CASE STUDY: The kids in med school are all right!

FRIDAY, JANUARY 13, 2023

Our blue tribe elites, not so much: Today, as our endless Case Study ends, we offer some excellent news.

We're here today to tell you this:

The kids seem to be all right!

At this point, let us make a correction. The "kids" to whom we refer aren't exactly kids. 

We refer instead to the 328 medical students and medical residents who participated in a UVa study back in 2015. The UVa study's alleged results are frequently by the journalists of our failing blue time. 

With that correction made, we're here today to tell you this:

Those medical students and residents seem to be all right!

At this point, another correction! In the UVa study in question, many of these trainees were surveyed, but only some of the trainees' responses were reported. 

Responses from 108 "nonwhite" trainees were essentially disappeared. For reasons which went unexplained, the authors only reported the responses of 222 "white" trainees.

According to ongoing blue tribe lore, those 222 "white" trainees plainly weren't all right. According to one account of the study, this is how disturbing and shocking those (whites only) trainees actually were:

SABIN (1/6/20): “Black people’s nerve endings are less sensitive than white people’s.” “Black people’s skin is thicker than white people’s.” “Black people’s blood coagulates more quickly than white people’s.”

These disturbing beliefs are not long-forgotten 19th-century relics. They are notions harbored by far too many medical students and residents as recently as 2016. In fact, half of trainees surveyed held one or more such false beliefs, according to a study published in the Proceedings of the National Academies of Science. I find it shocking that 40% of first- and second-year medical students endorsed the belief that “black people’s skin is thicker than white people’s.”

For the record, the headline on Professor Sabin's essay said that these were white medical trainees. And, as you can plainly see:

In Sabin's assessment, half of those white trainees just weren't all right at all. 

Those (white) trainees held false beliefs about biological differences between whites and blacks which were shocking and disturbing. Indeed, their false beliefs seemed to come straight out of this nation's long forgotten 19th century past.

Everyone knows what such claims mean. In our tribe, we find such claims highly pleasing.

We've never seen anyone overstate this case to the extent that Professor Sabin did. The professor's a good and decent person, but she really ran wild in that piece.

That said, this UVa study is frequently cited by blue tribe journalists, and the assessment is always the same. The assessment goes like this:

Half a group of (white) medical students gave voice to racist-adjacent beliefs!

With all that said, we're here today to share the good news! Those 328 medical trainees, white and nonwhite alike, don't seem to have given voice to shocking and disturbing beliefs after all! 

Also this, not that it actually matters:

The nonwhite medical trainees—the trainees whose responses were disappeared—were actually slightly more inclined to give voice to allegedly false beliefs! That said, the difference between these two groups of trainees was very slight. 

Once again, here's some of what Kevin Drum found deep inside the bowels of the widely cited UVa study. Drum is using "S&Rs" as shorthand for "[medical] students and residents:"

DRUM (1/4/23): S&Rs were allowed to mark an answer as "possibly," "probably," or "definitely" true or false. With one exception, which I'll get to, virtually every single person who marked a false statement as true said it was only "possibly true." Among all the false statements, there were 229 marks of "possibly true" and only 9 marks of "probably true." There was not a single mark of "definitely true."

I said there was one exception, and this is it: I didn't count the marks from the question about the thickness of Black skin. This is the huge outlier, with 41% of first and second-years believing it and even 23% of thirds and residents believing it.

[...]

Beliefs of white and nonwhite respondents are virtually identical. In particular the average score for nerve endings is 1.94 vs. 1.83 (nonwhite S&Rs are more likely to believe it) and 1.76 vs. 1.73 for skin thickness. Overall, the belief in false statements is 2.06 vs. 1.98, meaning that nonwhite S&Rs are slightly more likely to believe them than white S&Rs.

Belief in false statements is not a problem. The percentages are low and the responses are almost all tentative.

We'll strongly disagree with Drum on one important point. He seems to accept the absurd procedure in which respondents were judged to have scored an allegedly false statement as being "true" if they merely marked the statement as being "possibly true."

Since respondents weren't allowed to say that they simply didn't know if a given statement was true, that procedure strikes us as a baldly absurd—as a ludicrous, disgraceful and ugly thumb clamped down on the scale. 

(For the record, we assume that this is what Drum means when he says that responses which stated agreement with an allegedly false belief "[were] almost all tentative." We presume he means that all such responses merely said that a statement was possibly true.)

We very strongly disagree with Drum on that point. Other than that, Drum brought the good news from his laborious deep dive into the UVa study:

The medical trainees, white and nonwhite together, voiced agreement with very few false statements. Indeed, they almost never said that any of the allegedly false statements were probably or definitely true.

In this way, these (white and nonwhite) medical students seem to be all right! They didn't stand in line to voice agreement with the shocking, disturbing racist beliefs of our 19th century past.

That said, Drum noted one exception—the allegedly false statement "about the thickness of black skin." The statement in question was this:

Black people’s skin has more collagen (i.e., it’s thicker) than White people’s skin.

Is that statement true? Is it true that black people's skin is thicker than white's people's skin?

For ourselves, we don't have the slightest idea. Nor do we know of any basis on which any such minor biological difference could possibly matter. 

That said, quite a few respondents "agreed with" that allegedly false statement. This means that they at least chose to say that the statement was possibly true (as most statements are). 

In that vastly limited sense, quite a few respondents "agreed with" that particular statement. Let us say this about that, this time with reference to this pregnant footnote from Drum:

DRUM: However, it's worth noting that this is an active area of research that has produced some contradictory results. See here, here, here, and here. 

