The Student Room Group

How has your local area changed due to immigration?

Scroll to see replies

Reply 80

Original post
by tazarooni89
Quite a different argument from before then. Your previous statement was: “The likelihood of an event occurring doesn't magically change just because you don't know what the true underlying likelihood is.” Well actually under Bayesian probability it does precisely that.
We don’t need to assume a uniform prior. We can assume no prior distribution in particular, and what I’ve said will still hold.
Sure, some priors would make certain outcomes more likely than others. But for every such prior, there is another one available that makes those same outcomes less likely than others, to the same extent. For example, you could choose a prior where the would-be British community would be from amongst the top 5% of litterers, but there’s no reason to favour that over a prior where they’re from amongst the bottom 5% of litterers. To do so would be to make an assumption. If you avoid making assumptions and hence treat all possible priors as equally valid, you effectively end up with a uniform one - or at the very least a symmetric one anyway.
That’s why we can say all numbers on a dice are equally likely to come up. Physically speaking they may not be; perhaps the dice is weighted towards a 6. But it could just as easily be weighted towards the 1. Unless we’re prepared to start making assumptions about weighting, all numbers are equally likely because all possible weightings in any direction are equally plausible from our perspective.

It doesn't do that under Bayesian probability. The true likelihood will always remain the same (be it a fixed value or random variable), your prior knowledge or personal beliefs only change what the estimate of that likelihood is.

If you don't have a prior distribution then you're not able to make any probablistic statement since you've got not probability distribution to work with. Treating all possible priors as equally valid is also an assumption (a very bad one at that). To make no assumptions would be to make no claims about priors at all, which is not what you're doing.

Reply 81

Original post
by SHallowvale
It doesn't do that under Bayesian probability. The true likelihood will always remain the same (be it a fixed value or random variable), your prior knowledge or personal beliefs only change what the estimate of that likelihood is.


Okay, well that’s what we’re talking about here, isn’t it? I’m explaining my estimate of the likelihood based on my observations - not the objectively true, definite outcome based on an omniscience of information.

If you don't have a prior distribution then you're not able to make any probablistic statement since you've got not probability distribution to work with. Treating all possible priors as equally valid is also an assumption (a very bad one at that). To make no assumptions would be to make no claims about priors at all, which is not what you're doing.


What I’m doing nothing more than using the principle of insufficient reason.

“The principle of indifference (also called principle of insufficient reason) is a rule for assigning epistemic probabilities. The principle of indifference states that in the absence of any relevant evidence, agents should distribute their credence (or "degrees of belief") equally among all the possible outcomes under consideration.

In Bayesian probability, this is the simplest non-informative prior.”

Reply 82

Original post
by tazarooni89
Okay, well that’s what we’re talking about here, isn’t it? I’m explaining my estimate of the likelihood based on my observations - not the objectively true, definite outcome based on an omniscience of information.
What I’m doing nothing more than using the principle of insufficient reason.
“The principle of indifference (also called principle of insufficient reason) is a rule for assigning epistemic probabilities. The principle of indifference states that in the absence of any relevant evidence, agents should distribute their credence (or "degrees of belief") equally among all the possible outcomes under consideration.
In Bayesian probability, this is the simplest non-informative prior.”

Right, and I'm pointing out that your estimate relies on more assumptions than alternative explanations.

It doesn't matter if you're following the principle of insufficient reason. Treating all priors as equally valid, or by just directly using a uniform prior, is still making an assumption about what the underlying probability distribution is.

Reply 83

Original post
by SHallowvale
Right, and I'm pointing out that your estimate relies on more assumptions than alternative explanations.


So which alternative explanation requires fewer assumptions exactly?

It doesn't matter if you're following the principle of insufficient reason. Treating all priors as equally valid, or by just directly using a uniform prior, is still making an assumption about what the underlying probability distribution is.


No, we’re not making an assumption about the underlying probability distribution. Quite the opposite: we’re saying saying we have no reason to favour any underlying probability distribution over another. That means the best estimates are obtained by using the simplest non-informative prior, so they’re not biased one way or another. It does not mean we’re claiming that the true underlying distribution would be uniform; it could be anything.

The whole point of the principle of insufficient reason (or more specifically, the principle of maximum entropy) is that it deals with the absence of knowledge about the underlying probability distribution. What you called “an assumption and a very bad one at that” is actually a very well established principle in Bayesian Probability.

