S. Wessely - Death threats, abuse, smear campaigns - Standing up for Science: 29 March
- Thread starter Yogi
- Start date
SW said (from Janet Eastham's twitter feed) "they changed the recovery measure because they realised they had gone too extreme & they would have the problem that nobody would recover"
Talk about revealing yourself, & them, without realising what you just did!
That was my thought too.... reeeeaally Simon? is that what they did? because it's precisely what theyve been accused of doing all this time, & they said they didnt. - Their version of events (afaiaa) was that they decided to change the thresholds for other reasons.
-the ''too extreme'' threshold in the original protocol was 85 - which as i understand it is the normative for a healthy working age person - which is presumably why they chose it in the first place - because it's what any normal intelligent person would reasonably interpret "recovered" to mean.
But according to Simon, when they realised no one was going to hit it, & "they would have the problem that nobody would recover", instead of acknowledging the reality of their situation... ie THAT NO BODY HAD RECOVERED FOLLOWING THEIR INTERVENTIONS.... they decided that those results didnt fit in with their reality so they had to change something. So basically they went "eh? we were sure this would work, but it doesnt appear to have done so.... it cant be that our hypothesis is wrong, it cant possibly be that we have to change our beliefs in light of the facts, our version of reality is set in stone, so it must be that we made a mistake in how we planned to interpret the data. We are so certain that people will recover with CBT/GET that what we define as recovered will have to change to support our reality".
and if they want to argue that a score 60 is more normative than 85, then why did they recruit a bunch of healthy people for a trial testing treatments for sick people? Which ever way you look at it it's a load of bollocks.
I know we know that already, but i think old SW slipped up there...., saying, publicly, & for impressive/humourous effect, that they knew nobody was going to hit 85 so they changed it.... why did they know nobody would hit it? because either they recruited people they knew were not ill & were on the 'normal' scale to begin with. Or, they thought a bunch of people would hit it & when they saw that it hadnt worked & no one would recover, they couldnt have that result, it didnt fit with their reality, & therefore they changed the threshold for recovery to worse health than those they recruited in the first place.
I know we all know this, but this guy's hubris will be the down fall of the lot of em if he keeps 'showing off' in this way to his audience in these arenas.
I used to feel infuriated by him being given a platform to spout & be admired....
I'm also now smiling & thinking - yes good, you go on, give him plenty of opportunities for adulation, because he'll start to feel even more untouchable .... 'give em enough rope'
Talk about revealing yourself, & them, without realising what you just did!
That was my thought too.... reeeeaally Simon? is that what they did? because it's precisely what theyve been accused of doing all this time, & they said they didnt. - Their version of events (afaiaa) was that they decided to change the thresholds for other reasons.
-the ''too extreme'' threshold in the original protocol was 85 - which as i understand it is the normative for a healthy working age person - which is presumably why they chose it in the first place - because it's what any normal intelligent person would reasonably interpret "recovered" to mean.
But according to Simon, when they realised no one was going to hit it, & "they would have the problem that nobody would recover", instead of acknowledging the reality of their situation... ie THAT NO BODY HAD RECOVERED FOLLOWING THEIR INTERVENTIONS.... they decided that those results didnt fit in with their reality so they had to change something. So basically they went "eh? we were sure this would work, but it doesnt appear to have done so.... it cant be that our hypothesis is wrong, it cant possibly be that we have to change our beliefs in light of the facts, our version of reality is set in stone, so it must be that we made a mistake in how we planned to interpret the data. We are so certain that people will recover with CBT/GET that what we define as recovered will have to change to support our reality".
and if they want to argue that a score 60 is more normative than 85, then why did they recruit a bunch of healthy people for a trial testing treatments for sick people? Which ever way you look at it it's a load of bollocks.