Is it possible that black people's skin really is somewhat thicker? Almost everything is possible, and Drum offers links to four studies which seem to suggest that this allegedly false statement could possibly be true.

You can journey to Drum's report to click on his four links. For the record, the headlines on his second link read exactly like this:

Black Skin
Asian and black skins have a thicker dermis than white skin, the thickness being proportional to the intensity of pigmentation.

Briefly googling, we found a fair number of reports which seemed to make roughly similar claims. Lacking knowledge in this area, we have no idea what the (utterly pointless) truth may actually turn out to be.

That said, is it possible that some of those medical students had been exposed to such reports? Could that be why this one allegedly false statement drew much more "agreement" than any of the others, even among the third and fourth year trainees?

We don't know how to answer that question. Neither does anyone else.

In Drum's assessment, this UVa research effort was "a dog's breakfast of a study." We'll quote him at greater length:

DRUM: Overall, this is a dog's breakfast of a study. The authors end up focusing on whether S&Rs who harbor more false beliefs also tend to rate pain lower in Black patients compared to better-informed S&Rs. It turns out they don't, but they do rate pain in white patients higher. However, the amount is smallish; it makes little difference in treatment; and the statistics presented seem cherry-picked and gnawed at a little too carefully. I'm not really sure I put much stock in the authors' conclusions.

"The statistics seem cherry-picked," Drum unkindly says. In our view, several basic procedures involved in the study are astoundingly hard to explain.

First among those strange procedures was the unexplained decision to disappear the responses by the "nonwhite" medical trainees—responses which Drum found to be "virtually identical" to those of the "white" trainees.

From there, the jet plane rumbled down the runway, lugging tribal narrative as its dominant load. Soon, our blue tribe had the latest exciting claim which helps to drive the self-defeating, ugly Storyline we very stupidly like.

The woods are lovely, dark and deep—but our failing blue tribe is deeply in love with the spotting of racists. Joe McCarthy found Commies under every bed. Today, our blue tribe goes him one better.

Alas, our poor failing nation! The red tribe traffics in crazy beliefs. Our own deeply unimpressive tribe traffics is conduct like this.

The Others can often see how disordered we are. That disorder is a basic state of affairs we can't seem to see in ourselves.

As we close, we return you to another account of the UVa study. This account appeared in the New York Times Sunday Magazine, written by a healthcare journalist who is a good, decent person and who currently sits at the very top of the field:

VILLAROSA (4/11/18): In 2016, a study by researchers at the University of Virginia examined why African-American patients receive inadequate treatment for pain not only compared with white patients but also relative to World Health Organization guidelines. The study found that white medical students and residents often believed incorrect and sometimes “fantastical” biological fallacies about racial differences in patients. For example, many thought, falsely, that blacks have less-sensitive nerve endings than whites, that black people’s blood coagulates more quickly and that black skin is thicker than white.

According to Professor Sabin, the UVa study showed that These White Medical Students Today held shocking and disturbing beliefs straight out of the 19th century.

("For example, many thought, falsely, that blacks have less-sensitive nerve endings than whites?" Alas! Of the 87 third and fourth year white trainees who participated in the UVa study, only one failed to rate that statement as some version of untrue. In that sense, we'd say that those trainees were all right. We're less sure about the acclaimed medical writer who made the statement we've quoted.)

According to Linda Villarosa, the UVa study showed that These White Medical Students Today often, or at least sometimes, believe “fantastical” biological fallacies. It's very, very, very hard to affirm Villarosa's account of the UVa study.

The white kids surveyed in this study held shocking, disturbing beliefs. They sometimes believed "fantastical" notions. 

"Virtually identical" responses by the nonwhite kids were disappeared by these tribal elites. This is the way the game is played by our own failing tribe at this time.

Sadly but plainly, our failing, deeply unwell blue tribe loves claims of this type. We seem to love these invidious claims more than we love life itself. 

The red tribe now runs on crazy belief. As can be seen on a daily basis, our own tribe now runs on this!


83 comments:


  1. tl;dr

    Yawn. Yes, dembottery is everywhere; what else is new?

    ...your tribe needs it, so it's being spread, ubiquitously. Get over it, der Bob; get used to it...

    ReplyDelete
    Replies
    1. "what else is new?"
      Not much. Self-proclaimed anti-gay Conservatives are still preying on young men.

      Delete
  2. Trump stole the top secret documents. Biden merely misplaced them.

    ReplyDelete
    Replies

    1. Sounds like liberal tribe's bosses hint the Big Guy that he should go away quietly in 2024. Or else.

      Delete
    2. Trump said the documents were his and hid them, refusing to turn them over when requested. Trump filed lawsuits to continued to keep them. Biden's staff failed to return documents and Biden was unaware he had any. His lawyers found them and he turned them over immediately when they were discovered. There are also differences in the number and type of documents stolen by Trump, classification levels and subject matter. Trump had a purpose for stealing certain documents. Biden's retention of classified material was accidental.

      Delete
    3. The whole top secret document thing is a sideshow created by Republicans to take attention away from the Democrat's legislative focus on helping the poor.

      Delete
    4. As President, Trump arguably had a right to take any docs he wanted to, since he had the power to declassify anything. One may disagree with this claim, but at least Trump had an argument for his right to take the docs.

      OTOH VP Biden had no power to declassify docs. He had no argument to justify taking them. So, in one way, what Biden did was worse.

