Sep 29, 2022·edited Sep 29, 2022Liked by Modern Discontent
The graphs and figures section is also usually pretty hard.
My main personal frustration is this: I have limited time, perhaps 2-4 hours a day. I wan to find something interesting, that I could write about truthfully to inform, entertain and engage my readers.I like articles that could lead to unexpected conclusions.
So I have to read articles, twitter posts, news items etc. When looking at science articles I am not sure if the article has any amazing juicy material worth reporting on. For example (made up)
"Covid-19 incidence trends in prediabetic population of Kansas City, MO"
Such an article may be total dreck, or it may contain explosive findings. When I start the article out, I do not know. The abstract also does not help because it will say that the "vaccine is safe and effective" and abstracts are often highly misleading (even for more boring matters). So I have to look at the article to figure out if I should spend 2, 5, or 30 minutes. The risk is waste of time or missing a super amazing material.
To help myself decide which way to go, I often jump to figures and graphs to see what kind of data it provides and go from there. Any mention of "unvaccinated" is usually a good indicator that the article needs to be explored.
When I write, I always imagine a smart and nasty fact checker standing over my shoulder, ready to notice any mistakes and misrepresent my post by playing the mistakes up. So I am careful to qualify when I am not sure and write as transparently as I can.
Also I am mindful of wasting my readers time. If I waste 1 minute of each person who opens my post, I would end up wasting roughly ten 24-hour person-days total. So I end up deleting a lot of extraneous stuff.
I also give up about 2 out of 3 article ideas because my ideas were wrong, the results may be uninteresting, etc. Also, I try to have one idea, or at most 1.5 ideas in a post to avoid distraction. For example, yesterday, in my comparison of two breast milk studoes, I also wanted to discuss Gorski's critique of the 2022 breast milk study (and I am being generous to Gorski here), but decided against it to avoid idea overloading.
I loved the "How I read articles" post and I think that every substacker reporting on science needs to at least read it.
You and Brian play a very important role here.
Re: press. Sadly, their job is not to inform us, their job is serve their owners and "influence" us as desired, so my expectations are low. I wish that every journalist reads your "how to read articles" post and at least makes an effort to go beyond the headline and the last paragraph of the abstract. A tall order for people who chose journalism due to their inability to do science, I know.
I work in academia (in the humanities) and I dove into several studies for a project I was working on with race, class, and Covid -- mostly social science-type stuff. The biggest handicap for me is that I am not formally trained in statistical analysis, so I cannot fully understand some of the methodological choices. I have thought about taking a course in statistics through a university extension, simply because my interest lie in policy issues... Now, my biggest frustration with the literature is the mismatch between the results and the conclusion. To give you an example, many of the studies I read show that class was a bigger factor than race in terms of predicting bad COVID outcomes; in most studies, race ceased to be relevant once you controlled for income. Despite these findings, the conclusion read something akin to "the takeaway of this study is that we need to work to bring down systemic racism and white supremacy." What?! Because I work in academia, I know that academic journals will demand ideological conformity (this project on covid, race, and class ended up in journal purgatory because it drew the wrong conclusions), but there is something truly shocking about seeing a paper that deals with data outright contradict the results to fulfill the dogma du jour. I noticed something similar with vaccine papers; regardless of the results, the conclusions always highlighted the importance of vaccination against Covid.
I would suggest reporters, journalists,actually read through the entire publication. Keep a dictionary, medical dictionary, too, by your side. Read and think re the tables and figures included. Not easy.
Journalists and reporters make wide-sweeping statements without consideration of the shortcomings or limitations of the study results. I suspect their headlines and superficial treatments of findings is deliberate - clickbait and to add to the propaganda.
Sep 29, 2022·edited Sep 29, 2022Liked by Modern Discontent
Helpful would be an official "study reading forum." So like researchreadinggate as opposed to resarchgate. I am not sure if there are any unofficial ones. But this would be a place where you are reading a study, have a question, post it, maybe get an answer. The following example will show how this can't possibly go wrong:
BOBINOMAHA:
Hi, on this chart, does bla bla bla example question?
