1. Leveraging AI Effectively and Ethically:
   Designing, refining, and critically evaluating AI prompts to generate useful outputs (e.g., tutorials, code, descriptive text) while recognizing and mitigating limitations such as algorithmic bias, attribution issues, and hallucinations.

2. Foundational Data Handling and Analysis:
   Using basic tools—from spreadsheets to Python programming—to combine, clean, organize, transform, and explore data. This includes practical skills in managing everyday datasets and recognizing tool limitations.

3. Quantitative and Statistical Reasoning:
   Applying basic statistical methods (e.g., calculating descriptive metrics, understanding variance, range, and quartiles) to interpret data, differentiate between correlation and causation, and support sound data-driven conclusions.

4. Predictive Modeling and Machine Learning:
   Familiarity and use of machine learning models using Python libraries, including proper data partitioning and the application of Machine Learning (ML) algorithms (regression, decision trees, K-nearest neighbor) to solve real-world problems.

5. Data Communication and Value Creation:
   Translating analytical insights into clear, compelling visual stories using tools like matplotlib, integrating these insights into business contexts to create value and drive data-informed decision-making.

1. AI Ethics, Prompt Engineering (7%, 3 hrs) 

Limitations of AI, Iteratively designing useful AI large language model (LLM) prompts

1. 1. Algorithmic Bias
      Students will be able to analyze and identify sources of bias in AI algorithms
      Example: Examine how a bank’s AI-driven loan approval engine might disproportionately decline applicants from certain ZIP codes based on historical data.

   2. Privacy
      Students will be able to evaluate privacy and consent considerations in everyday AI applications
      Example: Assess the ethical implications of personal data collection when using a smartphone voice assistant to set reminders or send messages.

   3. Design and refine prompts
      Students will be able to design and refine prompts for an AI large language model (LLM) to elicit specific, useful outputs
      Example: Iteratively craft and adjust a prompt so that an AI chatbot generates a comprehensive study outline summarizing the key concepts and example problems from a recent statistics lecture.
      
      

2. Basic Spreadsheets  (7%, 3 hrs) 

Basic spreadsheet use, size limitations of spreadsheets

2. 1. Organize and Analyze Everyday Data Using Spreadsheets:
      Students will be able to utilize basic spreadsheet functions—such as sorting, filtering, and applying formulas—to effectively manage and analyze small datasets.
      Example: A student uses a spreadsheet to record daily expenses for a month, applies formulas to calculate total spending and average cost per day, and creates a simple chart to visualize spending patterns.

   2. Using Pivot Tables
      Students will be able to create and interpret pivot tables in Google Sheets to organize and summarize data.

Example: Using a spreadsheet for personal finances, a pivot table will allow grouping monthly grocery purchases by category to identify which food group has the most expenses.

2. 3. Identify and Evaluate the Size Limitations of Spreadsheets:
      Students will be able to assess the performance and scalability constraints of spreadsheets and compare them to Big Data tools for handling larger datasets.
      Example: A student discusses how a personal budget spreadsheet is effective for tracking a few hundred records but becomes impractical when attempting to manage thousands of transaction records, such as company sales data for an entire year.
      
      

3. Value Creation (7%, 3 hrs) 

How to determine what adds value and how data analytics can provide value.
Incorporating the Revenue/Cost/Profit model, Lean Canvas, or Business ModelCanvas (BMC).

3. 1. Analyze Value Creation Using Data Analytics and Financial Models:
      Students will be able to apply basic data analytics techniques and the Revenue/Cost/Profit model to identify and quantify factors that add value in a business context.
      Example: Given a simplified dataset for a local micro-finance program from some Chicago neighborhood, a student will determine which factors most strongly contribute to increased household income, calculate basic profit margins, and explain how these insights can guide decisions to improve financial stability.

   2. Integrate Data-Driven Insights into Business Model Frameworks:
      Students will be able to apply and populate frameworks such as the Lean Canvas or Business Model Canvas (BMC) with data analytics findings, synthesizing insights into a coherent, actionable business strategy.
      Example: A student receives a case study on a startup aimed at promoting economic stability through affordable financial services. They will populate a Business Model Canvas with key data points—such as projected revenue streams and cost structures—and propose strategic improvements based on their analysis, illustrating how data-driven insights can enhance a business model..
      
      

4. Basic Statistics (9%, 4 hrs) 

Basic statistics, dealing with Types of Inference, Correlation vs. Causation.

Metrics and Bounds (Variance/standard deviation, range, quartiles).

4. 1. Calculate and Interpret Descriptive Statistics (Metrics and Bounds):
      Students will be able to use tools to compute descriptive statistics—including range, quartiles, variance, and standard deviation—and interpret what these metrics indicate about a dataset's spread and central tendency.
      Example: Given a dataset of weekly grocery spending over a month, a student will calculate the range (difference between the highest and lowest spending), determine the quartiles, and compute the variance and standard deviation. They will then explain how these statistics illustrate the consistency or variability in a family's spending habits.

   2. Differentiate Between Correlation and Causation Using Basic Inference:
      Students will be able to analyze relationships between two variables by calculating correlation coefficients and explain the difference between correlation and causation using everyday examples.
      Example: With a dataset comparing daily outdoor temperature and the number of iced drinks sold at a local convenience store, a student will compute the correlation between the two variables. They will then discuss how, although higher temperatures may be correlated with increased iced drink sales, this does not necessarily mean that higher temperatures cause more sales—other factors (like holidays or promotions) might also influence sales.
      
      

5. Python programming (13%, 6 hrs) 

Python programming with concepts presented within a problem context. Variables, Assignment, Logical and boolean operators, Functions & parameters.

Libraries: numpy, pandas.

5. 1. Apply Python Fundamentals to Analyze Data:
      Students will be able to use Python to define variables, assign values, use logical and boolean operators, and create functions with parameters, using these to manipulate and filter datasets and derive insights.
      Example: Given a dataset containing information on individual income, education level, and employment status, a student will create variables to store these data points, apply logical operators to filter out records for individuals with low income and limited education, and assign the resulting subset to a new variable that identifies communities at risk of limited social mobility.

   2. Design and Implement Functions with numpy and pandas to Examine Data:
      Students will be able to develop custom functions with parameters and leverage the numpy and pandas libraries to load, process, and analyze datasets, calculating descriptive statistics and visualizing data trends, providing actionable insights.
      Example: Using a dataset detailing household income and expenses across Chicago neighborhoods, a student will write a function that accepts a pandas DataFrame and a specified income threshold, utilize numpy to compute average income, variance, and quartiles, and generate visual plots that reveal disparities in financial stability, offering insights that could inform local economic support initiatives.
      
      

6. Exploratory Data Analysis (EDA) (18%, 8 hrs) 

Exploratory Data Analysis (EDA) to extract insight, identifying patterns and trends, touching on Extract / Transform / Load (ETL) to retrieve, clean & transform data.

Use online tools (CODAP), build our own using Python, then use an AI agent

6. 1. Data extraction and transformation
      Students will be able to perform data extraction and transformation on real‐world datasets
      Example: Import the weekly sales CSV provided by a marketing team, cleaning up missing transaction records, and standardizing date formats in Google Sheets to prepare the data for analysis.

   2. Apply Exploratory Data analysis (EDA) techniques
      Students will be able to apply exploratory data analysis techniques to uncover patterns and trends in business data
      Example: Use pivot tables and summary statistics to analyze social media metrics and identify which types of posts drove the highest engagement.

   3. Derive Insight
      Students will be able to interpret and communicate actionable insights from cleaned data
      Example: Visualize foot-traffic counts from a store survey and draft a one-page report recommending optimal staffing times based on observed peaks and troughs.
      
      

7. AI Tool Creation Using LLMs (13%, 6 hrs)

Automate routine tasks using a custom toolchain

Create and deploy a knowledge-sharing chatbot using  custom LLM tools

7. 1. Analyze and design a custom prompt-toolchain to automate routine tasks.
      Given a business scenario, students will be able to map out and configure a sequence of LLM prompts (a “toolchain”) that automatically processes input data, performs a transformation, and delivers a structured output.
      Example: Design an LLM workflow that, when given a list of appointments in plain text, extracts dates, formats them into calendar invites, and emails them to attendees, just like using a smart assistant to turn your scribbled to-do list into scheduled meetings.

   2. Develop and deploy a knowledge-sharing chatbot using custom LLM functions.
      Students will build a simple web-hosted chatbot by integrating one or more custom LLM tools that retrieve, summarize, and answer questions over a shared document corpus.
      Example: Create a “project helper” bot that, when coworkers ask “What are the main takeaways from last week’s planning meeting?”, searches a stored slide deck, generates a concise summary, and returns it in chat, much like texting a coworker for quick bullet-point notes.

8. Machine Learning (7%, 3 hrs) 

Machine Learning: partitioning data (Train vs Test), Learning methods (Supervised, Unsupervised, Reinforcement)

Basic Machine Learning Algorithms: Regression, Decision Trees, Nearest Neighbors. 

8. 1. Partition Data and Differentiate Learning Methods in Machine Learning:
      Students will be able to split a dataset into training and test sets and explain the differences between supervised and unsupervised learning using everyday examples.
      Example: A student is given a dataset of daily commuting times and traffic patterns. They partition the data into a training set (to build a predictive model for commute duration) and a test set (to validate the model’s accuracy), then explain that predicting commute times based on historical data is an example of supervised learning, while grouping similar traffic patterns without labels represents unsupervised learning.

