Discussion Overview
The discussion revolves around the use of logarithmic scales to represent COVID-19 growth data, exploring how different scales can affect the interpretation of the data. Participants share insights on the advantages of logarithmic versus linear scales, express interest in visualizing the data on a log-log scale, and discuss the implications of using such representations.
Discussion Character
- Exploratory
- Technical explanation
- Debate/contested
Main Points Raised
- Some participants propose that logarithmic scales provide a clearer understanding of COVID-19 growth, as linear scales can obscure data from countries with lower case numbers.
- Others agree that on a log scale, the total cases outside of China appear to be growing exponentially, maintaining a straight line, which suggests consistent exponential growth.
- Several participants express interest in visualizing the data on a log-log scale, questioning how it might differ from log-linear representations.
- One participant raises concerns about the use of a logarithmic date axis, noting that it lacks a well-defined zero and may not effectively represent the data.
- Another participant mentions that while a log-log plot could revert the shape back to exponential, interpreting the time axis in logarithmic terms could be challenging.
- A participant shares that they obtained data from the source code of a webpage, specifying that all logs are to base 10 and noting a specific point in time where growth began to increase significantly.
Areas of Agreement / Disagreement
Participants generally agree on the potential benefits of using logarithmic scales for data representation, but there is no consensus on the effectiveness or interpretability of log-log plots, and some express skepticism about the use of a logarithmic date axis.
Contextual Notes
Participants mention limitations regarding data access for re-plotting and the challenges associated with interpreting logarithmic scales, particularly concerning the time axis.