I'll be teaching a course in the fall on data communication.
-
@chu if you are interested in #OpenData you may want to consider the following, although it is quite UK specific:
The transport energy and environment data tables: https://www.gov.uk/government/statistical-data-sets/energy-and-environment-data-tables-env
UK National Atmospheric Emissions Inventory (NAEI) showing various emissions https://naei.energysecurity.gov.uk(There is a bigger list of open data collated here https://anisotropi4.github.io/shed/opendata.html
It is less relevant to the ask but is stuff I have used in talks or visualisations.) -
@chu My all-time favorite for this is "How to Lie with Statistics" https://dn710604.ca.archive.org/0/items/HowToLieWithStatistics_201608/How-to-Lie-With-Statistics-1954-Huff_text.pdf
@mdwaroff @chu I love How to Lie with Maps (title is a tribute to How To Lie with Statistics) https://en.wikipedia.org/wiki/How_to_Lie_with_Maps
-
@mdwaroff @chu I love How to Lie with Maps (title is a tribute to How To Lie with Statistics) https://en.wikipedia.org/wiki/How_to_Lie_with_Maps
-
@chu yes. There is the EU Global Human Settlement Layer (GHSL) @CopernicusECMWF data here https://human-settlement.emergency.copernicus.eu/GHSLDataTools.php
I've used urban classification data and note "Climate Change" data https://climate.copernicus.eu/ but as I have never used it, I'm not sure how much use this is.There is also @WorldPopProject
data https://hub.worldpop.org/. They have population and Global flight data https://hub.worldpop.org/project/categories?id=13(Noting I'm a dilettante with a focus on European public transport and heavy rail...)
-
@chu yes. There is the EU Global Human Settlement Layer (GHSL) @CopernicusECMWF data here https://human-settlement.emergency.copernicus.eu/GHSLDataTools.php
I've used urban classification data and note "Climate Change" data https://climate.copernicus.eu/ but as I have never used it, I'm not sure how much use this is.There is also @WorldPopProject
data https://hub.worldpop.org/. They have population and Global flight data https://hub.worldpop.org/project/categories?id=13(Noting I'm a dilettante with a focus on European public transport and heavy rail...)
@chu (this then off the intent of your question but there is then stuff like the @openstreetmap which also provides an interesting map resource for features and names https://www.openstreetmap.org/#map=6/47.42/20.92.
Just to get a plug in here, here are some examples of #DataVisualisation from the #30DayMapChallenge which shows what you can draw with this date, if nothing else
https://anisotropi4.github.io/shed/30daymapchallenge.html) -
@chu yes. There is the EU Global Human Settlement Layer (GHSL) @CopernicusECMWF data here https://human-settlement.emergency.copernicus.eu/GHSLDataTools.php
I've used urban classification data and note "Climate Change" data https://climate.copernicus.eu/ but as I have never used it, I'm not sure how much use this is.There is also @WorldPopProject
data https://hub.worldpop.org/. They have population and Global flight data https://hub.worldpop.org/project/categories?id=13(Noting I'm a dilettante with a focus on European public transport and heavy rail...)
@wnd @CopernicusECMWF @WorldPopProject
Thanks.
Will go through all these resources as I really start to put the course together over the summer. Appreciate it
-
@chu — Supreme Court decision rates. You can make it appear the Court is hyperpartisan or hypercollegial depending entirely on how you define a Supreme Court case.
The majority of SCOTUS cases are disposed of via GVR. In a single order they (G)rant the request for SCOTUS review, (V)acate the lower court decision, and (R)emand to the lower court for further hearings — usually with a Post-It attached explaining to the lower court judge where they screwed up and what they should've done instead.
GVRs represent the majority of their output and are overwhelmingly (>90%) 7-2, 8-1, or 9-0.
Cases that get invited for full argument before the Court represent the smallest chunk of their output, and tend to break down 5-4 or 6-3 along predictable fault lines.