      Delete
    5. No, Trump did not have the right to steal classified documents after he left office, not even declassified ones. There is a law that requires presidents to maintain and turn over documents, letters, papers, photos, everything, to the National Archives after the conclusion of his term in office. Trump's claims have lost in court, repeatedly.

      Biden did not know about the documents and he has turned them over immediately, as required.

      So, no, what Biden did was not worse.

      Delete
    6. I'm waiting for confirmation on Biden's "Top Secret" document. That's the sort of thing that gets misreported (how'd that happen?) corrected once, and then ignored as the talking points stay the same.

      Delete
    7. Biden's lawyers labeled some of the documents as top secret.

      Delete

    8. Shouldn't y'all be screaming that the documents were planted?

      Read your talking points as soon as they arrive, you lazy dembots.

      Delete
    9. Trump's stack of documents was thicker because they had more collagen.

      Delete
    10. @11:48, they weren't concerning nuclear secrets (like Trump's were), Biden's were largely old daily briefings, from what I've been reading.

      Delete
    11. They were top secret documents about Ukraine and Iran.

      Delete
    12. David, you are arguably the most dishonest sonofabitch to comment on this blog.


      *************************
      The Vice President’s Declassification Authority
      By Steven Aftergood • February 16, 2006

      “Is it your view that a Vice President has the authority to declassify information?” Vice President Cheney was asked yesterday by Fox News’ Brit Hume.

      “There is an executive order to that effect,” replied the Vice President.

      This was a simple answer to a straightforward question, but the matter is actually a bit more complicated.

      The executive order in question is E.O. 13292 on classified national security information, issued by President Bush in March 2003.

      It states in section 1.3 that “The authority to classify information originally may be exercised only by: (1) the President and, in the performance of executive duties, the Vice President; (2) agency heads and officials designated by the President in the Federal Register…”

      Remarkably, the phrase “and, in the performance of executive duties, the Vice President,” which dramatically elevates the Vice President’s classification authority to that of the President, was added to the executive order in 2003.

      Prior to that, the Vice President only had classification authority comparable to that of an agency head, having been delegated such authority in a 1995 presidential order.

      So much for classification authority. What about declassification?

      Declassification authority is defined in Section 6.1(l) of E.O. 13292. It is granted to: “(1) the official who authorized the original classification…; (2) the originator’s current successor in function; (3) a supervisory official of either; or (4) officials delegated declassification authority in writing by the agency head or the senior agency official.”

      So the Vice President has authority to declassify anything that he himself classified. He also clearly has authority to declassify anything generated in the Office of the Vice President, which he supervises.

      But is the Vice President, like the President, “a supervisory official” with respect to other executive branch agencies such as the CIA? Did the 2003 amendment to the executive order which elevated the Vice President’s classification authority also grant him declassification authority comparable to the President’s?

      “The answer is not obvious,” said one executive branch expert on classification policy.

      Delete
  3. "At this point, another correction! In the UVa study in question, many of these trainees were surveyed, but only some of the trainees' responses were reported. "

    The results for all of the participants were reported in the study. The results for the nonwhite subjects were analyzed in the same manner as those for the white subjects, but they showed no significant differences and thus were not described in detail. This is standard practice when reporting research results because journal space is expensive and limited, so wasting space describing data that showed no differences across conditions is not done. This is true for all studies, not just this one. The findings of the study consisted of the significant differences in the white group, which were reported in detail and were the focus of the study because of their pain undertreatment of the black but not the white target in the treatment task.

    Here Somerby again introduces the word "trainee" to describe medical students. Medical students are not trainees because they are not permitted to do any medical treatment, only observe, because they are unlicensed and have not yet finished medical school, not graduated yet and are not doctors in any sense, not even trainee doctors.

    The only trainees in the group were the medical residents, who are engaged in the hands on treatment of patients. The rest of the subjects were not trainees at all, but medical school students. Medical school involves sitting in classrooms or attending lab sessions, while learning topics such as anatomy and physiology, pharmacology, medical ethics, and biology.

    "...medical school typically begins with health-related science classes in fields like biochemistry, physiology, pharmacology, anatomy, histology, microbiology and genetics. It also includes lessons on how to interview and interact with patients..."

    The specifics of how to treat people are acquired in practicums where they observe (shadow) physicians, and attend rounds at a hospital, and assist by learning to take a patient history. They only learn to practice medicine when they become interns, or in the first year of a residency (should they decide to specialize). That is when they are actual trainees. No interns were subjects in this study. Only the residents could be considered trainees. Residency is advanced training that occurs after the interns have finished their training and taken their board exams to become licensed physicians legally permitted to practice medicine.

    Why is this important? Because the use of this inappropriate term (to advance his argument) shows Somerby's ignorance about what medical students have learned and what they have done in their medical education. If he doesn't know this, he doesn't know much at all about medicine, including and especially the requirements imposed on hospitals for treatment of pain, which is now considered the fifth vital sign.

    No one blames the medical students for their ignorance, and no one blames Somerby either. Unlike the medical students, Somerby seems to be unaware of his own ignorance but has, in a bumbling manner, vigorously asserted several mistaken critiques of a study and made derogatory claims about it based on nothing but his own hubris and a desire to avoid racial findings in a report he does not like. Somerby is wrong and he prevents criticism of his own essay by failure to read his comments, so he has continued to be wrong, day after day after tiresome day. That isn't how journalism is done and it sure doesn't contribute anything to anyone's knowledge. But that is not Somerby's goal. For the past few weeks he has been a propagandist, working for the right wing to avoid any appearance of racism, even where it indisputably exists.

    ReplyDelete
    Replies
    1. I don't think you've taken into account all the relevant factors.

      Delete
    2. I think if you reexamine the data you will see you have misrepresented a number of methodologies and practicums.

      Delete
    3. You seem to be unaware of your own ignorance. You've moved the goal posts considerably in different ways,

      Delete
    4. Last Wednesday Somerby specifically addressed the issues you bring up here about the term trainee.

      Delete
    5. On January 11, Sabin apparently used the word trainee in one quoted sentence. Somerby then ran with it, repeating the word trainee over and over (19 times), especially as he discussed the nonwhite subjects. Drum says trainee 0 times. That is when I first objected to the word.

      Nowhere does Somerby explicitly "address," much less "specifically," the use of the word trainee himself, or by Sabin. It comes across as bigoted when Somerby does it, because he combined the frequent use of the word with discussion of nonwhite students, supposedly (but not actually) disappeared by the study.

      It is unclear whether Sabin herself used the word "trainee" in the opinion piece that Somerby links to, since it appears in an introductory paragraph not written by the author herself (she is referred to in 3rd person):

      "Half of white medical trainees believe such myths as black people have thicker skin or less sensitive nerve endings than white people. An expert looks at how false notions and hidden biases fuel inadequate treatment of minorities’ pain."

      This appears at the top of Sabin's opinion piece in the journal Academic Medicine. It may be that an editor introduced the term in order to shorten the description of subjects for a nontechnical audience.

      That doesn't change what Somerby did. And nowhere does he "address" the issues I brought up.

      Delete
    6. Sorry, you're right. It was the Wednesday before that that he addressed it.

      Delete
    7. Yes it was. Just use your computer to access the data. Don't forget to include the New Year's holiday.

      Delete
    8. To heck with the New Year’s holiday.

      Delete
  4. "("For example, many thought, falsely, that blacks have less-sensitive nerve endings than whites?" Alas! Of the 87 third and fourth year white trainees who participated in the UVa study, only one failed to rate that statement as some version of untrue."

    Notice how Somerby cooks the books on this one by including only the 3rd & 4th year medical students. This is a subset of the entire group of subjects, which also included 1st year students and residents. Why does Somerby only mention this more limited group? Presumably because those are the results that make his case, not the wider group upon which the study was conducted.

    Why not simply exclude all of the subjects who failed to mark that statement as some version of false entirely? That would make Somerby's case even stronger. He could say no one considered it true, and imply that no subjects believe in any misinformation. If Somerby is going to cherrypick among the results, why not go the whole way?

    Somerby never explains how it is possible to have a significant difference between those white subjects with more information and those with less, if there were no differences in how they endorsed misinformation, as both Somerby and Drum claim. And if there were no differences in misinformation (a statement not supported by the results), then where did the significant undertreatment for pain come from, observed in the treatment recommendations for the white vs black target patients?

    If Somerby and Drum are correct about the misinformation being too minimal to have affected behavior, then the results of the treatment task (which are significantly different for white vs black targets) must have some other cause. We are left explaining that significant result by subject racism, because misinformation cannot be causal, according to Somerby and Drum.

    Somerby perhaps thinks that if he can discredit the survey of misinformation, then the racism doesn't exist either. It doesn't work that way. He has done nothing to explain away the significant finding that white subjects undertreated the black targets for pain.

    I don't find Drum's analysis convincing because isolating a composite scale into individual questions and then claiming none are significant on their own, is inappropriate and ignores the procedures used by the researchers to assign subjects to high and low misinformation groups, and because Drum does not grapple with the FACT that significant differences were observed either. Drum should know better but he too is engaging in specious criticism that ignores the point of the study, the treatment task.

    ReplyDelete
  5. "The white kids surveyed in this study held shocking, disturbing beliefs. They sometimes believed "fantastical" notions. "

    These are not kids. All were at least 22 and the residents were 5 years older than that. Given that many medical students take a gap year or longer between undergrad and med school, to prepare applications, do interviews and prepare for MCAT exams, they may be even older.

    Is Somerby trying to infantilize med students in order to make them appear to be victims of those nasty researchers? At least some of these subjects appear to be apt to make some hapless black person endure needless pain in an ER or hospital room, without explicit training. The med students are not the victims.

    Which is worse: (1) having someone imply you might be ignorant about black physiology or perhaps bigoted, not by name, but in the findings of a study or news report you probably won't read, or (2) enduring throbbing or sharp pain from a broken limb while the doctor tells you to suck it up and insists you should do fine with a couple of Tylenol?

    ReplyDelete
    Replies
    1. No, as you get older you start to view kids in their 20s... as kids. You have a consistent thread through all of your analyses of being hamstrung as it were about simple explanations like this, always trying to read into things.

      How to explain? It's like a driver that swears "this guy is blocking me on purpose!" and then pulls up to see the other driver is simply on their cellphone and not paying attention.

      Delete
    2. This is a scientific paper that refers to its subjects as subjects, participants or medical students/residents, not trainees (unless they are) or kids (unless they are children). You don't use everyday language to report scientific studies. You are precise.

      Somerby is making a play on the title of a movie called The Kids are Alright (or perhaps a Who song with the same name). It is cute as a lead sentence, but when he continues, he is trying to portray subjects as children (young, innocent) who are being victimized by those mean ole researchers (who are trying to call them racist, except they are actually saying that -- they are calling them misinformed about black physiology and thus more likely to undertreat black people for pain).

