LLM Text Generation in R: Practical Prompt Engineering - A Short Course

By the end of this course, you will be able to:

• • Set up and configure LLM APIs (OpenAI and Groq) within the R environment.
  • Apply advanced prompt engineering techniques including role prompting, few-shot learning, chain-of-prompting, and chain-of-thought reasoning.
  • Generate high-quality survey items and research materials grounded in psychological theory.
  • Extract key concepts from research papers (PDFs) to inform item or text development.
  • Implement prompt chaining for complex, multi-step analysis tasks.
  • Create automated pipelines for large-scale text generation and analysis.
  • Minimize hallucinations through evidence-based grounding techniques.

This is a hands-on course with instructor-led software demonstrations and guided exercises. These guided exercises are designed for the R programming language, so you should use a computer with a recent version of R (version 4.1.3 or later) and RStudio (version 2022.02.1+461 or later). The names of the packages needed for the course, and how to install them, will be provided during the course.

To follow along with the course exercises, you should have good familiarity with the use of R, including opening and executing data files and programs, as well as performing very basic data manipulation and analyses.

If you’d like to take this course but are concerned that you don’t know enough R, there are excellent online resources for learning the basics. Here are our recommendations.

Anyone interested in prompt engineering and text generation with OpenAI Models who has a good working knowledge of R

Day 1: Exploring the fundamentals of OpenAI models

Introduction to LLMs and API Setup (OpenAI Models: GPT-3.5-TURBO/GPT-4, GPT-4o, GPT-5, the open source OpenAI models GPT-OSS-120b, GPT-OSS-20b, plus other open source models such as Llama-3.3, Llama-4, and others)

• • Understanding the concept of language models
  • Differentiating between the models
  • Differentiating API systems with GPUs and LPUs
  • Exploring real-world applications

Setting up R environment for LLMs

• • Installing necessary libraries and packages
  • Authenticating with the OpenAI API
  • Authenticating with the Groq Cloud System API
  • Obtaining API keys

Basic text generation

• • Creating your first LLM request in R via OpenAI and Groq Cloud APIs
  • Generating text for simple prompts in R
  • Anatomy of effective prompts for research applications
  • Parameter tuning (temperature, max_tokens) for optimal outputs
  • Handling responses and formatting output in R
  • Hands-on practice

Day 2: Advanced applications and prompt engineering with OpenAI and open-source models

Role prompting and persona development

• System vs. user messages for behavioral control
• Creating expert personas (psychometrician, market researcher, educational psychologist)
• Comparative analysis of role-specific outputs
• Hands-on practice: Generating Big Five personality items using expert personas

Few-shot learning and multi-domain item generation

• Few-shot prompting with domain-specific examples
• Building automated item generation functions
• Creating complete survey batteries with CSV output
• Quality assessment and validation techniques
• Hands-on practice: Job satisfaction scale development across multiple domains

Chain-of-thought reasoning

• Step-by-step problem solving for complex research questions
• Methodological decision-making with LLM assistance
• Statistical analysis planning and assumption testing
• Hands-on practice

Evidence-based grounding and PDF integration

• Minimizing hallucinations through document grounding
• PDF text extraction and preprocessing techniques
• Concept extraction from research literature
• Hands-on practice: Generating theory-informed items from academic papers

Generating text data for topic modeling as an exercise in R

• Text Generation via OpenAI models (and open models) for topic modeling
• Hands-on practice

“Dr. Golino’s course is an engaging combination of background, conceptual, and applied knowledge. It helped me move from being interested and curious to developing a solid baseline understanding of the topic and its issues and then applying that knowledge to my own work.”
  Maggi Mackintosh, U.S. Department of Veterans Affairs