   2. Implement Basic Machine Learning Algorithms to Solve Real-World Problems:
      Students will be able to apply fundamental machine learning algorithms—such as regression, decision trees, and nearest neighbor—to analyze and make predictions from everyday datasets.
      Example: Using a dataset of household energy consumption, a student builds a regression model to predict monthly utility bills, uses clustering to group households with similar usage patterns, and develops a decision tree to classify days as “high” or “low” consumption, demonstrating how these algorithms can support financial planning and energy-saving strategies.
      
      

9. Data Visualization & Storytelling (9%, 4 hrs) 

Data Visualization and using data for Story-Telling

Visualization using matplotlib and seaborn Python libraries

9. 1. Develop Data Visualization and Storytelling Skills Using Tableau:
      Students will be able to create interactive dashboards in Tableau and translate data into clear visual narratives that explain everyday trends and patterns.
      Example: A student gathers daily commuting times from a public transit dataset, builds a Tableau dashboard with charts and maps, and presents a story that explains how weather and time of day impact travel duration in their community.

   2. Implement Advanced Visualizations Using Python’s matplotlib and seaborn Libraries:
      Students will be able to design, generate, and interpret various visualizations using matplotlib and seaborn to effectively communicate insights from data in everyday contexts.
      Example: A student uses a dataset of monthly grocery spending to create a line chart with matplotlib that tracks spending trends and a boxplot with seaborn to illustrate spending variability, then explains how these visualizations reveal patterns that could help in personal budgeting decisions.
      
      

10. Projects & Durable Skills (11%, 5 hrs) 

Create three semester group projects, along with a comprehensive final group project.

Each project will include:

10. 2. 1. 1. Pre-project writing: 

             1. What is your understanding of the task? What are the goals that create the most value?  What questions are you attempting to answer? 

             2. What is your plan?  What is your “success criteria”? What do you already know, and what do you need to learn to accomplish your task?  How long do you think it will take?

          2. A presentation - the white paper in presentation form 

             1. Data - numbers, graphs, dashboards, etc.

             2. A reflection on the process, problems encountered, learnings of best GenAI practices

             3. The “one pager” - a summary report that is either a retrospective report on “what happened” or recommendations for action based on data findings.  We see X therefore we should do Y.  Or Given data XYZ, our options are ABC.

          3. Post-project reflection using the STAR method: Situation, Task, Action, Result
             This will be framed as an internship or job interview scenario.

     Learning Objectives:

10. 2. 1. Plan and Execute Data Science Projects Using AI-Enhanced Methods:
          Students will use a pre-project writing strategy to develop comprehensive project plans by clearly defining the task, setting measurable success criteria, and outlining a learning roadmap—including what they already know, what they need to learn, and estimated timelines.
          Example: In a group project analyzing local traffic data to improve public transportation, students articulate their understanding of the task, identify key questions (e.g., “What factors cause peak congestion?”), and outline a plan that includes gathering data, cleaning it, and determining success criteria such as reducing average commute time by a target percentage.

       2. Present and Communicate Data-Driven Insights Effectively:
          Students will be able to create professional presentations and summary reports (white papers and one-pagers) that incorporate numbers, graphs, and dashboards to tell a compelling story about their data findings, and translate their analysis into actionable value-driven recommendations.
          Example: A group uses sales data from a local coffee shop to build a dashboard and design a white paper presentation that explains trends in customer spending. They then synthesize their findings into a one-page report stating, “Given that sales peak on weekends, we recommend targeted promotions during weekdays to boost customer visits.”

       3. Reflect Critically on Project Processes Using Best Practices and the STAR Method:
          Students will be able to evaluate their AI-enhanced data science project execution by documenting challenges, actions, and outcomes through structured reflections.  These will  include their pre-project writing along with post-project writing using the STAR method (Situation, Task, Action, Result), highlighting their learnings in practice. Their results will be presented in a simulated job interview scenario.
          Example: After completing a group project on analyzing recycling rates in their neighborhood, a student uses the STAR method to describe the situation (inconsistent data sources), explain the task (to clean and analyze the data using GenAI tools), detail the actions taken (collaborative data cleaning, prompt refinement for code generation), and summarize the results (improved data accuracy and actionable insights for local recycling programs).
          
          

In addition to technical content area coverage, projects and reports will be cast in a business context, pressing into course goals of providing knowledge and skills that are helpful in finding a job.

ChatGPT o4-mini Prompt used to help create Learning Objective drafts:

Google Colab