▶️ Total olympic medals won in Paris 2024 and Human Development Index 🏅
➡️ https://www.businesstimes.com.sg/opinion-features/what-olympic-medal-table-really-tells-us
After reading the article we made this #boxplot using #LabPlot, an open source data analysis and visualization software.
The plot doesn’t provide answers, it rather invites some thinking.
#Olympics #Olympics2024 #France #China #USA #UnitedStates #UnitedKingdom #UK #Brazil #Australia #Japan #Italy #Canada #Germany #Italy #Netherlands #DataAnalysis #DataScience #OpenSource #FOSS
This is not beautiful, this is confusing.
Why are the Nederlands with a index of 10 more ro the right than Australia that happens to have an index of 10 as well?
The “information” of the x-axis is completely random or so it seems.
If you plotted medals (y-axis) over index (x-axis) there might be information in there.
Bit this? C’mon, thats embarrassing.
“doesn’t provides answers but invites thinking”… Nope. Doesn’t even help that as the X-Axis is unlabelled
A boxplot is a 1-dimensional plot. The data points are jittered along the x-axis to make them less crowded.
More on boxplots here:
➡️ https://labplot.kde.org/2021/08/11/box-plot/
➡️ https://userbase.kde.org/LabPlot/2DPlotting/BoxPlot
Yeah, not a good way to visualize as the relationship between medal count and HDI is not obvious as only outliers get highlighted and the lack of information on other countries actually invite doubt as to the story that the plot is trying to tell (for example, Singapore and Hong Kong have extremely high HDI but the sheer smallness of their population is a factor against a higher medal count). There’s nothing wrong with a traditional 2D scatter plot and axis-related box plots plotted against each axis separately
A boxplot is a visualization tool to quickly get an idea of how the data is distributed. In this population the outliers are so large that the info the real box + whiskers give is very low.
In your title you suggest investigating a relationship between total Olympic medals and HDI - why not choose a scatter plot here?
That the number in square brackets refers to the HDI rank only get’s clear on the second look.
The outliers being distributed over the X-Axis is confusing.
Sorry but this visualization is not beautiful, rather the wrong method used that cannot display the hypothesis stated in the title.
@LabPlot @dataisbeautiful
Doesn’t make sense unless you calculate in population size. Best way to do this is to have “# medals per capita ratio” on the vertical axis instead of simply # medals.
This doesn’t make any sense at all, it’s trying to force correlation to be causation as some political agenda that I can’t quite understand.
@stupidcasey Ok, let me explain: if you look at the chart it looks like the US is doing much much better than Australia. Twice the # of medals and about same score on human development index. Truth is US has over 12x the population of Australia.
If you adjust per my suggestion you’d see that Australia is doing ~6x better than US instead of US doing ~2x better than Australia as it is in the chart now. Much more realistic, isn’t it?
IM hearing a lot of correlation and not A whole lot of causation there, did the US have 12x the people competing in the Olympics?, did Australia pick its people from an even distribution of its populous or maybe just maybe did they Cherry pick from places that are better than the US like Sydney or Melbourne?
@stupidcasey And what all of this has to do with any political agenda; beats me! 🤣🤣🤣
Thank you for all your comments. A jittering of data points along the x-axis was used to avoid over-plotting. But yes, a scatter plot with a boxplot attached along the y-axis (to show outliers) may be more informative in this case.