In seriousness, a good forum needs a certain percentage of humanity to be interacting with the subject; anything too esoteric and the forum turns into a backwater where the same three contributors answer the same questions with the same text snippets and it's random chance whether their approach is actually good or not. Likewise, reputation scoring (to weed out the SUEINSTLOUISes) probably needs a certain volume to work.
Sep 30, 2022·edited Sep 30, 2022Liked by Modern Discontent
I wish journalists with no scientific background would not report on studies, and realize they don't know enough to assess the validity of a study. If journalists report on a study, they need to get help from someone who can understand the methods used and can understand statistics if there is statistical analysis involved.
Off-topic, but since you're into those "molecule" things and what not you might be better at reviewing this one than me. Was brought up by Merogenomics youtube - https://www.mdpi.com/1420-3049/27/17/5405/htm
I would really like to do the book club style study. The problem for me is that I read a lot of abstracts and read a lot of content from doctors, etc and almost always they are written as if we all know what they are talking about. For instance, here is an article snippet by Dr. Peter McCullough which is in english but I really don't know the 'rules' for reading this. Most of his intro's are impossible to read or understand and he never gives links. Anyway, I have been writing software for 35+ years and I can tell you it's no different when I start talking about code, even in a general way, there is just no background in what I do for most people. I suspect this is much the same for medical studies for the average person.
A medical study of United Kingdom healthcare workers who had already had COVID-19 and then received the vaccine found that they suffered higher rates of side effects than the average population. Rachel K. Raw, et al., Previous COVID-19 infection but not Long-COVID-19 is associated with increased adverse events following BNT162b2/Pfizer vaccination, medRxiv (preprint), (last visited June 21, 2021).
Wow. I too have been thinking about this subject for some time. I tried look up YouTube videos but many of them were useless. They either had too much info or not enough.
The most difficult part of the studies is understanding the numbers...I want to be able to tell if they are off, be able to spot errors, & have the ability to do some of the math myself
I would love to get better acquainted with confidence intervals & hazard ratios lol
OF COURSE I forgot to include materials and methods! I'm not sure how to fix this but I may just include another poll question.
The graphs and figures section is also usually pretty hard.
My main personal frustration is this: I have limited time, perhaps 2-4 hours a day. I wan to find something interesting, that I could write about truthfully to inform, entertain and engage my readers.I like articles that could lead to unexpected conclusions.
So I have to read articles, twitter posts, news items etc. When looking at science articles I am not sure if the article has any amazing juicy material worth reporting on. For example (made up)
"Covid-19 incidence trends in prediabetic population of Kansas City, MO"
Such an article may be total dreck, or it may contain explosive findings. When I start the article out, I do not know. The abstract also does not help because it will say that the "vaccine is safe and effective" and abstracts are often highly misleading (even for more boring matters). So I have to look at the article to figure out if I should spend 2, 5, or 30 minutes. The risk is waste of time or missing a super amazing material.
To help myself decide which way to go, I often jump to figures and graphs to see what kind of data it provides and go from there. Any mention of "unvaccinated" is usually a good indicator that the article needs to be explored.
When I write, I always imagine a smart and nasty fact checker standing over my shoulder, ready to notice any mistakes and misrepresent my post by playing the mistakes up. So I am careful to qualify when I am not sure and write as transparently as I can.
Also I am mindful of wasting my readers time. If I waste 1 minute of each person who opens my post, I would end up wasting roughly ten 24-hour person-days total. So I end up deleting a lot of extraneous stuff.
I also give up about 2 out of 3 article ideas because my ideas were wrong, the results may be uninteresting, etc. Also, I try to have one idea, or at most 1.5 ideas in a post to avoid distraction. For example, yesterday, in my comparison of two breast milk studoes, I also wanted to discuss Gorski's critique of the 2022 breast milk study (and I am being generous to Gorski here), but decided against it to avoid idea overloading.
I loved the "How I read articles" post and I think that every substacker reporting on science needs to at least read it.
You and Brian play a very important role here.
Re: press. Sadly, their job is not to inform us, their job is serve their owners and "influence" us as desired, so my expectations are low. I wish that every journalist reads your "how to read articles" post and at least makes an effort to go beyond the headline and the last paragraph of the abstract. A tall order for people who chose journalism due to their inability to do science, I know.
