Population Demography in Swirl

Author(s): Emily Weigel

Georgia Institute of Technology

Summary:
The students will learn to generate, test, and graphically represent basic hypotheses on data distributions using large datasets.

Licensed under CC Attribution-ShareAlike 4.0 International according to these terms

Version 5.0 - published on 10 Oct 2022 doi:10.25334/E760-Z411 - cite this

Description

This lesson centers around comparing the survivorship of two groups using a life table derived from tombstone information. Students will work through calculating some elements by hand via a worksheet alongside working in R. This swirl lesson goes step-by-step to show students how to do basic data manipulations, graphing, visual inspection of the data, and statistical tests as they work an example dataset with respect to a given hypothesis (whether males and females have the same life history curve). Students are then expected to repeat the coding steps they learned in R to test a hypothesis of their own.

Learning objectives:

 Complete a prelab reading to familiarize themselves with the goals of the lesson and the background ecological information of life tables Practice calculations and what the numbers in a life-table mean; generate associated hypotheses Practice coding elements necessary to generate datasets and test hypotheses based on the data Apply the coding learned through swirl to their own hypothesis to answer whether two groups in their data have the same survivorship

Notes

Please feel free to adapt this for use with other cemetery, longevity, or death datasets, provided that you are always respectful of the people your data represent.

Note:

Version 2.0 reorders the lessons and has code to get around package-dependences. Also, based on field tests, the lesson plan now includes code to help students with computer-specific issues on package installation found to be specific to Macs/Linux systems.

Version 3.0 builds on the last version by adding in commands which ensure matrices are built correctly in the newest packages. This version also adds a warning to examine the data within a generated matrix, specifically regarding 0's in comparison groups which can cause errors in running chi-square and G-test functions in edge cases.

Version 4.0 adds additional instructions to aid with the code when run in base R (outside of swirl/RStudio).

Version 4.1 corrects a minor typo and provides an alternate instruction for graphing code use outside of swirl.

Version 5.0 corrects a minor typo and doubles the dataset size based on updated records from the source cemetery.

Cite this work

Researchers should cite this work as follows: