My students have been collecting bad COVID-19 graphs and I love them so far so I am going to post them in this thread
graphs like this are why scientists hate pie charts. this graph won't photocopy well, and there are too many slices to be able to clearly read.I am intrigued by the fact a single cruise ship is as big a slice as Switzerland is, though. the data in this is outdated, but still, wow
if you don't label your axis or provide a legend, how are we supposed to know what's happening? also, it's only ~sometimes a good idea to contrast two lines of data. sometimes it's better to make two graphs. I can't say if it's a good idea here because I don't know what this is
we choose to present things in graphs instead of tables when the visualization makes it easier for our audience to see a pattern. do you see a pattern here? I don't.
this won't photocopy well, but it's also quite cluttered. if you were going to put this on TV? there are better ways to do this.
I will add more as they turn their assignments in and I grade them!
This graph appears to show a trend and it's easy to follow... except the data isn't particularly meaningful because every country has a different population. It would be ~better if the graph showed cases per 100K people, but still not perfect since testing rates vary.
This is overwhelming to look at and the spots overlap in ways that obscure the data. Apparently zooming in is possible, but you want visuals to be QUICKLY readable. No one should have to squint or study or hurt their brains trying to figure them out.
QUICK SCICOMM TUTORIAL
Good visuals:
1) are easy to read
2) provide meaningful, unambiguous data
3) are tailored to an audience & purpose
4) do a better job at communicating information than words/tables alone
5) don't distort data
6) don't contain more or less info than needed
My student is "pretty sure" this one is a joke. Lol.
The number of positive tests is not meaningful in the absence of 1) populations 2) testing availability

Of COURSE big populations have big numbers. & There are no +'s w/o tests

To be meaningful, this would need to be per 100K people with estimates for availability of testing
Lets forget for a moment that there are no axis labels and we don't actually know what these bars mean. How tf do you get five bars from "good or bad"?????
In the first tweet I said "bad graphs" and I should specify the assignment was "bad graphs and visuals" because these aren't all ~graphs per se
I don't even know where to start with this one.

It's Florida... but that's the whole Southeastern US. What do the dots mean? Why are there different colors? Is that important somehow? Why is there a bar graph next to the map? Is that new cases in the whole US? Or just Florida?
I'm going to go on the record as not liking any visual that compares a lack of cases in counties that may not be testing to the # of cases in counties that are testing as if that means anything.

Also, like, the unnecessary chaos here. Does this tell us anything but population?
Jesus take the wheel.
I actually have seen some good graphs, too. In the midst of this thread of terrible graphs, I just want you to know good ones are possible. I have seen several good ones from https://ourworldindata.org . Meaningful, clear ones that include context
This one is not one of the good ones, though.

👏🏻use 👏🏻legends 👏🏻please👏🏻
Do I need to tell you why this one is bad or
As far as maps go, I don't hate this one. But here is the thing
Most maps only tell us where people live. And it's the PEOPLE that matter.
I have a PhD and I do not understand this graph
One thing a lot of scientists/data-people do wrong when it comes to graphs is they don't adjust to their audience. That last graph might be FINE if it's the sort you're used to looking at. But if you're not? For public-facing graphs, you need to adjust.
Logarithmic graphs are very helpful for epidemiologists! Normal people? Not as much.
Logarithmic graphs are very helpful for epidemiologists! Normal people? Not as much.
Logarithmic graphs are very helpful for epidemiologists! Normal people? Not as much.
Cherry picking data that isn't meaningful just to push an agenda makes me angry. This graph is COMPLETELY not meaningful because of COURSE countries less affected--or countries hiding true numbers, or with better management--will not have the same sorts of lockdowns. FURTHERMORE
Let's look at what happened to Sweden after their non-lockdown. (This is another GOOD graph. Shoutout to https://ourworldindata.org ; I am becoming a big fan of y'all)
idk if this is about COVID19 or not but lmao
So, graphs like this can be useful. But I'm still not a big fan. Every country has a different population--and every population is different. Raw number of deaths doesn't tell us very much if you're comparing the number of deaths in a small country against a big one.
what on earth
👏🏻 don't 👏🏻 use 👏🏻 logarithmic 👏🏻scales👏🏻for👏🏻 lay 👏🏻audiences 👏🏻
I am speechless
I had to look at this three times before I realized there were two sets of data in this graph. My student suggested a form change from bar graph to line graph would improve this and I don't disagree.
lol
Someone find me a GOOD map challenge
*breathe in for four counts, breathe out for eight counts*
My student just wrote underneath a graph, "if anyone could explain to me what is happening here I would be so grateful" and lmao what a mood
So many of you tagged me in this one that I made it my header