      Language matters, especially to propagandists like Somerby. People choose words for a reason. This is nothing like making an incorrect causal attribution while driving.

      Delete
  6. "It's very, very, very hard to affirm Villarosa's account of the UVa study."

    Especially when you arent' even trying.

    Somerby started out claiming that no nonwhites were included in the study. That suggests he didn't read it carefully, which surely must be the starting point.

    ReplyDelete
  7. With all the seditionists Republicans put in Committee Chairs, I'm starting to think the GOP hates America as much as they hate Americans.

    ReplyDelete
    Replies
    1. Have you considered that the best way to prove that the government is incompetent is by making it so?

      Delete
  8. ""Virtually identical" responses by the nonwhite kids were disappeared by these tribal elites. This is the way the game is played by our own failing tribe at this time."

    There were no significant findings for the nonwhite subjects (who were not "kids"). The study clearly says that. Nothing was "disappeared," much less by "tribal elites."

    Has the entire science enterprise now become part of a red or blue tribe, simply because Somerby doesn't like a specific finding of one study? Is Somerby saying that Susan Fiske, a highly respected professor of Social Psychology and the Editor of the Proceedings of the National Academy of Sciences who approved publication of this study, is herself a tribal elite? She is a distinguished professor at Princeton (that is a job title, not praise) with a great deal more experience than Somerby or Drum (combined) in how to conduct experimental studies. Is Somerby accusing her now of being a Democrat or liberal? On what basis? Is he saying she is playing some game? These are serious claims and Somerby should exercise some care when he plays his own propaganda games using real people's lives, as he has done for weeks now by maligning the authors of this study, simply because he doesn't want to admit that racism is a real phenomenon in today's world.

    ReplyDelete
  9. Sadly, the moral of this story is to never trust media reports of scientific studies. That goes double when the study supposedly supports their preferred narrative.

    ReplyDelete
    Replies
    1. And don't trust Somerby either. Go read the study for yourself. Oh, no link? What does that tell you about Somerby's honesty?

      Delete
    2. anon 11:33 - in his earlier posts about the study there were links to the study. Perhaps TDH should have included the links in subsequent posts. I looked at the study based on the prior links, and the study seems ridiculous. One example that I have brought up in previous posts is this. Answers by subjects that a "wrong" answer is "possibly true" are deemed by the study as wrong answers - the same as if the answer was "probably" or "definitely" true. But answers by subjects who said the wrong statements were "possibly false" were deemed to be correct answers, the same as if they said the statement was "probably" or "definitely" false. This is ass-backwards. Saying something is "possibly true" is basically the same as saying it is "probably false." Saying that the statement is "possibly false" is the same as saying it is "probably true." Nobody here has responded to this point (insanely), as usually happens, or otherwise) Aside from that, lumping subjects who said something was "possibly true" with those (very few) who said it was "probably" or "definitely" true makes no sense, as TDH has repeatedly noted.

      Delete
    3. And what is the “preferred narrative” in this case, David?

      It is that “Black Americans are systematically undertreated for pain relative to white Americans.” (That’s a quote from the study. )

      Except that isn’t a made up narrative. Multiple studies have established this. They are mentioned in both the study and in Sabin’s article. We’ve all had about a month to go and peruse those studies. Except Somerby didn’t. Did you?

      Based upon this, you are the one believing a “narrative.”

      Delete
    4. “ lumping subjects who said something was "possibly true" with those (very few) who said it was "probably" or "definitely" true makes no sense, as TDH has repeatedly noted.”

      AC, What do you mean it “makes no sense?” The more such answers the white med students gave, the more likely they were to under treat for pain in the black patient. A single “possibly true” for a false statement didn’t result in a large gap in pain rating, but the more such answers, the lower the pain rating, resulting in a significant gap. How many times do we have to go back over this?

      Delete
    5. AC: “Saying something is "possibly true" is basically the same as saying it is "probably false."”

      But those two responses did not correlate in the same way to the pain rating.

      Do you understand that, AC?

      Delete
    6. "Answers by subjects that a "wrong" answer is "possibly true" are deemed by the study as wrong answers - the same as if the answer was "probably" or "definitely" true. But answers by subjects who said the wrong statements were "possibly false" were deemed to be correct answers, the same as if they said the statement was "probably" or "definitely" false. "

      In the research literature, it is common for researchers to ask a question and then ask the subject to rate the confidence with which they hold that answer. Are they sure they are right about it or do they have some doubt. This is called a "confidence rating" and it can be done in any study where you ask a subject to make a judgment of some sort, including general knowledge and trivia questions.

      In the case of the possibly true and possibly false choices, the confidence rating is part of the answer itself, instead of being separated into two question, one giving the answer (true/false) and the other giving the degree of confidence about that answer. On a true/false question, true and false are not the same answer. Confidence in the answer given is distinct from the answer itself. Definitely, probably, possibly are confidence ratings. True and false are the answer themselves. That is why possibly true and possibly false are two distinct answers. The word "possibly" only refers to how sure the subject is about the answer chosen (true/false). And low confidence in a response is not the same as saying "I don't know". This is a forced choice response format in which subjects must guess when they don't know. That doesn't mean the answer is meaningless, because the guesses will be guided by something, slight knowledge, stereotype, chance, or a tendency to choose true vs false (or vice versa) in situations of uncertainty (called a guessing bias). Researchers study such things and there is a large and complex literature about how people answer questions.

