Pages

Tuesday, March 24, 2020

Spreading fear, intentionally or unintentionally

Is there really a plague upon us all?
Sometimes, even the best-intentioned sources can give off the wrong messages. Look at this map from the Johns Hopkins site that is tracking “Coronavirus COVID-19 Global Cases”.

From the look of it, it appears as if there is “a plague upon us all”. Certainly uncontrolled outbreaks of novel (new) viruses can get that way, and they have, historically, in real life, been incredibly damaging, and incredibly frightening.

But some charts and maps (like this one) aren’t helping us really to understand what’s going on, and where we are in real life.I think that the way the graphics are set up, they are more inclined to spread fear, than to explain things accurately.

It’s true that we don’t yet have all the best numbers, because not everyone’s testing is either accurate or adequate. I trust that’s a situation that is correcting itself as things move forward. In some cases, such as South Korea, we do have that kind of numbers. In other places, maybe not.

But as better information becomes distributed, we can expect that we’ll be able to see more clearly those locations where the curve is flattening, and where it is not, and the things they are doing to help things along. The video here, though it’s a few weeks old, will give you an idea of what to look for.

Exponential charts can’t work in real life.

Here’s a video that shows why that’s so, and when and how to expect and understand when things begin “flattening”.



How Exponential Growth Charts Work

31 comments:

  1. Of course, the equal and opposite error is to downplay, soften, or whitewash the facts, intentionally or unintentionally.

    ReplyDelete
    Replies
    1. In that vein, it's nice to see you accept correction from other commenters that your previous post was based on incorrect facts because it had originally stated most states (except NY) didn't have any new coronavirus cases when that was mistaken.

      Delete
    2. I will always incorporate new and better information when it’s presented.

      😉

      Delete
    3. "I will always incorporate new and better information when it’s presented."

      In that case:

      1. You may wish to correct the last paragraph in this post because it's likewise based on the incorrect facts presented in your previous post.

      2. I think you might be reading too much into the graphics - i.e. the size or distrubtion of red circles - on the Johns Hopkins website. I never bother looking at the size or distribution of the red circles. Instead I look at the numbers.

      3. I think you might be expecting too much from the Johns Hopkins website. I think it's generally reliable, but it's not as up-to-date as it should be. It'd be more up-to-date to look at each nation's own statistics. However that would be time-consuming if one did it on a routine basis. Hence compendiums like the Johns Hopkins website exist as reliable approximations.

      4. By the way, if anyone wishes to see the data repository for the Johns Hopkins website, they've made it available on GitHub.

      5. Actually, I think the YouTube video you link to could both argue for as well as against what you're attempting to show. What would make the difference (or at least a difference) is the data that's fed into the model.

      Delete
    4. I think you might be reading too much into the graphics - i.e. the size or distrubtion of red circles - on the Johns Hopkins website. I never bother looking at the size or distribution of the red circles. Instead I look at the numbers.

      Well, maybe not everyone's as sophisticated as you are? As I mentioned elsewhere, sometimes it's "how you feel", and that's caused by impressions often caused by shiny objects, and not necessarily a reasoned understanding of red-circle usage on major medical websites.

      In addition, I think that the Worldometer website is trying to fill the gaps that the Johns Hopkins site apparently has with respect to US state data. It updates based on official state health department sites, but it also doesn't do so very quickly.

      Delete
    5. It has nothing to do with sophistication. Rather it's simply about looking at numbers. Such as how many total cases there are or how many deaths there are. That hardly requires sophistication. Besides, if you can understand Worldometer, you can also understand Johns Hopkins. They're both essentially presenting the same information.

      Delete
    6. Good to see you remove the last paragraph in this post.

      Delete
    7. Seriously, you look at numbers. I would guess that very few people who are looking for info will dive into the numbers as you would. They'll look at the map, maybe click around on a few of the numbers, get lost, and move on to the next thing.

      This is why companies will spend millions of dollars for a 30 second TV commercial during the Super Bowl. Because people are coming and going, not looking at and comparing the nutritional value of the various beers that are being advertised. It's "impressions", and even "first impressions", and how they leave people feeling. No analysis at all. That's true for millions, tens of millions of people who watch those things.

      With that said, I've never seen the state data on the John's Hopkins site -- there is the odd state or even metro area that shows up. But not in a chart, with all the statistics from a particular state, in a nice chart like the Worldometer site. I'd be interested if you could say how to get to that information.

      Delete
    8. The state data is there. Click on US and then go to the Admin 2 tab at the bottom to see the individual state data. The 3rd tab Admin 3 has the county data

      Also it gives you an incorrect impression when you are zoomed out. Zoom in to get a more accurate portrayal. The distances and relative sizes of the data come into focus.

      Delete
    9. "Seriously, you look at numbers. I would guess that very few people who are looking for info will dive into the numbers as you would. They'll look at the map, maybe click around on a few of the numbers, get lost, and move on to the next thing. This is why companies will spend millions of dollars for a 30 second TV commercial during the Super Bowl. Because people are coming and going, not looking at and comparing the nutritional value of the various beers that are being advertised. It's "impressions", and even "first impressions", and how they leave people feeling. No analysis at all. That's true for millions, tens of millions of people who watch those things."

      1. I think you underestimate millions of people.

      2. I think most people know how to distinguish between how they felt during a Super Bowl commercial vs. the number of cases in a pandemic. A person can be moved to tears by a commercial or a movie, but that doesn't they can't count numbers.

      3. Again, if you can understand the "numbers" on Worldometer, you can understand the "numbers" on Johns Hopkins.

      4. Otherwise I don't really know how to reply to people who are inordinately impressed with "impressions often caused by shiny objects" and "how they leave people feeling", but puzzled by basic numbers and basic counting.

      "With that said, I've never seen the state data on the John's Hopkins site -- there is the odd state or even metro area that shows up. But not in a chart, with all the statistics from a particular state, in a nice chart like the Worldometer site. I'd be interested if you could say how to get to that information."

      1. ScottO already answered your question. I tried what he said and it worked like he said. I'll add:

      2. I think this goes back to my earlier point that you might be expecting the Johns Hopkins website to be something it wasn't meant to be.

      3. You don't have to pick one website over another website. You could use both. Or three or four or five or whatever. For example, Johns Hopkins, Worldometer, the WHO, the University of Washington, etc.

      4. If you prefer the Worldometer site, then use that instead.

      5. I already gave you a link to Johns Hopkins' data repository at GitHub above.

      Delete
  2. The state data is there. Click on US and then go to the Admin 2 tab at the bottom to see the individual state data. The 3rd tab Admin 3 has the county data

    ReplyDelete
    Replies
    1. ScottO, it only gives the number of confirmed cases. Take a look at the Worldometer site now. It is really filling in, with a chart full of statistics that are easily comparable:

      https://www.worldometers.info/coronavirus/

      Delete
    2. Yes it appears they are updating a bit cleaner

      Delete
  3. Data is simply data. The underlying dispositions people have toward the data and what they do with it ultimately says more about the interpreters than anything else.

    The same concept loosely applies to lots of things actually, money for example.

    ReplyDelete
    Replies
    1. ///what they do with it ultimately says more about the interpreters than anything else///

      Of course this is true. but there are always multiple layers of "interpretation", and the initial "interpreters" of the data (perhaps falsely assuming that everyone is both technically adept and sufficiently analytical) have created a website, for which one's first impression, is a global map that is practically inundated with big bright red circles that connote "alarm" in our culture.

      This blog post is more about the method of communication, ("interpretation"), than anything else. For people who want to communicate "what is actually happening" in the world, the first impression given by the global map at the Johns Hopkins site is not one of hope, (and that may not be its intention), but one rather of alarm.

      Delete
  4. This blog is saying that looking at red dots is spreading fear???