I know we know that already, but i think old SW slipped up there...., saying, publicly, & for impressive/humourous effect, that they knew nobody was going to hit 85 so they changed it.... why did they know nobody would hit it? because either they recruited people they knew were not ill & were on the 'normal' scale to begin with. Or, they thought a bunch of people would hit it & when they saw that it hadnt worked & no one would recover, they couldnt have that result, it didnt fit with their reality, & therefore they changed the threshold for recovery to worse health than those they recruited in the first place.
I know we all know this, but this guy's hubris will be the down fall of the lot of em if he keeps 'showing off' in this way to his audience in these arenas.
I used to feel infuriated by him being given a platform to spout & be admired....
I'm also now smiling & thinking - yes good, you go on, give him plenty of opportunities for adulation, because he'll start to feel even more untouchable .... 'give em enough rope'
85 is below normal for a healthy working age person.
17% of the working population fall below 85. 22% are at 85 or less. The median (typical value is 100).
If we take out those people with long term health conditions (i.e. to get healthy working age people) the about 8% of people are under the 85 range with 61% scoring 100, 17% 95, 9% on 90 and 4% are on 85.
So 85 is way below the norm and 85 covers the majority of the healthy working age population.
17% of the working population fall below 85. 22% are at 85 or less. The median (typical value is 100).
If we take out those people with long term health conditions (i.e. to get healthy working age people) the about 8% of people are under the 85 range with 61% scoring 100, 17% 95, 9% on 90 and 4% are on 85.
So 85 is way below the norm and 85 covers the majority of the healthy working age population.
Ah. thanks for that, I stand corrected.
Do you have a source/ref for those figures please? Not that i dont believe them, it makes sense to me, i'd just like to be able to quote them & to do that I'd like to be able to back them up with where i got them.
Do you have a source/ref for those figures please? Not that i dont believe them, it makes sense to me, i'd just like to be able to quote them & to do that I'd like to be able to back them up with where i got them.
I think Wessely was trying to imply that SF36 Physical Functioning score of 85 is too stringent. It's completely bullshit of course, and he relies on the ignorance of his audience to get away with it. But he may have been trying to avoid telling an outright lie, and ended up telling a bit too much of the truth in the process.
I can believe it. Prosecuting someone for this would be extraordinarily difficult, unless there's a record of them somewhere saying "Let's do some fraud!", more or less. But I can easily see it killing their careers, and I think that's a much better outcome to aim for.
This is why it would be very hard to prove outright fraud. In most crimes, intent is obvious in that someone carried out a physical act which requires intentional actions. It's less clear when the inappropriate action could be due to gross incompetence instead of only intent. But for the purposes of ending their careers, it doesn't matter. They've been doing research for decades, and can't claim that they just need a second chance to learn how to conduct research properly.
I can believe it. Prosecuting someone for this would be extraordinarily difficult, unless there's a record of them somewhere saying "Let's do some fraud!", more or less. But I can easily see it killing their careers, and I think that's a much better outcome to aim for.
This is why it would be very hard to prove outright fraud. In most crimes, intent is obvious in that someone carried out a physical act which requires intentional actions. It's less clear when the inappropriate action could be due to gross incompetence instead of only intent. But for the purposes of ending their careers, it doesn't matter. They've been doing research for decades, and can't claim that they just need a second chance to learn how to conduct research properly.
Last edited:
- Messages
- 65
- Likes
- 159
https://www.eventbrite.co.uk/e/london-march-for-science-tickets-32229648747?aff=erelexpmlt
"London March for Science
DESCRIPTION
The March for Science celebrates publicly funded and publicly communicated science as a pillar of human freedom and prosperity. We unite as a diverse, nonpartisan group to support science that upholds the common good, and political leaders and policymakers who enact evidence-based policies in the public interest.
On Saturday, 22 April scientists and science enthusiasts will come together in a massive, diverse, and celebratory demonstration of public support for science that highlights challenges facing the scientific community and will act as a catalyst for ongoing action around specific issues.
The exact route is being finalised, this information will be made available in due course.
ALL ARE WELCOME!
