4 ms·
Now if only we could get the media to report on useful statistics, rather than total number of cases between geographic areas that differ widely in size and pop
by kbenson 6y ago
Now if only we could get the media to report on useful statistics, rather than total number of cases between geographic areas that differ widely in size and population.
California's not even in the top 20% for new cases over the last 7 days per capita, and is barely in the top 50% overall per capita, but we have the most individual cases overall because we have such a large population. That it gets reported along with Florida every time there's a "new record" broken in new cases (gripe number two, for a similar reason) is just the media trying to make the headlines seem as relevant as possibly to the largest audience, even if it's more fear-mongering than useful.
- refurb 6y agoThis annoys me to no end as well. In my local area new cases are quite low, maybe 10 new cases in a population of 750,000. Yesterday they mentioned a record number of new cases and a “potential hotspot”. The cases went up to 12.
- war1025 6y ago> Now if only we could get the media to report on useful statistics The website for my state [1] has a map with infection counts by county. The irritating thing about this map is it's total numbers since February. That tells me nothing useful in terms of how places are currently doing. You can click on individual counties and it'll give you the 14 day rolling total, which is pretty useful for seeing upticks and improvements. But for whatever reason, we've decided that "number of people who ever got sick" is the benchmark we need to focus on. It seems completely unhelpful. [1] https://coronavirus.iowa.gov/pages/case-counts https://coronavirus.iowa.gov/pages/case-counts
- dredmorbius 6y agoEven Wikipedia only gives cumulative counts for Iowa :( https://en.wikipedia.org/wiki/COVID-19_pandemic_in_Iowa https://en.wikipedia.org/wiki/COVID-19_pandemic_in_Iowa
- war1025 6y agoI like how the entire map on that page is dark red. Somehow 0-1% seems like too low a range for the scale there. Maybe just needs to be rebalanced by someone.
- thephyber 6y agoI hate to always blame the media because it's a chicken or egg problem. No news outlet wants to be the one to lose business because they increase the friction of their news consumption, so we all suffer from Tragedy of the Commons. Sad truth: news readers don't read the entire article. News agencies craft news in the format that readers are willing to digest at scale (a very brave few stayed in long form publications, but they don't report daily). The important details that are relevant to you are nuanced and aren't subject to easy headlines and even if they are, the articles don't get clicks (for better or worse). They are in the body of many articles from the better outlets, but usually the quality content is behind paywalls. Until we find a better news business model that doesn't incentivize news to optimize for clickbait and social media to optimize news feeds for maximum emotion, we are going to continue to suffer from Tragedy of the Commons in this space.
- mistermann 6y agoThe CDC could publish a single, authoritative dataset, accompanied by spreadsheet-like UI's, sensible and informative charts, with useful filtering. I would think that would be part of their mandate, to inform the public.
- thephyber 6y agoThe CDC already does it.[1] It's probably worth a discussion about why we weren't all exposed to it. Interesting related note: the CDC was pretty inventive with their marketing back in ~2011 when they created a Zombie Preparedness[2] campaign. [1] https://www.cdc.gov/coronavirus/2019-ncov/communication/graphics-buttons.html https://www.cdc.gov/coronavirus/2019-ncov/communication/grap... [2] https://www.cdc.gov/cpr/zombie/index.htm https://www.cdc.gov/cpr/zombie/index.htm
- mistermann 6y agoThose are just a few graphics, I'm wanting comprehensive and reasonably trustworthy data, and a variety of useful visualizations of the data. It's surreal to see the sophistication of this sort of thing that exists in private industry, compared to the dearth of comparable platforms in the government sector, which we're told is populated by highly skilled experts who should be listened to. The sales pitch certainly seems to have worked like a charm, but how completely escapes me.
