Interesting/surprising that several of your own predictions are so extreme (98, 98, 95, 5). You must feel very confident on those.
I love footnote 5 :)
I'm wondering if I've misunderstood the net migration question. Wasn't the previous year's figure close to a million? So are you saying you're 65% confident that it will have reduced by half (or more) this year? What's that based on?
With hindsight the 98s should probably have been 95.
- For the three leaders I put at 95 or 98, given their parliamentary positions and/or polling positions, I think only death/severe illness/Party-gate level scandal would see them removed, which seem unlikely.
- It seems very unlikely that Denmark or Panama cede territory, at least this year (even if they opened negotiations or something, I can't see such a sensitive matter being done).
- The base rate of Presidential assassinations is about 1 per century. I've upped this to 5% as Trump has had two attempts already; this implies a 20% chance of him being assassinated during his tenure, which isn't entirely unreasonable but I'm already worrying is on the high side.
Re immigration, the Conservative Government brought in a load of more restrictive measures which largely hadn't kicked in for this year, but will for the next (the one in the forecast). I had thought when I made the forecast that the OBR was predicting it was therefore going to drop to about 350k; they have historically underestimated this, so I still gave a 35% chance it would be over 500k. However, I now see that the ONS is predicting it will be considerably higher than 500k, so I suspect I've got that question wrong.
I'm pretty sure it's just an error - the total number doing it this time was <2% of entrants, which seems consistent with people making a mistake. The people who do this do it for all the entries (some just use yes or no, others throw in maybes and probablies).
I didn't get any other deliberate invalid answers (like negative answers). I get a few typos - e.g. someone writing 900 instead of 90 on a question, with the rest filled out correctly. There are few enough of these that I correct them manually if it's obvious, or change them to 50 if it's not.
Regarding footnote 3, which is more important - reducing discrimination or maximising effectiveness? If it were the case that diversity improved organisational effectiveness, or that organisation leaders thought it did, then aiming to increase diversity might well be discriminatory, but would be a rational attempt to maximise effectiveness. Would you still oppose it?
I'm not a complete purist: I accept the Catholic/Protestant quotas in Northern Ireland in certain areas, under the Good Friday agreement, as a necessary evil. If we were in a war situation, and there was strong evidence that mixed units (or single sex units) were clearly more effective, then OK.
In general though, we should be have a high prior that claims that discrimination will improve effectiveness are false, given the number of times hokum claims have been deployed to justify people's ideologies (everything from Victorian racial pseudoscience to the debunked McKinsey study on board diversity). In general, selecting the person whose best for the job - or allowing anyone who passes a standardised test (e.g. in fitness, say for police force roles, or an academic one, for medicine) to do a role - is going to give better results, be fairer and - crucially - less open to abuse.
Yes, my limited understanding is that the best, recent evidence suggests gender and ethnic diversity has no effect (positive or negative) on organisation performance. If we had strong evidence, I'm not sure why a war situation would be different to running a company, hospital, etc. but it's moot. Some long-running questions to ask about how you select the best person for the job but we do have some useful ideas and evidence there, I think.
Interesting/surprising that several of your own predictions are so extreme (98, 98, 95, 5). You must feel very confident on those.
I love footnote 5 :)
I'm wondering if I've misunderstood the net migration question. Wasn't the previous year's figure close to a million? So are you saying you're 65% confident that it will have reduced by half (or more) this year? What's that based on?
With hindsight the 98s should probably have been 95.
- For the three leaders I put at 95 or 98, given their parliamentary positions and/or polling positions, I think only death/severe illness/Party-gate level scandal would see them removed, which seem unlikely.
- It seems very unlikely that Denmark or Panama cede territory, at least this year (even if they opened negotiations or something, I can't see such a sensitive matter being done).
- The base rate of Presidential assassinations is about 1 per century. I've upped this to 5% as Trump has had two attempts already; this implies a 20% chance of him being assassinated during his tenure, which isn't entirely unreasonable but I'm already worrying is on the high side.
Re immigration, the Conservative Government brought in a load of more restrictive measures which largely hadn't kicked in for this year, but will for the next (the one in the forecast). I had thought when I made the forecast that the OBR was predicting it was therefore going to drop to about 350k; they have historically underestimated this, so I still gave a 35% chance it would be over 500k. However, I now see that the ONS is predicting it will be considerably higher than 500k, so I suspect I've got that question wrong.
I'm curious about the people who continue to write "yes" or "maybe" even after you've repeated the instructions many times in all caps.
Are they just trolling, do you think?
Do they do that for all the questions, or do they mostly answer with numbers and then throw in the occasional "maybe"?
Do they write longer commentary / value judgements, like "In your dreams" or "I bloody well hope not"?
Do the same people also give other invalid answers, like negative numbers or numbers over 100?
I'm pretty sure it's just an error - the total number doing it this time was <2% of entrants, which seems consistent with people making a mistake. The people who do this do it for all the entries (some just use yes or no, others throw in maybes and probablies).
I didn't get any other deliberate invalid answers (like negative answers). I get a few typos - e.g. someone writing 900 instead of 90 on a question, with the rest filled out correctly. There are few enough of these that I correct them manually if it's obvious, or change them to 50 if it's not.
Regarding footnote 3, which is more important - reducing discrimination or maximising effectiveness? If it were the case that diversity improved organisational effectiveness, or that organisation leaders thought it did, then aiming to increase diversity might well be discriminatory, but would be a rational attempt to maximise effectiveness. Would you still oppose it?
Interesting hypothetical.
I'm not a complete purist: I accept the Catholic/Protestant quotas in Northern Ireland in certain areas, under the Good Friday agreement, as a necessary evil. If we were in a war situation, and there was strong evidence that mixed units (or single sex units) were clearly more effective, then OK.
In general though, we should be have a high prior that claims that discrimination will improve effectiveness are false, given the number of times hokum claims have been deployed to justify people's ideologies (everything from Victorian racial pseudoscience to the debunked McKinsey study on board diversity). In general, selecting the person whose best for the job - or allowing anyone who passes a standardised test (e.g. in fitness, say for police force roles, or an academic one, for medicine) to do a role - is going to give better results, be fairer and - crucially - less open to abuse.
Appreciate your response.
Yes, my limited understanding is that the best, recent evidence suggests gender and ethnic diversity has no effect (positive or negative) on organisation performance. If we had strong evidence, I'm not sure why a war situation would be different to running a company, hospital, etc. but it's moot. Some long-running questions to ask about how you select the best person for the job but we do have some useful ideas and evidence there, I think.