Further information on Fall 2020 courses


Sandy Johnson: Numbers in the News – Lessons from the Pandemic


Expanded description

Readings for the course will include Naked Statistics: Stripping the Dread from the Data by Charles Wheelan, available at Content Books or new or used online for less than $15. Also suggested (for those who do want to read about the math) is How Not to Be Wrong: The Power of Mathematical Thinking by Jordan Ellenberg. Discussion material will come from data generated in class and from recent newspaper and internet reports. Participants will be encouraged to bring examples and questions for discussion.


Schedule (subject to modification)

Week One: How Much? How Many? How Long?
Reading: Wheelan Chapter 2: Descriptive Statistics
Discussion: What do we really want to know? How do we count or measure it? How do we know if we are being accurate? And when we have the data, what do we compare our number to?

Week Two: What’s the Difference?
Reading: Wheelan Chapter 9: Inference
Discussion: How do two groups or two events differ? How can we tell why they are different? What differences are “significant” (and what do we mean by “significant”)?

Week Three: What’s the Connection?
Reading: Wheelan Chapter 4: Correlation
Discussion: Lots of events and characteristics are linked together; how do we tell which linkages are important and which ones are not? And which linkages demonstrate a causal relationship? (spoiler: none of them)

Week Four: Who Did We Ask?
Reading: Wheelan Chapter 10: Polling
Discussion: We can’t gather data from everyone (every event, every time) – so how do we decide who and when to do our research? Do our choices reflect biases in our thinking? Do they reflect social and financial inequality? How can we tell?

Week Five: What If It Is More Complicated?
Reading: Wheelan Chapter 11: Regression Analysis, and Chapter 12: Common Regression Mistakes
Discussion: Most events, personal traits, and accomplishments have many contributing causes. How do we consider them simultaneously? Does “controlling statistically” tell us what we want to know?

Week Six: What about algorithms? (and what are they?)
Reading: TBA
Discussion: Algorithms are the formulas that are used to determine what Amazon thinks you want to buy, who qualifies for a mortgage, sometimes even who gets into college. What data go into these formulas, and who develops them? Is there any accountability for them? How do algorithms contribute to Artificial Intelligence? To Facial Recognition Technology?

Week Seven: What If They are Trying to Fool Me?
Reading: Selections from How To Lie with Statistics (available online)
Discussion: How can the same data give different answers? What are common ways that data can be misrepresented? Who decides what will be reported?

Week Eight: How Do We Make Real Decisions?
Reading: Wheelan Conclusion
Discussion: How do statistics help us make good decisions about important questions? When are they likely to distort information or confuse our reasoning?


Potential sources for Real-life data:

            Sports Teams (especially Twins)
            Minnesota Zoo
            Public Schools (testing data)
            Retailers (Target?)
            UCC Research Office
            DNR (wildlife surveys)
            Political Polling (reported in the media)