Reply 84

Original post
by tazarooni89
So which alternative explanation requires fewer assumptions exactly?
No, we’re not making an assumption about the underlying probability distribution. Quite the opposite: we’re saying saying we have no reason to favour any underlying probability distribution over another. That means the best estimates are obtained by using the simplest non-informative prior, so they’re not biased one way or another. It does not mean we’re claiming that the true underlying distribution would be uniform; it could be anything.
The whole point of the principle of insufficient reason is that it deals with the absence of knowledge about the underlying probability distribution. What you called “an assumption and a very bad one at that” is actually a very well established principle in Bayesian Probability.

The explanation would stop at why the differences in littering occur between the current communities (i.e. because of poverty). You've gone a step further to make a prediction about what the would-be British community would be like instead, which cannot be done without making an assumption.

The purpose of a prior distribution is to estimate what the underlying probability distribution is. By saying, 'we have no reason to favour any underlying probability distribution over another', is to make an assumption about it (i.e. that all possible distributions are equally likely). To make no assumption would be to make no prediction at all.

Non-informative priors are useful as a stepping stone for calculating posterior distributions once new data has been collected, which is not what you're doing (in your case they are effectively useless). It's also almost always the case that all priors are not equally valid (be it in population statistics or otherwise), that's why uniform priors are bad in general unless you have good reason to believe they're valid.

Reply 85

Original post
by SHallowvale
The explanation would stop at why the differences in littering occur between the current communities (i.e. because of poverty). You've gone a step further to make a prediction about what the would-be British community would be like instead, which cannot be done without making an assumption.

The purpose of a prior distribution is to estimate what the underlying probability distribution is. By saying, 'we have no reason to favour any underlying probability distribution over another', is to make an assumption about it (i.e. that all possible distributions are equally likely). To make no assumption would be to make no prediction at all.

Non-informative priors are useful as a stepping stone for calculating posterior distributions once new data has been collected, which is not what you're doing (in your case they are effectively useless). It's also almost always the case that all priors are not equally valid (be it in population statistics or otherwise), that's why uniform priors are bad in general unless you have good reason to believe they're valid.


I was perfectly happy to stop at saying that there’s a difference in the inclination to litter between these immigrant communities and the general British population (without even needing to add an extra layer of assumptions by saying “it’s because of poverty”, as you have done).

I’ve only gone a step further and made a prediction about the would-be British community because you specifically asked that question. Obviously I don’t know the answer, so I made my best estimate using the least biased method (a uninformative prior). Sure, I could make even fewer assumptions by simply refusing to answer your question, but since you asked it I just assumed that answering it was part of the exercise.

Reply 86

Original post
by tazarooni89
I was perfectly happy to stop at saying that there’s a difference in the inclination to litter between these immigrant communities and the general British population (without even needing to add an extra layer of assumptions by saying “it’s because of poverty”, as you have done).
I’ve only gone a step further and made a prediction about the would-be British community because you specifically asked that question. Obviously I don’t know the answer for sure, so I made my best estimate using the least biased method (a uninformative prior). Sure, I could make even fewer assumptions by simply refusing to answer that question, but since you asked it I assumed that answering it was part of the exercise.

If you stop at saying there's a difference in littering due to poverty then you cannot conclude that the immigrant communities have "ruined" Birmingham. That would require an assumption about the would-be British community (which stemmed our discussion about probability, etc).

Reply 87

Original post
by SHallowvale
If you stop at saying there's a difference in littering due to poverty then you cannot conclude that the immigrant communities have "ruined" Birmingham. That would require an assumption about the would-be British community (which stemmed our discussion about probability, etc).


No, because as I’ve already clarified, when I say that “immigrant communities have ruined parts of Birmingham”, I mean relative to what those areas were like before. I don’t mean relative to what they “would have” been like had some other unknown community moved there instead. My estimates on that have nothing to do with my original statement; they’re just an answer to a specific question that you posed.

Reply 88

Original post
by tazarooni89
No, because as I’ve already clarified, when I say that “immigrant communities have ruined parts of Birmingham”, I mean relative to what those areas were like before.
I don’t mean relative to what they “would have” been like had some other unknown community moved there instead. My estimates on that have nothing to do with my original point; they’re just an answer to a specific question that you posed.

