[OC] Aktualisiert und jetzt animiert! Welches Vitamin erhält die meiste Aufmerksamkeit für Forschungsarbeiten? – Mehr in Post
Von madkeepz
[OC] Aktualisiert und jetzt animiert! Welches Vitamin erhält die meiste Aufmerksamkeit für Forschungsarbeiten? – Mehr in Post
Von madkeepz
11 Kommentare
How does animating this help?
Why are the line colors changing?
Interesting data but it is kind of hard to read. There are alot of vitamins which means that the colours are similar hues. Worse.. they seemingly change intensity over time. Especially the orangey trace that finishes much higher than the rest – it seems to go from light to dark over time. It leaves me very unsure what colour I’m supposed to match to the legend.
This graph is an updated version from [this post](https://www.reddit.com/r/dataisbeautiful/comments/1m2mclw/oc_turns_out_one_vitamin_gets_way_more_research/).
In short, this shows the amount of articles in PubMed indexed by MeSH term for each vitamin from 1950 to 2023. Done with R. Data from rentrez package
Changes:
Moved to a more conventional style but decided to keep some stuff because it just looks pretty.
Animated with gganimate
Added a shift in alpha for each line equivalent to the inverse modulo of change between each year. I wanted to play with this and try to get something out of the animation. I ended up with this which is kinda cool because it sort of shows variation without having to introduce another color and messing things up
Shorted x axis to the end of 2023 because of publication bias. there IS a drop in publication that is at least 2 years before now but im not making any assumptions
So that big spike was definitely vitamin C, D, or E.
What is the value of the animation, except to make me wait for the information to appear?
The fact that the colors of all the vitamins is so close together and that they change over time makes this ugly. Would have preferred the lines stay solid.
Why do they change colours? Makes it unnecessarily difficult to tell which is which.
Instead of absolute values on Y (total publications), perhaps a area stacked percentage chart? That would highlight the swallowing majority vitamin(s) with diminishing minority vitamin(s) — e.g. https://www.highcharts.com/demo/highcharts/area-stacked-percent
Next up, label the #1 vitamin of the year under the date.
Data are looking beautiful.
How about graph the ones that are interesting in a few colors and the others in monochrome gray. That would improve the graph.