to Promote the Development of Critical Thinking in Primary Education Students. (2024). Editorial Octaedro. https://doi.org/10.36006/09643-1-04
Cohen, L., Manion, L., & Morrison, K. (2017). Sampling (pp. 202–227). Routledge. https://doi.org/10.4324/9781315456539-12
Fasahat, M., Bandeali, D., Sultan, Y., & Rehan, M. (2025). The Role of Artificial Intelligence in Fostering Critical Thinking among University Students. The Critical Review of Social Sciences Studies, 3(2), 1812–1827. https://doi.org/10.59075/ka4pqm42
Joyner, D. A. (2024). A Teacher’s Guide to Conversational AI. Routledge. https://doi.org/10.4324/9781032686783
Shahzad, S. A., Raza, M. M., Malhi, S. D., Ain, Q., & Malhi, D. K. (2025). Integrating Artificial Intelligence To Foster Critical Thinking And Problem-Solving In 21st-Century Classrooms. Research Journal for Social Affairs, 3(5), 635–648. https://doi.org/10.71317/rjsa.003.05.0362
Singh, G. (2026). The Role of Artificial Intelligence (AI) in Personalized Learning. International Journal of Scientific Research in Engineering and Management, 10(02), 1–9. https://doi.org/10.55041/ijsrem56942
Yazidi, R. E. (2023). Strategies for Promoting Critical Thinking in the Classroom. International Journal of English Literature and Social Sciences, 8(2), 026–040. https://doi.org/10.22161/ijels.82.5

0 Komentar

Kirim Komentar

Alamat email Anda tidak akan dipublikasikan. Ruas yang wajib ditandai *

style="display:block" data-ad-format="autorelaxed" data-ad-client="ca-pub-4224751382866031" data-ad-slot="1131314884">