Generally, having the Y axis start at 0 gives a better sense of the magnitude of a change relative to the previous value.
The graph you showed isn’t dishonest, but at a glance doesn’t give an intuitive understanding of the how significant the increase is. It would be nice if the graph provided a visual answer to “what is the percentage increase vs pre covid?” As it stands, I need to do the math myself :(
E.g. if the exact same graph had the Y axis from 5,000 to 5,100 (instead of 5,000 to 8,500), it would still show an increase, but a significantly less severe one. By starting the Y axis at 0, the change is put into context much more intuitively.
The graph you showed isn’t dishonest, but at a glance doesn’t give an intuitive understanding of the how significant the increase is.
That’s because that’s not what the graph is originally for. The graph is being repurposed from Labor and Census statistics unrelated to covid.
Also, I really don’t agree with you on starting with zero. That’s exactly how I would present it if I was trying to downplay it. I would take advantage of the fact that we live in a eugenic social murdering society that doesn’t give a shit about 30 million disabled and would swamp that additional 5 million in a tiny little uptick at the top right corner to let people continue not giving a shit.
edit: Also, wtf. How is the percentage change relevant to the argument being made? is an additional 50 million disabled since 2020 okay if it’s only an increase of .2%?
Woah there, I’m just trying to answer your question, not attack you or try to downplay the severity of the situation. It’s strictly about how data is presented.
Graphs with a Y axis that don’t start at zero is a common gripe for anyone presenting any graph, not just this.
The reason it is done is to make changes look more dramatic (stock market CRASHES with a graph showing a huge-looking decrease… of 0.5%).
Wanting graphs to start at zero is just about trying to remove spin from data presentation.
Except the zero is irrelevant because the purpose of both graphs is to display a shift in the ongoing trend. Have you even visualized what the graph you’re asking for what would look like? Literally 80% whitespace.
Because you’ve ignored every criticism of why that preference is irrelevant to conversation at hand. I did not make my original comment with the intention of debating graph aesthetics, I did so to remind people that they are taking part in a mass disabling event by people who are openly trading in eugenics rhetoric. Much of that rhetoric being directed towards the people of this very community, which is largely not showing solidarity with the other people who are victims of said rhetoric and their associated policies.
I can definitely acknowledge that graph aesthetics aren’t as important as the real-world implications of the underlying data. I did not start the thread about graph aesthetics, though.
You asked miz about their graph aesthetic preference, and I answered bc it’s a common preference. I assumed that you were asking because you wanted to learn, and I answered in good faith. If you didn’t want to talk about it, just don’t ask questions about it?
I ignored your criticisms because I don’t really care about other people’s graph aesthetic preferences, because it’s not important and they’re my preferences.
If you had said “what flavor ice cream do you prefer?” and I said “many people prefer chocolate because <whatever>”, it would weird if you got upset and said “actually vanilla is better because <whatever> and also why are we even talking about ice cream preferences it’s not important”.
At any point, you could have said “I think this representation is appropriate, given the severity of the situation, but I understand the other perspective.” and that would have been the end of the conversation.
Generally, having the Y axis start at 0 gives a better sense of the magnitude of a change relative to the previous value.
The graph you showed isn’t dishonest, but at a glance doesn’t give an intuitive understanding of the how significant the increase is. It would be nice if the graph provided a visual answer to “what is the percentage increase vs pre covid?” As it stands, I need to do the math myself :(
E.g. if the exact same graph had the Y axis from 5,000 to 5,100 (instead of 5,000 to 8,500), it would still show an increase, but a significantly less severe one. By starting the Y axis at 0, the change is put into context much more intuitively.
That’s because that’s not what the graph is originally for. The graph is being repurposed from Labor and Census statistics unrelated to covid.
Also, I really don’t agree with you on starting with zero. That’s exactly how I would present it if I was trying to downplay it. I would take advantage of the fact that we live in a eugenic social murdering society that doesn’t give a shit about 30 million disabled and would swamp that additional 5 million in a tiny little uptick at the top right corner to let people continue not giving a shit.
edit: Also, wtf. How is the percentage change relevant to the argument being made? is an additional 50 million disabled since 2020 okay if it’s only an increase of .2%?
Woah there, I’m just trying to answer your question, not attack you or try to downplay the severity of the situation. It’s strictly about how data is presented.
Graphs with a Y axis that don’t start at zero is a common gripe for anyone presenting any graph, not just this.
The reason it is done is to make changes look more dramatic (stock market CRASHES with a graph showing a huge-looking decrease… of 0.5%).
Wanting graphs to start at zero is just about trying to remove spin from data presentation.
Except the zero is irrelevant because the purpose of both graphs is to display a shift in the ongoing trend. Have you even visualized what the graph you’re asking for what would look like? Literally 80% whitespace.
Correct. Some folks, myself and miz included, would prefer that.
Clearly, we just had two different interactions.
My perspective is that you asked a question (how would you prefer it?) and I answered your question.
It seems like you didn’t like that. Why?
Because you’ve ignored every criticism of why that preference is irrelevant to conversation at hand. I did not make my original comment with the intention of debating graph aesthetics, I did so to remind people that they are taking part in a mass disabling event by people who are openly trading in eugenics rhetoric. Much of that rhetoric being directed towards the people of this very community, which is largely not showing solidarity with the other people who are victims of said rhetoric and their associated policies.
I can definitely acknowledge that graph aesthetics aren’t as important as the real-world implications of the underlying data. I did not start the thread about graph aesthetics, though.
You asked miz about their graph aesthetic preference, and I answered bc it’s a common preference. I assumed that you were asking because you wanted to learn, and I answered in good faith. If you didn’t want to talk about it, just don’t ask questions about it?
I ignored your criticisms because I don’t really care about other people’s graph aesthetic preferences, because it’s not important and they’re my preferences.
If you had said “what flavor ice cream do you prefer?” and I said “many people prefer chocolate because <whatever>”, it would weird if you got upset and said “actually vanilla is better because <whatever> and also why are we even talking about ice cream preferences it’s not important”.
At any point, you could have said “I think this representation is appropriate, given the severity of the situation, but I understand the other perspective.” and that would have been the end of the conversation.