    ReplyDelete
    Replies
    1. It's about first impressions, and the shallowness of people in general, who are moved emotionally by shiny things. You look at the map as I've posted it, and it looks as if the whole world is being immolated. True, some people will dig deeper and find out that's not so. But for others, who see that the Johns Hopkins site is probably one of the best for understanding the scope of the pandemic, one could come away with the notion that the pandemic is worse than it really is.

      Delete
  5. Bro you are getting ALOT of mileage out of how alarming red dots are supposed to be. Is Johns Hopkins supposed to use happy yellow smiley faces instead?

    ReplyDelete
    Replies
    1. Johns Hopkins is not the only example. It's one that's been recommended, but it's much used. And it's much used not only by analytical people (of whom there are a lot), but also of non-analytical (maybe "right-brained" people). And what I'm saying is to illustrate a principle of how communications works, and not specifically to write about Johns Hopkins. Johns Hopkins illustrates a broader principle.

      Delete
    2. 1. I think John's basic "principle" is some people are more persuaded by gut feelings and impressions than reason and evidence.

      2. Of course, that's hard to deny. For example, I'm sure there are people who have become progressives or socialists primarily due to their gut feelings or impressions that progressive or socialist values are good for society.

      3. At best, one might have a gut feeling about a good cause. However, even if so, the problem is this could have simply been due to sheer luck. That's because gut feelings aren't reliable. They could be right, they could be wrong. Who knows?

      4. In other words, gut feelings in and of themselves aren't necessarily reliably aimed at truth. Rather gut feelings need to be anchored in truth. And to be anchored in truth requires reason and evidence.

      Delete
    3. Hawk, you continue to miss that I'm talking about the communication of a thing. But I expect that from you now, after our recent exchanges. For some reason, I'd think this is a concept that a smart guy like you could understand, but maybe not. (Please accept this friendly correction in the spirit in which it's intended).

      There are effective ways to communicate, which are clear and which convey what the communicator wants to convey, and such things as this map, which very initially sends the wrong message.

      Delete
    4. 1. The information on the Johns Hopkins website is quite clear for anyone who doesn't simply look at the "red circles" and becomes "alarmed" by looking at the red circles.

      2. I was attempting to put the best possible spin on your argument. To steelman your argument, as it were. However, if that's not what you're arguing, then feel free to offer a different and better argument.

      3. And yes we have gone back and forth in debates, both here and elsewhere, but I believe I've done my level best to be fair and reasonable with you. In fact, anyone can read our debate in this very post of yours and decide for themselves.

      Delete
    5. By the way, I'm hardly the only person even in this post who has disagreed with you. There are others here who seem to disagree with you.

      Same with your previous post in which I didn't even comment. Rather other commenters commented to voice their disagreement (though you eventually closed the comments on the post).

      And I've disagreed with other Tbloggers. Such as Alan over his position on vaccines and vaccination.

      Delete
    6. I understand people disagreeing. You can disagree with Alan over medical issues and that's valid because you are a medical specialist. You are not, however, a communications and media specialist, and as this is a matter of opinion (whether dots cause people to fear or not), your opinion carries relatively less weight.

      Delete
    7. 1. Sure, that's fair. No one has to accept my "opinion" on this.

      2. I never made an argument from authority in this post or with Alan.

      3. It'd be nice to have a good study on whether "red circles" "cause people to fear or not".

      4. I think I'm more or less saying the same things the other commenters in this post are saying.

      5. I wouldn't consider myself a medical specialist. I'm a nobody. I prefer to let my arguments stand or fall on their own.

      Delete
    8. Aside from yourself, there really only was one person objecting on this comment thread. ScottO was telling me how to find state and county data -- which was there, but it wasn't organized well, and Coram Deo merely made a statement about data, and I responded (seemingly) to his satisfaction. The only person who had the least objection to it was DQ, who really didn't offer a positive alternative.