I work in academia (in the humanities) and I dove into several studies for a project I was working on with race, class, and Covid -- mostly social science-type stuff. The biggest handicap for me is that I am not formally trained in statistical analysis, so I cannot fully understand some of the methodological choices. I have thought about taking a course in statistics through a university extension, simply because my interest lie in policy issues... Now, my biggest frustration with the literature is the mismatch between the results and the conclusion. To give you an example, many of the studies I read show that class was a bigger factor than race in terms of predicting bad COVID outcomes; in most studies, race ceased to be relevant once you controlled for income. Despite these findings, the conclusion read something akin to "the takeaway of this study is that we need to work to bring down systemic racism and white supremacy." What?! Because I work in academia, I know that academic journals will demand ideological conformity (this project on covid, race, and class ended up in journal purgatory because it drew the wrong conclusions), but there is something truly shocking about seeing a paper that deals with data outright contradict the results to fulfill the dogma du jour. I noticed something similar with vaccine papers; regardless of the results, the conclusions always highlighted the importance of vaccination against Covid.
I would suggest reporters, journalists,actually read through the entire publication. Keep a dictionary, medical dictionary, too, by your side. Read and think re the tables and figures included. Not easy.
More than 100,000 views in 24 hours, approaching 1,000 comments
"Safe and Effective - A Second Opinion" documentary
oraclefilms.com/safeandeffective
Journalists and reporters make wide-sweeping statements without consideration of the shortcomings or limitations of the study results. I suspect their headlines and superficial treatments of findings is deliberate - clickbait and to add to the propaganda.
Helpful would be an official "study reading forum." So like researchreadinggate as opposed to resarchgate. I am not sure if there are any unofficial ones. But this would be a place where you are reading a study, have a question, post it, maybe get an answer. The following example will show how this can't possibly go wrong:
BOBINOMAHA:
Hi, on this chart, does bla bla bla example question?
SUEINSTLOUIS:
VIRUSES DON'T EXIST BOB!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
In seriousness, a good forum needs a certain percentage of humanity to be interacting with the subject; anything too esoteric and the forum turns into a backwater where the same three contributors answer the same questions with the same text snippets and it's random chance whether their approach is actually good or not. Likewise, reputation scoring (to weed out the SUEINSTLOUISes) probably needs a certain volume to work.
I wish journalists with no scientific background would not report on studies, and realize they don't know enough to assess the validity of a study. If journalists report on a study, they need to get help from someone who can understand the methods used and can understand statistics if there is statistical analysis involved.
Off-topic, but since you're into those "molecule" things and what not you might be better at reviewing this one than me. Was brought up by Merogenomics youtube - https://www.mdpi.com/1420-3049/27/17/5405/htm
I liked that Igor mentioned you yesterday when he was comparing the recent lactation/vac shedding study to the one from a year or so ago
Conflicts of interest are the issues reporters should always include in their writing.
I would really like to do the book club style study. The problem for me is that I read a lot of abstracts and read a lot of content from doctors, etc and almost always they are written as if we all know what they are talking about. For instance, here is an article snippet by Dr. Peter McCullough which is in english but I really don't know the 'rules' for reading this. Most of his intro's are impossible to read or understand and he never gives links. Anyway, I have been writing software for 35+ years and I can tell you it's no different when I start talking about code, even in a general way, there is just no background in what I do for most people. I suspect this is much the same for medical studies for the average person.
https://www.americaoutloud.com/risks-of-vaccines-for-those-recovered-from-covid-19-krammer-raw-mathioudakis/
```
A medical study of United Kingdom healthcare workers who had already had COVID-19 and then received the vaccine found that they suffered higher rates of side effects than the average population. Rachel K. Raw, et al., Previous COVID-19 infection but not Long-COVID-19 is associated with increased adverse events following BNT162b2/Pfizer vaccination, medRxiv (preprint), (last visited June 21, 2021).
```
Wow. I too have been thinking about this subject for some time. I tried look up YouTube videos but many of them were useless. They either had too much info or not enough.
The most difficult part of the studies is understanding the numbers...I want to be able to tell if they are off, be able to spot errors, & have the ability to do some of the math myself
I would love to get better acquainted with confidence intervals & hazard ratios lol
A club is a great idea