Very few national pundits nattering about the "factionalism" of SCOTUS are ever honest enough to talk about the GVR rate, or other measures which do not support the factionalization narrative.
@rob @chu The current hyperpartisanism of the US Supreme Court has nothing to do with how votes are split but the policies they allow. From recent memories, they’ve allowed post-hoc political corruption (“it’s not a bribe anymore after the vote”), erosion of women’s reproductive right by repealing Roe vs Wade and enabled political segregation in Louisiana. None of the vote splits on these cases will speak louder than the political and social consequences of these decisions. -
@chu (this then off the intent of your question but there is then stuff like the @openstreetmap which also provides an interesting map resource for features and names https://www.openstreetmap.org/#map=6/47.42/20.92.
Just to get a plug in here, here are some examples of #DataVisualisation from the #30DayMapChallenge which shows what you can draw with this date, if nothing else
https://anisotropi4.github.io/shed/30daymapchallenge.html) -
I'll be teaching a course in the fall on data communication.
One of the assignments I hope to put together is a lesson on how data is manipulated. I want to show how easy it is for climate change deniers, anti vaxxers, etc to crop data, stretch or flip an axis and suggest the opposite of what the data is actually showing. Still thinking through the assignment and I'm thinking of having them make an honest representation and one less so.
I think there's value to such a lesson given how much downright lying we have from not just randos but even political circles these days.
Was just going to use publicly available data sources but then I am thinking that there must be researchers here who have awesome data they wouldn't mind seeing put into visual form. If you do have data you'd be willing to let me use, please drop me a comment or PM and let me know how to access it. Thanks!
(P.S. would appreciate a share for wider reach)
@chu Will you also explain the core idea of the Scientific Method, why it was a breakthrough, and how delicate (even while also powerful) it is even by itself, in context of the "replication crisis" and various perverse incentives ("publish or perish", pressure for positive results, non-double-blind results, poor/manipulated statistics, publishers monopolies, overloaded and underpaid reviewers)? To me realizing and understanding this (taught on a course during my studies) was a major worldview shock.
Also the classic "scientific news progression" from a paper saying foo and bar may be correlated through univ. press saying "our scientist says foo could be a cause for bar" to popular press "Uni of Herethere scientists say foo causes bar".
A few communications thing that come to my mind in photography and video: One, using even the same photo or film but cropping it differently can give very different impression (e.g. person A hits person B, vs. a longer footage when person B first heavily abuses, or even hits, person A). Or how to make a photo of whatever number of people look like a crowd by lowering camera position and tightening the frame (I cannot unsee it since I learned about it). Or, lying by showing different photos claiming they're from elsewhere (difficult to check if you're not an expert - show any hurt people and they can be anywhere in the world). Or how a neutral-expression face looks either angry or sad depending on how you primed the watcher by narrative.
Anyway, you're doing an amazing thing for those students and for the better world, thank you!
-
@chu Will you also explain the core idea of the Scientific Method, why it was a breakthrough, and how delicate (even while also powerful) it is even by itself, in context of the "replication crisis" and various perverse incentives ("publish or perish", pressure for positive results, non-double-blind results, poor/manipulated statistics, publishers monopolies, overloaded and underpaid reviewers)? To me realizing and understanding this (taught on a course during my studies) was a major worldview shock.
Also the classic "scientific news progression" from a paper saying foo and bar may be correlated through univ. press saying "our scientist says foo could be a cause for bar" to popular press "Uni of Herethere scientists say foo causes bar".
A few communications thing that come to my mind in photography and video: One, using even the same photo or film but cropping it differently can give very different impression (e.g. person A hits person B, vs. a longer footage when person B first heavily abuses, or even hits, person A). Or how to make a photo of whatever number of people look like a crowd by lowering camera position and tightening the frame (I cannot unsee it since I learned about it). Or, lying by showing different photos claiming they're from elsewhere (difficult to check if you're not an expert - show any hurt people and they can be anywhere in the world). Or how a neutral-expression face looks either angry or sad depending on how you primed the watcher by narrative.