It’s art. Truly. Wow.
Just gon leave this here
https://twitter.com/rodolfoalmd/status/1255203098764730369?s=21 https://twitter.com/rodolfoalmd/status/1255203098764730369
don’t 👏🏻distort👏🏻data👏🏻for👏🏻your👏🏻agenda👏🏻
https://twitter.com/charleststella/status/1261633629702111232?s=21 https://twitter.com/CharlesTStella/status/1261633629702111232
You are all my students now and I am going to add what you send me to this thread
Like this one. Read the replies (I know, you normally never should but this is an EXCEPTION) because there is a fascinating discussion about malice versus incompetence and how this happened. H/t @Lyndo12, thanks for sending me this disaster lol
https://twitter.com/andishehnouraee/status/1284237474831761408?s=21 https://twitter.com/andishehnouraee/status/1284237474831761408
how DNA is like inductors 🙃
https://twitter.com/thoughtsofaphd/status/1286006022788841474?s=21 https://twitter.com/thoughtsofaphd/status/1286006022788841474
included without comment
https://twitter.com/itsyabitsjace/status/1249708176242507778?s=21 https://twitter.com/itsyabitsjace/status/1249708176242507778
this one makes me feel embarrassed for all of us tbh
https://twitter.com/littlechambers/status/1290635729983471622?s=21 https://twitter.com/littlechambers/status/1290635729983471622
This one is not COVID but about 7 of you tagged me with it and it is very good (bad)

Two big (lol) errors:

1) y axis truncation
2) width/area variance of figures distorts data
https://twitter.com/kejames/status/1291555426706874369?s=21 https://twitter.com/kejames/status/1291555426706874369
Also here is my 6’0” Viking self on this graph
Chart seems fine idk
We have a winner for the “someone make me a good map” challenge and it is @KylaPadbury
https://twitter.com/kylapadbury/status/1292340150182141954?s=21 https://twitter.com/KylaPadbury/status/1292340150182141954
1) it has a clear audience (Twitter) and clear purpose
2) it made me laugh
3) it shows us something OTHER than population
I am just gon randomly include non COVID ones now
I laughed at this one because Heather & I had just been talking about nonbinary gender & how hers manifests, which is a little different than other people’s & some of these actually aren’t bad for her, but it’s a ridiculous representation of ~all enbies
https://twitter.com/ciarkents/status/1303009539650760708?s=21 https://twitter.com/ciarkents/status/1303009539650760708
The one blue pirate leg one is @L_Historienne, her gender is “one blue pirate leg”
You can follow @Swilua.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled:
There are a lot of misconceptions about the Turkish diaspora in Germany. Since i just finished writing an essay about this topic I thought it's time to give other people
Read more
Las mejores calculadoras online en 2020A thread http://totumat.com/2020/11/26/las-mejores-calculadoras-online-en-2020/ Cuando se estudian asignaturas que requieren de cálculos complicados, nada mejor que contar con una buena calculadora.
Read more
Mitigating #ClimateChange is as much about taking up new techs as about leaving others behindMost studies focus on #coal's global downfall but another #energy tech decline is loomingRead my new
Read more