      Somerby's and Drum's speculations about question responses are no substitute for actual knowledge, which is part of being trained in research methods. These guys do not know what they are talking about. Your reliance on your own "common sense" is misleading you in similar ways.

      The study is far from ridiculous. So is the problem black people face when undertreated for pain. That you would consider such a study to be ridiculous says more about you, than it does about the study. It is the mark of an ignorant person. The kinds of criticisms raised by Somerby and Drum, and parrotted by you, reveal your ignorance. There are criticisms to be made of this study (or any study), but you don't know what they are -- neither do Drum and Somerby -- and I sure as hell am not going to tell you. In the estimation of a famous editor and 5-7 anonymous peer reviewers, this study had something to contribute to the knowledge of other researchers. That is worth a lot more than you, Somerby and Drum's ignorant attempts to discredit scientific inquiry in favor of their own ignorant and uninformed judgment.

      You all have the right to make fools of yourselves. Rest assured that you are doing a splendid job of that.

      Delete
    7. Y'know, if you take a random bunch of people and divide them randomly into 2 groups, there is 100% chance that one group will have, on average, thicker skin, quicker blood coagulation, etc.

      Therefore, "possibly true" and "possibly false" (interchangeably) are always the correct answers to this meaningless bullshit list of questions.

      ...and that's all there is to it...

      Delete
    8. Somerby's core thesis as of yet remains completely unchallenged.

      Delete
    9. Somerby’s core thesis (and Drum’s) is that racism is no longer a problem.

      There are mountains of evidence demonstrating this thesis is utter nonsense.

      Let’s clarify for the confused: there is no such thing as right wing ideology; right wingers are people obsessed with dominance, often expressed via racism, or other forms of oppression. This obsession with dominance is likely a function of unresolved childhood trauma, which can affect brain development.

      Somerby and Drum are right wingers, no different in their core motivations than any other Republican.

      Delete
    10. That is not their core thesis. Go away troll.

      Delete
    11. I agree with @5:46. He is not a troll.

      Delete
    12. This comment has been removed by the author.

      Delete
    13. mh -- the preferred narrative in this case is that blacks are greatly harmed by racism and that whites are very substantially racist. (BTW to say that this is the preferred narrative is not to say that it's correct or incorrect.)

      Delete
    14. Racism harms everyone. People vary in their racism.

      Delete
    15. David, is this narrative possibly true or possibly untrue?

      Delete
  10. Under Biden, the cost of donuts has decreased for average Americans.

    ReplyDelete
  11. Notice that Somerby has once again failed to cite the UVa study itself today by providing a link. He has tended to 'disappear' the study itself more often than not during this lengthy discussion. Today he links twice to Drum, but not to the study. Does he not trust his readers -- is this how he cooks the books and puts his own thumb on the scales? Yes, he wants you to take his word for the study's deficiencies, but shouldn't we be allowed to judge for ourselves by reading the actual study?

    ReplyDelete
  12. "The red tribe now runs on crazy belief. As can be seen on a daily basis, our own tribe now runs on this!"

    What is the "this" that Somerby decries? Some might call it science.

    I'd rather run on science than on crazy.

    ReplyDelete
  13. Bob now believes that liberal reverse racism witchhunt is worse than McCarthyism. An easy conclusion when, for years, you totally ignore racism when it appears in the traditional form.
    Easy to dismiss any point on MSNBC when you never address what they say, and merely smirk about liberals wanting to put people in jail.
    The Daily Howler has become a useless hate screed.

    ReplyDelete
    Replies
    1. Communists are bad. Racists are worse.

      Delete
    2. There are fewer communists in the US today than there are racists. The racists are in positions of power where they can harm others, not so much for the communists. Unless you are the kind of idiot who thinks that everyone left of Matt Gaetz is a communist.

      Delete
  14. Voter fraud is real!

    https://talkingpointsmemo.com/news/iowa-republican-politicians-wife-indicted-for-alleged-voter-fraud-scheme

    ReplyDelete
    Replies
    1. Well, tell the Republicans to knock it off!

      Delete
  15. Somerby stakes an objection on the truth or falsity of one statement among the 11 considered false, the one about skin thickness:

    "Briefly googling, we found a fair number of reports which seemed to make roughly similar claims. Lacking knowledge in this area, we have no idea what the (utterly pointless) truth may actually turn out to be.

    That said, is it possible that some of those medical students had been exposed to such reports? Could that be why this one allegedly false statement drew much more "agreement" than any of the others, even among the third and fourth year trainees?

    We don't know how to answer that question. Neither does anyone else."

    Here is what Somerby is missing. First, that one question may have accounted for some of the endorsements of false statements, but it didn't account for all of them. Second, if the statement is true and the students had read about it, then why did their endorsements correlate with undertreatment of the black target patient for pain? Third, it is possible that the subjects genuinely believed in the truth of every false statement they endorsed. How would that change the FACT that they undertreated the black target patient for pain? Is Somerby seriously arguing that if any of the statements considered false were actually true, it would justify undertreating black people for pain? Fourth, given that endorsement of the statement about the thickness of skin was correlated with undertreatment of black patients, how is that knowledge "utterly pointless"? Is concern about the health care of black people also utterly pointless to Somerby? Finally, Somerby says he has no way of knowing whether the students previously saw such an article (such as the studies cited by Drum) and then Somerby says: "and neither does anyone else." Somerby is apparently unaware that studies frequently ask subects whether they have seen stimuli or certain kinds of material before. It is a way of validating their stimuli. Concerns about prior exposure to questions on a survey are part of research. Such questions are usually part of the debriefing of a subject, but they are commonly included in careful research. They wouldn't be reported in the published study report unless they affected the outcome.