How can you conclude that the immigrant community have "ruined" these parts of Birmingham if not by considering what they would be like if they hadn't moved there? If these areas would have declined anyway then immigration wouldn't have "ruined" anything.

Reply 89

Original post
by SHallowvale
How can you conclude that the immigrant community have "ruined" these parts of Birmingham if not by considering what they would be like if they hadn't moved there? If these areas would have declined anyway then immigration wouldn't have "ruined" anything.


The same way I can say “the twin towers were destroyed by aeroplanes” without having to think “ah but had it not been for the aeroplanes, maybe they would have been destroyed by earthquakes”. For the purposes of my statement it doesn’t matter what “would have” destroyed them, it matters what actually destroyed them.

Similarly, another community may or may not have ruined the areas by littering just as much had they lived there instead, but it’s besides the point. I’m talking about who actually ruined them, not who “would have” ruined them.

Reply 90

Original post
by tazarooni89
The same way I can say “the twin towers were destroyed by aeroplanes” without having to think “ah but had it not been for the aeroplanes, maybe they would have been destroyed by earthquakes”. For the purposes of my statement it doesn’t matter what “would have” destroyed them, it matters what actually destroyed them.
Similarly, another community may or may not have ruined the areas by littering just as much had they lived there instead, but it’s besides the point. I’m talking about who actually ruined them, not who “would have” ruined them.

That's not a valid comparison. Terrorist attacks and natural disasters are exceptional (very rare) events which have a direct causal impact on what they're affecting, we also have the very long history of modern architecture to tell us that buildings don't randomly become destroyed otherwise.

What you're observing, an area becoming worse over time (through littering, antisocial behaviour, etc), is hardly a rare phenomenon. It's entirely the point if another community would have "ruined" these areas instead. It's disingenuous, and in my opinion very toxic, to blame an immigrant community for doing something that a British community would have done anyway.

Reply 91

Original post
by SHallowvale
That's not a valid comparison. Terrorist attacks and natural disasters are exceptional (very rare) events which have a direct causal impact on what they're affecting, we also have the very long history of modern architecture to tell us that buildings don't randomly become destroyed otherwise.

What you're observing, an area becoming worse over time (through littering, antisocial behaviour, etc), is hardly a rare phenomenon. It's entirely the point if another community would have "ruined" these areas instead. It's disingenuous, and in my opinion very toxic, to blame an immigrant community for doing something that a British community would have done anyway.


Littering and antisocial behaviour doesn’t “randomly” happen either. They’re carried out by the people of the area. There is a direct causal impact here as well: When people drop litter, a place becomes littered. It doesn’t happen on its own.

How is it toxic to blame someone for something that someone else might have done anyway? By that logic you can never blame anyone for anything, because it’s always possible that someone else would have done it. Someone else also might have murdered the CEO of United Healthcare; in fact people die all the time anyway, and he eventually would have too.

Reply 92

Original post
by tazarooni89
Littering and antisocial behaviour doesn’t “randomly” happen either. They’re carried out by the people of the area. There is a direct causal impact here as well: When people drop litter, a place becomes littered. It doesn’t happen on its own.
How is it toxic to blame someone for something that someone else might have done anyway? By that logic you can never blame anyone for anything, because it’s always possible that someone else would have done it. Someone else also might have murdered the CEO of United Healthcare; in fact people die all the time anyway, and he eventually would have too.

This misses the point. The decline of an area and the rise of antisocial behaviour are reasonably common phenomena, terrorist attacks and earthquakes are not. It's easy to argue that an earthquake destroyed a building, it's not easy to argue that a community of people "ruined" a city. The murder of the United Healthcare CEO is also not comparable here, murder is something that happens extremely rarely.

If you accept that poverty is the underlying cause of antisocial behaviour and that a different community would have done the same then you're needlessly blaming the problem on immigration and culture. It's odd that you take issue with me describing this as toxic but not disingenuous...?

Reply 93

Original post
by SHallowvale
This misses the point. The decline of an area and the rise of antisocial behaviour are reasonably common phenomena, terrorist attacks and earthquakes are not. It's easy to argue that an earthquake destroyed a building, it's not easy to argue that a community of people "ruined" a city. The murder of the United Healthcare CEO is also not comparable here, murder is something that happens extremely rarely.


What has being rare or common got to do with it? If murder was very common would that mean Luigi Mangione is now blame-free? Exactly how common does it need to be to absolve him?

