Knud Berthelsen
SLAM 2031 Applied Data Analysis at Tulane, paper collage of data points streaming over the Mississippi River into a New Orleans skyline

SLAM 2031 · Tulane University

Applied Data Analysis

Students turn real questions into evidence, with AI and without it.

  • 3 credits
  • No prerequisites
  • Open to every major

The course

Sooner or later, someone hands over a spreadsheet and asks what it means.

SLAM 2031 is about finding out, with AI and without it. Students turn a real question into a testable one, design a survey, collect their own data, clean it, and run descriptive, diagnostic, and predictive analysis. AI does some of this well and some of it badly, and students learn to tell which.

They also see how data goes wrong: bad samples, statistical fallacies, correlation mistaken for cause, charts made to mislead. Then they design, analyze, and defend their own project, and one with a team.

By the end, students know what their numbers do and don't show, and how to say so.

Applied data analysis, paper collage of a hand holding a magnifying glass over scattered data points and a spreadsheet with a few cells highlighted
3

credits

A full Tulane course

0

prerequisites

Open to every major

12

topics

From first question to ethics

2

projects

One individual, one with a team

What it counts for

One course, several requirements

SLAM 2031 fits into the core curriculum and two minors at once.

Course requirements, paper collage of three puzzle pieces fitting together with data dots flowing between them

Newcomb-Tulane College core

Formal Reasoning

Satisfies the core curriculum requirement

SLAM minor

Analytics Outcome

Strategy, Leadership & Analytics Minor

SLAM minor

Economic & Financial Literacy

Satisfies this requirement too

AI Literacy Minor, proposed

Problem Solving with AI

Proposed for this category; approval is pending

Students should confirm with their advisor how the course fits their degree plan.

How students learn

Every topic, three ways

Each method is learned by hand first, then with AI, and then the two are compared.

Analysis with and without AI, paper collage of a hand checking a spreadsheet with a pencil, linked by coral threads to the same table on an AI screen
1

Without AI

Students learn the method and apply it themselves, so they know what a correct answer looks like.

2

With AI

They do the same work with current AI tools and see how much faster it goes.

3

Compare

They explain what changed, what the AI got wrong, and when to trust it.

Students are responsible for the evidence behind the answer.

Students do not build AI models. Like all of Professor Berthelsen's courses, SLAM 2031 is built on experiential learning and frequent, low-stakes assignments, using Google Sheets and free AI tools.

What students learn

From a question to a recommendation

The course follows the full analysis pipeline, through all five stages of analysis: descriptive, exploratory, diagnostic, predictive, and prescriptive.

Data analysis pipeline, paper collage of a question mark dissolving into data tiles that become ordered dots, rising bars and an arrow pointing forward
  1. 01

    Ask

    Turn a real problem into a question data can answer

  2. 02

    Collect

    Find sources and design surveys

  3. 03

    Clean

    Fix the messy data every real project starts with

  4. 04

    Analyze

    What happened, why, and what comes next

  5. 05

    Tell the story

    Charts and narratives that move a decision

  6. 06

    Call out misuse

    Spot misleading charts, bad samples, and false causation

Topics

Twelve topics

From the first question to the ways data goes wrong: truncated axes, bad samples, and correlation mistaken for cause.

Misuse of data, paper collage of a bar chart with a broken, truncated axis next to a scatter plot with a tangled thread labeled correlation is not causation
  1. 01

    What is data analysis?

  2. 02

    Tools for data analytics

  3. 03

    Formulating questions and research design

  4. 04

    Data types and sources

  5. 05

    Data collection and survey design

  6. 06

    Data cleaning and quality

  7. 07

    Descriptive and exploratory analysis

  8. 08

    Diagnostic analysis and pattern recognition

  9. 09

    Predictive and prescriptive analysis

  10. 10

    Data visualization and storytelling

  11. 11

    Statistical fallacies and misuse of data

  12. 12

    Validity, limitations, and ethics

Team data analysis project, paper collage of three students around a table of printed charts, one pointing at a heat map

Projects

Two projects, presented and defended

  • An individual projectEach student pitches a question, collects and analyzes the data, and presents and defends the findings.
  • A team projectTeams analyze data together and present and defend their conclusions.
  • Peer feedbackTeam members give each other confidential feedback on their work together.

Questions that could become projects

Paper collage of a crowd in front of an outdoor concert stage

Campus life

What predicts attendance at student events?

Paper collage of a stadium seen from above with price tags floating around it

Sports and culture

Which factors are linked to higher ticket prices?

Paper collage of hands holding a megaphone that sends a message out to a crowd

Organizations

Which outreach efforts bring in more volunteers?

Examples only. Students choose questions they care about.

Who it's for

Built for every major

  • No prerequisitesThe course builds from the basics.
  • No textbook to buyMaterials are provided on Canvas.
  • No final examGrades come from assignments, participation, the two projects, and peer assessment.
  • A laptop and a free Google accountFor Google Sheets and free AI tools.

When it's offered

SLAM 2031 runs as an in-person course during the regular semester and as an online course in the summer. The topics, projects, and approach are the same; the schedule and format differ. Current and upcoming sections are listed on the Teaching page.

Data analysis course for every major, paper collage of a student at a sunny desk with a laptop showing a spreadsheet and charts
Data cleaning, paper collage of jumbled shapes pouring through a sieve and coming out sorted by color into four bins

AI and data

AI is expected. So is checking it.

The use of AI tools is expected and encouraged, and training is part of the course. An AI tool alone is not a credible source, though: students verify AI output, describe how they used AI, and support every claim with credible data and an inline reference to its source.

An open paper door in a dark green wall, with a river of data dots flowing out of a bright valley beyond

The instructor

Knud E. Berthelsen

Knud Berthelsen is a Professor of Practice at Tulane University and Senior Advisor for AI Literacy at the Connolly Alexander Institute for Data Science. He also teaches DATA 2150 Artificial Intelligence Tools and SLAM 3111 AI and Society, and consults for and trains organizations on how AI works and how to work with it. Before focusing on AI, he spent a decade teaching international strategy and new venture management at Tulane's Freeman School of Business.

More in his CV and on the Teaching page. Questions about the course? Contact Professor Berthelsen.