- dredmorbius 6y agoWorldometers have been pretty good in providing if not focusing on many of these values. You may have to dig into the display columns selector. Higher-level (country/world) reporting tends to be better and more consistent than regional local, as with most resources. "Cases / deaths / tests per X people" have been added in the past week or so: https://www.worldometers.info/coronavirus/#countries https://www.worldometers.info/coronavirus/#countries FT's coronavirus page shows excellent country-level detail with recent trending. As is often the case, business press is better than general news organisations at quantitative presentation, and the small-multiples charts are particularly effective: https://ft.com/coronavirus-latest https://ft.com/coronavirus-latest As is almost always the case for complex developing stories, Wikipedia exceeds virtually all conventional news organisations in synthesizing current knowledge. The COVID-19 / SARS-COV-2 pages have become a cluster of excellent information, from overview to detail: https://en.wikipedia.org/wiki/COVID-19_pandemic https://en.wikipedia.org/wiki/COVID-19_pandemic Organisation has evolved as well, with per-country stats now having its own page: https://en.m.wikipedia.org/wiki/COVID-19_pandemic_by_country_and_territory https://en.m.wikipedia.org/wiki/COVID-19_pandemic_by_country... For the US, the page now begins with a county-level map showing new cases per 100k population over the preceding 14 days (preseently highly focused on southern states): https://en.wikipedia.org/wiki/COVID-19_pandemic_in_the_United_States https://en.wikipedia.org/wiki/COVID-19_pandemic_in_the_Unite... State/local pages vary in quality, e.g., Illinois similarly reports prior two weeks cases per 100k, South Carolina only total infections per 100k, again, by county for each: https://en.wikipedia.org/wiki/COVID-19_pandemic_in_South_Carolina https://en.wikipedia.org/wiki/COVID-19_pandemic_in_South_Car... https://en.wikipedia.org/wiki/COVID-19_pandemic_in_Illinois https://en.wikipedia.org/wiki/COVID-19_pandemic_in_Illinois There is also realtime rt values by US state: https://rt.live https://rt.live And projections: https://covid19.healthdata.org/projections https://covid19.healthdata.org/projections Few news organisations can or will provide useful quantitative or graphical information. Fortunately there are other sources.
- dredmorbius 6y agoA more complete set of COVID-19 informational resources†: COVID-19 trends by region, country, state/province, U.S. metropolitan area, or U.S. county: https://outbreak.info https://outbreak.info COVID-19 projections: https://covid19.healthdata.org/projections https://covid19.healthdata.org/projections Financial Times Covid-19 tracker: https://ig.ft.com/coronavirus-chart/ https://ig.ft.com/coronavirus-chart/ CovidTrends; New cases (weekly average) vs. total cases, by country: https://aatishb.com/covidtrends/ https://aatishb.com/covidtrends/ Worldometers Coronavirus live statistics: https://www.worldometers.info/coronavirus/ https://www.worldometers.info/coronavirus/ US realtime state Rt estimates and tracker: https://rt.live https://rt.live The Economist excess mortality tracker: https://www.economist.com/graphic-detail/2020/04/16/tracking-covid-19-excess-deaths-across-countries https://www.economist.com/graphic-detail/2020/04/16/tracking... Covid19 projections: https://covid19-projections.com/ https://covid19-projections.com/ Imperial College London Covid tracking 7 estimation model (US): https://mrc-ide.github.io/covid19usa/ https://mrc-ide.github.io/covid19usa/ Our World in Data Coronavirus tracker: https://ourworldindata.org/coronavirus https://ourworldindata.org/coronavirus Johns Hopkins University Coronavirus map (US): https://coronavirus.jhu.edu/us-map https://coronavirus.jhu.edu/us-map ArcGIS visualizer: https://experience.arcgis.com/experience/7572b118dc3c48d885d1c643c195314e/ https://experience.arcgis.com/experience/7572b118dc3c48d885d... Rysiek's National comparison grapher: https://rys.io/covid/ https://rys.io/covid/ ________________________________ Notes: † Some duplicates w/ parent.