Turning data into findings

Total suggested time: 60 minutes

Class exercise: Expanding your comparisons

Suggested time: 20 minutes

Using pivot tables, see if you can answer the following small questions as a class:

  • Which Connecticut town saw the largest number of COVID-19 cases?
  • What about deaths?
  • Which town saw the highest rate of COVID-19 deaths per patient?
  • Which town saw the highest rate of COVID-19 deaths per population?
  • When you divide cities by facility category (assisted living vs. nursing homes) how do the highest rates of death per patient compare?

Class discussion: Considering data’s flaws

Suggested time: 15 minutes

As a class, discuss the following questions:

  • Why might our “infection rate per patient” measure be flawed?
  • What factors might complicate comparisons or rankings between long-term care facilities?
  • Why were the death rates between federally regulated and state-regulated facilities so different?

Group work: Crafting a sentence

Suggested time: 15 minutes

In small groups, investigate whether the population of Connecticut facilities might have impacted the number of deaths due to COVID-19.

  • Examine the data you have on the number of patients in each Connecticut facility and determine numerical cutoffs to categorize each facility into one of four sizes based on the patient population: tiny, small, medium and large.
  • In your source data, create a new column and write a formula to categorize each facility into your categories.
  • Create a pivot table comparing the rates of death for each size group.
  • As a group, write a single sentence explaining, as simply as possible, what your analysis shows.

Class discussion: Comparing methodologies

Suggested time: 10 minutes

After completing the work with their groups, students should share their sentences and discuss:

  • Why did you choose the population cutoffs you did?
  • Did different choices yield different results?
  • What about the operation of these facilities might explain your findings?
  • What other data not currently part of your analysis might help explain the trends you noticed across categories of facilities, size, etc.?