Anyway, you're doing an amazing thing for those students and for the better world, thank you!
Great ideas. I honestly haven't thought it all through yet. I'll get working on it over the summer so may hit you up in July and August when I'm ready to piece it all together.
Thanks
-
I'll be teaching a course in the fall on data communication.
One of the assignments I hope to put together is a lesson on how data is manipulated. I want to show how easy it is for climate change deniers, anti vaxxers, etc to crop data, stretch or flip an axis and suggest the opposite of what the data is actually showing. Still thinking through the assignment and I'm thinking of having them make an honest representation and one less so.
I think there's value to such a lesson given how much downright lying we have from not just randos but even political circles these days.
Was just going to use publicly available data sources but then I am thinking that there must be researchers here who have awesome data they wouldn't mind seeing put into visual form. If you do have data you'd be willing to let me use, please drop me a comment or PM and let me know how to access it. Thanks!
(P.S. would appreciate a share for wider reach)
@chu I presume you know the work of Tufte who does some really good examples of manipulated data presentation. I mean, he is a bit dated now, as a lot of the examples are from soviet propaganda, but the idea is still the same.
-
@chu I presume you know the work of Tufte who does some really good examples of manipulated data presentation. I mean, he is a bit dated now, as a lot of the examples are from soviet propaganda, but the idea is still the same.
I don't. I'll look it up. Thanks
-
Fair. Don't worry. It'll go back to zero.
I want to show how easy it is for climate change deniers, anti vaxxers, etc to crop data, stretch or flip an axis and suggest the opposite of what the data is actually showing.
Ironic how you're talking about educating others about questioning misrepresented data, then outspokenly misrepresenting data in the next breath.
I'm not arguing for bitcoin -- it's an absolute disaster of volatility -- but you can't look at the long-term data and deny that its long term trajectory is up, even if you only measure the lows. Also, it's history it littered with detractors who bet against it by shorting it on the open markets -- and then got destroyed when it suddenly increased.
I'm not ignoring prior history - tulips, beany babies, internet companies circa 1999... Maybe you need to add a section at the end on how data can be good, but an idea can still be bad... But I hope it doesn't undermine the entirety of the rest of your message.
-
R relay@relay.mycrowd.ca shared this topic
-
Thanks. Will look at this. My first time teaching this particular course so all resource suggestions are welcome
@chu @a_cubed @troublewithwords
vote against Tufte; I know I am wierd but IMO he is a blowhard
much better to use the quiet lady who gets no credit:
https://www.amazon.com/Creating-Effective-Graphs-Naomi-Robbins/dp/0985911123 -
I want to show how easy it is for climate change deniers, anti vaxxers, etc to crop data, stretch or flip an axis and suggest the opposite of what the data is actually showing.
Ironic how you're talking about educating others about questioning misrepresented data, then outspokenly misrepresenting data in the next breath.
I'm not arguing for bitcoin -- it's an absolute disaster of volatility -- but you can't look at the long-term data and deny that its long term trajectory is up, even if you only measure the lows. Also, it's history it littered with detractors who bet against it by shorting it on the open markets -- and then got destroyed when it suddenly increased.
I'm not ignoring prior history - tulips, beany babies, internet companies circa 1999... Maybe you need to add a section at the end on how data can be good, but an idea can still be bad... But I hope it doesn't undermine the entirety of the rest of your message.
I'm not looking at the graph of Bitcoin. I am talking about it's long term theoretical value.
Yes. It's trajectory has been up. But think about how it operates. It has no intrinsic value and its cost of maintenance over time is prohibitive.
At the end of the day I am first and foremost an energy person. I used to work on energy networks as a consulting engineer.
When the electricity infrastructure fails to provide enough for the next Bitcoin unit, by all accounts it will trend to zero. That day will come. Just a matter of when
-
R relay@relay.infosec.exchange shared this topic