    Somerby has a tendency to make ignorant remarks about topics he knows very little about. This is another such remark, since he is entirely untrained in research methods and apparently thinks everyone in the world is stupid except himself. Especially Einstein, Godel, and trained scientists in fields he has not studied.

    ReplyDelete
    Replies
    1. Poor people are untreated for pain. This is a class issue, not a race issue.

      Delete
    2. Yes, poverty is a problem, but it doesn't explain the findings in this study.

      Delete
    3. anon 1:44, you don't disprove, or even address, my main point. A response to a false statement that it is "possibly true" is treated as an incorrect response. However, "possibly true" is the equivalent of "probably false." So a correct answer is treated as an incorrect answer in the study. There many "possibly correct" responses in the study. On the other hand "possibly false" responses are treated in the study as correct; but a "possibly false" response is the equivalent of a "probably true" response. Doesn't this treatment of correct responses as incorrect, and vice versa, lead to a distorted correlation between correct and incorrect answers and insufficient treatment? If not, could you kindly explain why?

      Delete
    4. I worked with someone named Jamille that was a white female. Her coworker told a story once that when she was told she would be sitting by "Jamille" that she panicked a bit and then was relieved to see it was a white girl.

      To my credit, I had no such reaction. I didn't expect any particular sex and race based on first hearing the name.

      These days, it might be different for me. It gets to you. The 24/7 discussion of people based SOLELY on their race, it has become front and center. It's doing damage.

      I can't even imagine how it will impact the next generation. Hopefully they are resilient enough to think it's all weird and have stopped paying attention.

      Delete
    5. Presumably, this happened to your coworker back before there was this modern focus on race. Her panic suggests it wasn't a woke racial focus but actual racism that did damage in her case. Imagine her horror if she had actually had to sit next to a black girl! That racist impact seems a lot more damaging to me than having to learn that people of different races can and should get along together, even to the point of sitting next to each other at school.

      Already, people in more tolerant areas of the country are intermixing, dating, marrying, having kids across races and noone is being damaged by that. I suspect the next generation will see this happening more widely across the country, to the point where even people in Idaho and Eastern Washington may sit in booths beside each other in diners. As Somerby says, anything is possible.

      Delete
    6. Name one area of the country that people are not intermixing, dating, marrying, having kids across races.

      Delete
    7. Yes, her panic was a mix of racism and sexism. She feared she would be sitting next to a black male. Like I said I didn't have any reaction to the name.

      You think any of the modern focus on race has helped this other worker that exhibited the racism? I don't. I bet it's alienated her and made her more racist. Has it helped me? No, it's taken away my ability to see people as 100% who they are and not think about their race. It hasn't made me more or less racist, just handicapped by having to notice race all the time when I used to not consider it.

      Thanks for reading my story, hope it was helpful to someone in some way.

      Delete
    8. AC/MA, when you read the report, it says that after the subjects marked their surveys, the responses to the true questions were discarded (not included) and only the responses to the 11 false questions were considered. These were not scored as correct or incorrect, but were coded as "1 = definitely untrue, 2 = probably untrue, 3 = possibly untrue, 4 = possibly true, 5 = probably true, 6 = definitely true" and the numbers corresponding to each of the 11 answers were then averaged together to produce a composite single score used to divide the group of subjects into low misinformation and high misinformation subjects (probably using the median as the cutoff). The authors say they did an analysis using all 15 of the questions and the results were the same. Note that this is simply classifying subjects into two groups, not examining the content of any of the specific questions (as Somerby and Drum did).

      If the questions were misclassified or the subjects were in the wrong groups, it would tend to work against finding a clear-cut difference between the two groups of subjects, and it would weaken the correlations. The authors do discuss the possible impact of misclassifying the bone strength question and explain why they think it didn't cause a problem. They didn't find a difficulty with the skin density question because it didn't produce a similar impact on the data (the subject responses). Drum's speculation that it could have done so, is specious if the data itself doesn't show any impact.

      Moving some of the high misinformation subjects into the low misinformation group (because a question was right instead of wrong), would have the effect of making the two groups slightly more similar to each other, not more different. Drum says that almost no subjects gave definite answers to any of the questions. They used "possibly" in their responses. It seems unlikely that shifting a response from possibly true (3) to possibly false (4) is going to shift the average (11 numbers averaged together across 255 subjects) sufficiently to shift any subjects from one group to the other. Nor is considering a 3 instead of a 4 (or vice versa) going to affect the correlations much, if at all. It is the composite of 11 scores averaged together that classifies a subject, not any single answer.

      That isn't the difficulty. Moving subjects into different groups wouldn't change their pain treatment decisions. The significant effect is in those treatment decisions and those will not have been changed just by reclassifying a question or one subject out of 255.

      The worst that might happen would be a less significant result. There wouldn't be a significant undertreatment finding due to a true statement mistakenly being included in the misinformation survey. The undertreatment decisions do not depend on the classification of the subjects as low or high misinformation. At worst, there might be a controversy over whether their explanation for why the undertreatment was happening is justified by the data. It wouldn't make the undertreatment or the correlation go away.

      Delete
    9. "The authors say they did an analysis using all 15 of the questions and the results were the same."