      If I were to offer a positive alternative, I would point to the Worldometer site. I liked the look of it initially, although at first it wasn't easy to determine how and when they were updating. The comments in that thread forced me to look more deeply into its functionality.

      As a tool for communication, it is much better than the Johns Hopkins site. There is no map, which is virtually meaningless in its context anyway. Plus, all of the "cases" and "new cases" and "deaths" and "new deaths" and "recovered" and "active cases" are all aligned by country, in a chart format.

      Where there are graphics, curves, they are easily toggled between linear scale and logarithmic scale, so the viewer can look at each country's figures and easily see where the inflection point is, and which way the virus is trending.

      All of that seems much more useful, and much less fear inspiring, than the big global map that Johns Hopkins is featuring.

      By the way, I thought I've seen it mentioned here that you have a medical background. If I got that wrong, I apologize.

      Delete
    9. 1. ScottO, Coram Deo, and DQ can speak for themselves if they wish, but I'll simply say I read them differently from how you read them.

      2. I've recommended Worldometer as well. For example, see my post "Which medical experts do I follow?".

      3. I imagine what's a good or bad "tool for communication" is partly person-dependent.

      4. For example, one person might not like the Johns Hopkins website because the "big bright red circles" somehow "alarm" him, while another person doesn't think much of that at all.

      5. No, I've never mentioned my background here. I've intentionally avoided doing so.

      Delete
    10. I guess there are at least two types of people in this world:

      1. The first type is the person who can look at "red circles" on a website and realize they're just symbols or representions. Just computer generated images.

      2. The second type is the person who looks at the same "red circles" on a website but thinks Armageddon is upon them.

      Delete
  6. The visual is somewhat ghastly, no offense. However its not the JHU meds who have created it. It is likely some software guys. The JHU docs researching this most likely only provide data to the coders. Yes, they can talk disease origins, symptoms, the normal curve, spread, etc., however they likely only provided the data to some code warrior, who then created the grim website + visuals. IT people often are not the greatest at aesthetics.

    Said U is down the road from me although the Bloomberg School of Health is in downtown. I think a doc from my church may be researching the data behind the visual up above. I can't confirm since we have not met for a couple of weeks. I point this out because this chap* could tell you about all sorts of stuff about Corona, however he would not know all that much about html, CSS, whether to construct heatmaps, line maps, etc. the stuff that goes into building the site up above.

    ~ In Him,
    Raj
    *If you can, squeeze in a prayer for him, since he is concerned about family and friends, esp. the elderly back home in India.

    ReplyDelete
    Replies
    1. Thanks, rgbrao.

      1. Good point. It's true IT guys aren't the greatest when it comes to designing visually appealing graphics. And I agree the Johns Hopkins' map is "ghastly" from an aesthetic perspective.

      2. However, if I can say, I don't think that was the issue. Rather I think the post is arguing there are some people who look at "big bright red circles" on the map and become "fearful" simply by seeing "red circles" on the map. Sort of like how a bull becomes angry when it sees red, these people become "scared" when they see red circles. This is turn is because there are some people who are moved by "impressions" and "feelings" so when they see "red circles" they become "alarmed". So the "red circles" on the map are causing people to be "scared" of the coronavirus.

      3. In fact, I don't disagree there are such people. There are people who do rely on their "feelings" or "impressions" or gut feelings. My point was that these sorts of people need to look beyond the "red circles".

      4. I think the actual numbers are "alarming" or at least concerning to some degree, though John doesn't think the actual numbers are "alarming". He thinks it's not all that bad if we look at the numbers. That's why he's arguing the big red circles are alarming people, but if we look at the numbers, then they won't alarm people. But I think either way you look at it, it's "alarming" or at least concerning to some degree.

      5. However I don't think we should panic. Rather I think we should act reasonably. Like it sounds like the doctor at your church is doing.

      6. By the way, I mentioned above that the data repository for the Johns Hopkins website is available on GitHub. Perhaps that would be of interest to your physician friend.

      Delete