DATE AND TIME
Sat 22 April 2017
11:00 – 15:00 BST
LOCATION
London"
"London March for Science
DESCRIPTION
The March for Science celebrates publicly funded and publicly communicated science as a pillar of human freedom and prosperity. We unite as a diverse, nonpartisan group to support science that upholds the common good, and political leaders and policymakers who enact evidence-based policies in the public interest.
On Saturday, 22 April scientists and science enthusiasts will come together in a massive, diverse, and celebratory demonstration of public support for science that highlights challenges facing the scientific community and will act as a catalyst for ongoing action around specific issues.
The exact route is being finalised, this information will be made available in due course.
ALL ARE WELCOME!
DATE AND TIME
Sat 22 April 2017
11:00 – 15:00 BST
LOCATION
London"
thanks. & great job working that out.
Sorry to be dense but what am i searching for then in order to find that report? or even Bowling's Paper? I've never heard of either & the people i would tell those figures to will want to check the facts. No worries just ignore me if it's a hassle.
Sorry to be dense but what am i searching for then in order to find that report? or even Bowling's Paper? I've never heard of either & the people i would tell those figures to will want to check the facts. No worries just ignore me if it's a hassle.
In the PACE recovery paper they quote a paper analysing a survey published by Bowling:
The reference for the Bowling paper is
The bowing work is based on a survey from the office of population and surveys.
From this the stats can be worked out. The assumptions they make in the PACE paper is that the results for the SF36 are distributed with a distribution that is parameterised by mean and SD. Which is not the case when you look at the distribution. This means using the median and percentile is more robust. The Bowling paper doesn't give these but they can be calculated from the data which is what I have done.
The reference for the Bowling paper is
The bowing work is based on a survey from the office of population and surveys.
From this the stats can be worked out. The assumptions they make in the PACE paper is that the results for the SF36 are distributed with a distribution that is parameterised by mean and SD. Which is not the case when you look at the distribution. This means using the median and percentile is more robust. The Bowling paper doesn't give these but they can be calculated from the data which is what I have done.
Wessely seems to realise he went a bit too far:
- Messages
- 902
- Likes
- 2,921
Oh what a relief. For once I thought he might actually have said something relevant.
Is "ambiguous" a word anyone used? SW probably wishes it was, and as usual thinks that if he says it, everyone will accept it as true. In fact looking through the twitter comments no-one has used the word "ambiguous", although I did spot "nonsense" and "disgraceful". Why can't he just say "It's been pointed out that my answer was disgraceful nonsense" instead of trying to reframe the story as usual?
- Messages
- 2,157
- Likes
- 17,726
''Before the analysis'' is very different from (or should that be different to...) ''before anyone knew any of the results''. They did not need to do any analysis to see that hardly anyone in any of the groups had reached 85 or more on SF-36 and that there wasn't a significant between group difference.
So even if they were blinded to which group of data was which, it was blindingly obvious the project had failed. All it would have taken was a glance at the figures. Similarly with the walk and fitness tests.
So even if they were blinded to which group of data was which, it was blindingly obvious the project had failed. All it would have taken was a glance at the figures. Similarly with the walk and fitness tests.
It's been pointed out that for once my answer wasn't sufficiently ambiguous.
...because it was blindingly obvious to us that CBT and GET weren't having the desired effects long before we started analysis. But hopefully by offering a 'clarification' on the *timing* of the change I can create the impression that I mispoke when I blurted out the *reason* for the change.
I'm deeply embarrassed that - despite my many years experience in the art of obfuscation - on this occasion I spoke a little too candidly. I will try not to let it happen again.
...because it was blindingly obvious to us that CBT and GET weren't having the desired effects long before we started analysis. But hopefully by offering a 'clarification' on the *timing* of the change I can create the impression that I mispoke when I blurted out the *reason* for the change.
I'm deeply embarrassed that - despite my many years experience in the art of obfuscation - on this occasion I spoke a little too candidly. I will try not to let it happen again.
Last edited: