Open or close the menu

Using Generative AI in Research

Time and place: University of Jyväskylä, August 24-25, 2027

Learning goal and objectives: This course aims to provide doctoral students with hands-on experience and practical skills in leveraging Large Language Models (LLMs) for various aspects of academic research. The course is designed to accommodate both beginners and those with some AI experience. By the end of the course, students will be able to:

  • Understand the fundamental principles, capabilities, and limitations of generative AI and LLMs
  • Apply appropriate AI tools ethically and effectively across different research phases
  • Develop critical skills in prompt engineering for research-specific tasks
  • Implement AI-assisted workflows for literature reviews and academic writing
  • Utilize AI tools for both quantitative and qualitative data analysis
  • Evaluate the ethical and legal implications of AI use in academic contexts
  • Set up and use local LLM installations for research privacy and customization
  • Creatively and critically evaluate when and how AI tools can enhance specific research processes
  • Understand and apply AI agent workflows, including tool use and the Model Context Protocol (MCP), for autonomous research tasks

Instruction and examination: The course is structured as a three-day intensive workshop with additional pre- and post-course assignments. Each day runs from 9:15 to 17:00 and includes:

  • Interactive lectures on key topics
  • Guided hands-on exercises using various current frontier AI platforms
  • Small group discussions and collaborative problem-solving
  • Practical demonstrations of research-specific AI applications

The course will cover practical applications of AI such as:

  • Visualization of regression results
  • Full analysis of datasets
  • Grounded theory analysis using interview data
  • Coding of large numbers of survey responses
  • Improving academic writing
  • Ideating discussions sections
  • Writing feedback to students
  • Grading exams and assignments
  • Building AI agent workflows that connect to external tools and data sources via MCP
  • Designing multi-step agentic research pipelines (e.g., automated literature search, data collection, and analysis)
  • Understanding AI agent safety, oversight, and responsible deployment in research contexts
  • Automating research workflows with agentic coding tools (e.g., Claude Code, Gemini CLI, OpenAI Codex, Cline)

Practical activities will include:

  • Setting up and using Ollama to run open-source LLMs locally
  • Creating effective research-oriented prompts for different AI models
  • Accessing and utilizing API endpoints with provided API keys to complete practical research tasks
  • Testing special-purpose AI tools (e.g. Deep Research)
  • Comparative analysis of different AI models for specific research tasks
  • Guided programming exercises in Python (no prior programming experience required)
  • Hands-on setup and use of AI coding agents for research automation tasks
  • Configuring MCP servers to connect AI agents with research databases, file systems, and web services
  • Building end-to-end agentic workflows: from data retrieval to analysis to report generation

A course dinner will be arranged during the workshop to facilitate networking and informal discussion.

Assessment will be based on:

  • Pre-course assignment: Reflective essay (1000-1500 words) based on provided readings about AI in academic research (20%)
  • Active participation during the workshop (40%)
  • Post-course assignment: Critical reflection on potential applications of AI in the student’s field or a small practical experiment with an AI tool of choice, documented with screenshots and analysis (40%)

Accepted participants will receive a reading package containing articles, book chapters, and videos on using AI that should be completed before the course.

Credits: 4 ECTS

Grading: 1-5 based on pre-course assignment, participation, and post-course assignment.

Prerequisites:

  • Basic computer literacy
  • Interest in integrating AI tools into research processes
  • Background in research methods is beneficial but not required
  • No prior experience with AI tools is necessary – the course is designed for both beginners and more advanced users
  • Computer programming knowledge is not required
  • Students will get the most out of the course if they have a paid subscription to Claude, Gemini, or ChatGPT. The course instructors have a limited budget for providing access to less expensive frontier models for students who do not have a subscription

Admittance: Maximum 25 students will be admitted to the course. Applications should include a brief description (max. 300 words) of the student’s research interests and why they want to participate in the course. Application deadline is 14.6.2027. Accepted students will be notified by 16.6.2027. Submit your application to the course Moodle page at https://moodle.jyu.fi/course/view.php?id=34982 (use the Haka login)

Instructors:

Dr. Mikko Rönkkö is an Associate Professor of Entrepreneurship at Jyväskylä University School of Business and Economics (JSBE) and a docent at Aalto University School of Science. He has published extensively on research methods in both methodology journals and applied journals in management and information systems. Dr. Rönkkö serves as an Associate Editor for Organizational Research Methods and is on the editorial board of Entrepreneurship Theory and Practice. He runs a research methods-focused YouTube channel with over 10,000 subscribers and maintains expertise in statistical methods, research design, and entrepreneurship. Dr. Rönkkö brings practical experience as a former entrepreneur to his teaching and research.

Dr. Ville Heilala is a University Researcher at the University of Jyväskylä specializing in computational methods, artificial intelligence, and machine learning applications in human sciences. With a PhD in computer science and background in education, his research focuses on leveraging technology to enhance human capabilities. Dr. Heilala’s interdisciplinary expertise spans computational human sciences, AI in education, multimodal learning analytics, and computational psychometrics. He supervises doctoral researchers across diverse fields including pilot training assessment, virtual reality learning, and AI literacy among educators, bringing practical insights on AI implementation across various research domains.

Course coordinator and contact information: mikko.ronkko@jyu.fi