      This is the equivalent of considering all of the true questions as false information and including them in the analysis. This means that the authors have already tested the impact of misclassifying false statements as true ones, and they found no impact on the results.

      Somerby and Drum may not have understood what that statement meant in the report.

      Delete
    10. "I bet it's alienated her and made her more racist."

      Racism is apparently now being blamed on anti-racist attempts to help everyone get along better.

      Delete
    11. 4:03 it doesn't seem as if you really understand academia.

      Delete
    12. anon 4:03, all very interesting, but doesn't address my point at all. Try re-reading it, maybe you'll see

      Delete
    13. I did address your point, assuming you were in good faith. I have nothing more to say to you.

      Delete
  16. Here is one of Drum's mistakes, quoted by Somerby:

    "The authors end up focusing on whether S&Rs who harbor more false beliefs also tend to rate pain lower in Black patients compared to better-informed S&Rs. It turns out they don't, but they do rate pain in white patients higher. However, the amount is smallish; it makes little difference in treatment; and the statistics presented seem cherry-picked and gnawed at a little too carefully. "

    When reading, it is important to carefully notice the difference between the pain ratings and the treatment decisions. These are two separate tasks that were given to all subjects following the false statement survey. The results of these two distinct tasks were graphed in Figure 2. Panel A shows the pain ratings and Panel B shows the treatment decisions. Both compare subjects classified as low and high misinformation based on their survey responses.

    Drum talks only about the pain ratings. Both high and low misinformation groups rated the black pain the same. They rated the white pain as lower for low misinformation subjects and higher for high misinformation subjects. This is interesting but not germane to the research question, which is whether high misinformation leads to undertreatment of pain. It suggests that subjects were not rating pain differently depending on misinformation, but they were rating white and black pain differently. A different pattern of results is shown there, which Kevin Drum dismisses.

    Here is the bad part. Kevin Drum does not discuss at all the main finding of the study -- that the two information groups made significantly different treatment decisions, with the high misinformation group undertreating the black target for pain but not undertreating the white target, and a significant correlation between the amount of misinformation and the degree of undertreatment. These are the results that Drum entirely dismisses (by never mentioning it). It is as if Drum doesn't recognize what was done in the study and doesn't bother looking at Panel B, where the effect is obvious.

    Then Drum has the nerve to accuse the study of cherry-picking its data and using dubious statistics. It is as if Drum only looked at part of the study, the phone rang, he got distracted, and he wrote a bunch of nonsense in support of Somerby. And then Somerby has the nerve to talk about things being "disappeared" when his own buddy Drum disappeared the main findings of the study.

    This is a clown show. I get it that Drum has been ill and Somerby is old, but they are attacking the reputations of serious scholars with these foolish attacks, where they try to convince the unwary that woke is run amok in academia. It is dishonest and beneath the reputations these two guys used to have in the blogosphere.

    ReplyDelete
    Replies
    1. Your writing is actually painful to read.

      I'm sure it's my fault as a reader. I'll invite you to explain my shortcomings. I'm sure you can divine them based on these few words I've written here.

      Delete
  17. The criticism that Somerby describes (raised by Drum), that the skin question may actually be true instead of false, would tend to work against obtaining a significant result for undertreatment of pain by high misinformation subjects (some of whom might actually be low misinformation). If low misinformation protects against undertreatment, including more low misinformation people in the high misinformation group (via mistakenly including a true statement among the false ones), should cause there to be less undertreatment because low information subjects tend not to undertreat black subjects. If this were happening, then there would be a stronger effect of undertreatment, not a weaker one.

    In short, this doesn't discredit the study at all. It suggests that the undertreatment effect observed may in fact be stronger than it appears in this study.

    It troubles me that Somerby and Drum do not seem to be able to reason well about the consequences of the problems they are raising. This is a two-step process. First you identify a weakness or challenge to the methods, second you examine how that weakness might effect the data, the findings claimed by the study. Existence of a flaw, by itself, does not overturn a study. There needs to be an impact on the data that would make the claimed findings untenable. Somerby and Drum have failed the second part -- they seem to have no idea what would happen to the pattern of results if a question were removed or some subjects were moved from one group into the other.

    ReplyDelete
    Replies
    1. You come across as someone who really doesn't understand how to interpret data or read reports.

      Delete
    2. Yeah I don't know, if you skim read their comments they seem like good analysis. But then pick one paragraph as an exercise and try to divulge it's meaning. Try this one for example:

      "Drum talks only about the pain ratings. Both high and low misinformation groups rated the black pain the same. They rated the white pain as lower for low misinformation subjects and higher for high misinformation subjects. This is interesting but not germane to the research question, which is whether high misinformation leads to undertreatment of pain. It suggests that subjects were not rating pain differently depending on misinformation, but they were rating white and black pain differently. A different pattern of results is shown there, which Kevin Drum dismisses."

      Delete
    3. Trolling. Notice that he doesn't contradict anything in the paragraph supposedly used as an example of skimming.

      Delete
    4. 323 p.m. is dealing with a number of outdated modalities. At this point, they are engaging in the intellectual equivalent of strapping young athletes snapping towels at each other's privates in the shower. Nothing more, nothing less.

      Delete
    5. You still scoping that booty like a big game hunter?

      Delete
  18. Bob ended this series the same week that the Russiagate delusion was finally exposed to be beyond all doubt a fraud, and the same week that Biden's presidency essentially ended amidst a criminal investigation.

    ReplyDelete
    Replies
    1. wishful thinking, this series will never end

      Delete