https://alhenry.shinyapps.io/covid-o-meter/
Covid-O-Meter (or covidometer) is an open access, daily-updated interactive web app designed to track and visualise various statistics related to the Covid-19 pandemic.
Covid-O-Meter is built on R Shiny framework using the John Hopkins CSSE Covid-19 Dataset.
Since the beginning of the Covid-19 outbreak, there are already many sophisticated stasticial models, descriptive statistics, and data visualisations arising from various projects from across the globe (for a list, see The Coronavirus Tech Handbook Infographics collection).
Each of these projects has its own perks and tells a different story. For a complex and worldwide problem such as Covid-19, however, narratives on specific problems on certain population or areas may not always fit well in other context. A combination of domain expertise and local knowledge is often required to make sense of a specific subset of the data.
To aid this time-consuming data sleuthing work, Covid-o-meter provides a simple interface for users to navigate through different parts of the dataset directly and to make their own discoveries.
- Cumulative Number of cases
- Number of new cases
- Cumulative number of deaths
- Number of new deaths
- Case fatality rate = Number of deaths (cumulative) / Number of cases (cumulative)
- Number of cases (per-day time-lapse)
- Number of deaths (per-day time-lapse)
The Y-axis represents the corresponding statistics on a log base 2 scale (default for number of cumulative / new cases /deaths) or a linear scale (default for case fatality rate).
The log base 2 scale is used to represent doubling rate due to the exponential nature of the disease spread (for discussion, see e.g. explainer by John Burn-Murdoch).
The X-axis represents day after first N cases (for case statistics) / deaths (for death statistics) on a one-unit increase linear scale.
The following example illustrates how the day variable is calculated for case statistics if N is set to 100:
| Date | Number of cases | Day |
|---|---|---|
| 1 March 2020 | 88 | 0 |
| 2 March 2020 | 93 | 0 |
| 3 March 2020 | 105 | 1 |
| 4 March 2020 | 134 | 2 |
| 5 March 2020 | 199 | 3 |
| ... | ... | ... |
On mouse hover, each data point (represented by dot) shows:
- Country name
- Date
- Day after first N cases / deaths (X-axis coordinate)
- Value of the statistics (Y-axis coordinate)
The draggable sidebar contains functionality to adjust data and plot parameters as follows:
| Parameter | Description |
|---|---|
| Countries | Add / remove countries to display (limited to a maximum of 20 to allow some physical distancing between data points) |
| Period | Adjust range of dates to use in calculation |
| Minimum cases / deaths for day 1 | Minimum number of cases / deaths to be counted as Day 1 |
| Range of days to display | Lower and upper limit of day (i.e. the X-axis) to display |
| Scale | Select between linear / log base 2 scale for Y-Axis |
| Show doubling rate | Show / hide doubling rate for number of cases / deaths per 1, 3, and 7 days (hover on the lines for description) |
After adjusting the desired parameters, press the Update plots / animation button to see the change
Note:
ScaleandShow doubling ratewill trigger automatic change on click- Generating a new animation could take ~2 minutes to complete
The download button will save a static version of the corresponding plot as .png file.
Use the width / height button to adjust the size (in inches).
Note: Animation can be downloaded as a .gif image via web browser interface
(e.g. on Google Chrome press right-click then select Save Image As...)
Pull requests and suggestions are welcome.
Email: [email protected]
This work uses the publicly available Covid-19 dataset provided by John Hopkins CSSE. Please read and follow the terms of use of the data.
This work is inspired by other similar works, e.g. Covid-19 charts by John Burn-Murdoch of Financial Times, 91-Divoc project by Wade Fagen-Ulmschneider, Covid-19 Tracker by Dr Edward Parker and Quentin Leclerc of The London School of Hygiene & Tropical Medicine