If you accept that poverty is the underlying cause of antisocial behaviour and that a different community would have done the same then you're needlessly blaming the problem on immigration and culture.


But I don’t accept that “poverty is the underlying cause” of anti-social behaviour and that another community would have done the same”. Those are assumptions on your part.

At most, I have accepted that poverty may be one of many influences, and that another community might have done the same. I also don’t believe this removes blame from the people doing it. The direct cause of littering is people who litter.

It's odd that you take issue with me describing this as toxic but not disingenuous...?

I take issue with both and await a full explanation as to why.

Reply 94

Original post
by tazarooni89
What has being rare or common got to do with it? If murder was very common would that mean Luigi Mangione is now blame-free? Exactly how common does it need to be to absolve him?
But I don’t accept that “poverty is the underlying cause” of anti-social behaviour and that another community would have done the same”. Those are assumptions on your part.
At most, I have accepted that poverty may be one of many influences, and that another community might have done the same. I also don’t believe this removes blame from the people doing it. The direct cause of littering is people who litter.
I take issue with both and await a full explanation as to why.

If an event is rare (e.g. natural disasters, terrorism, etc) then it's very reasonable to assume that without said event it's consequences (e.g. destroyed buildings, etc) wouldn't have occurred. To say the event "ruined" the thing it has affected would therefore be reasonable. In your case this doesn't apply, anti-social behaviour is by no means rare. If the murder of CEOs occurred very often then you could hardly argue that Luigi Mangione "ruined" anything. That doesn't absolve him of being a criminal, however, since he still broken the law.

If you don't think another community would have done the same then you've not explained why. If you don't know then you can't say Birmingham has been "ruined" by immigration, at best you can say 'I don't know'; after all, if another community were to do the same thing then immigration and culture wouldn't be the issue.

If something like littering were going to occur anyway (which seems very plausible judging from the areas you mentioned) then it's disingenuous to blame the immigrant community for "ruining" your city. I'm not sure how that would need explaining, it puts blame onto a community (and paints a very broad brush) for something that easily could have happened without them. All it's doing is using the immigrant community as a scapegoat for the deeper problems that exist in society (e.g. poverty).
Original post
by EuropeanIAm
Can you post your experiences?
This fuels hate.

Reply 96

Original post
by SHallowvale
If an event is rare (e.g. natural disasters, terrorism, etc) then it's very reasonable to assume that without said event it's consequences (e.g. destroyed buildings, etc) wouldn't have occurred. To say the event "ruined" the thing it has affected would therefore be reasonable.

Sounds like you’re using a uniform prior.

In your case this doesn't apply, anti-social behaviour is by no means rare. If the murder of CEOs occurred very often then you could hardly argue that Luigi Mangione "ruined" anything. That doesn't absolve him of being a criminal, however, since he still broken the law.

If you don't think another community would have done the same then you've not explained why. If you don't know then you can't say Birmingham has been "ruined" by immigration, at best you can say 'I don't know'; after all, if another community were to do the same thing then immigration and culture wouldn't be the issue.

If something like littering were going to occur anyway (which seems very plausible judging from the areas you mentioned) then it's disingenuous to blame the immigrant community for "ruining" your city. I'm not sure how that would need explaining, it puts blame onto a community (and paints a very broad brush) for something that easily could have happened without them. All it's doing is using the immigrant community as a scapegoat for the deeper problems that exist in society (e.g. poverty).


For the purposes of my statement it doesn’t make any difference whether someone else would have littered there anyway. If person A litters, then they are (contributing to) ruining the area. That’s regardless of whether person B would have littered there instead, had it not been for person A. If that were the case, person B would be ruining the area. Either way, someone is ruining the area - namely, whoever is littering. The place doesn’t litter itself.

And the people doing this are also breaking the law, so rarity vs. commonness / “someone else would have done it anyway” doesn’t absolve them either. Plus, poverty is hardly an excuse for it; plenty of people are poor and don’t litter. So why should those who do litter escape blame for it?

Reply 97

Original post
by tazarooni89
Sounds like you’re using a uniform prior.
For the purposes of my statement it doesn’t make any difference whether someone else would have littered there anyway.
If person A litters, then they are (contributing to) ruining the area. That’s regardless of whether person B would have littered there instead, had it not been for person A. In that case, person B would be ruining the area. Either way, someone is ruining the area - namely, whoever is littering. A place doesn’t litter itself.
And the people doing this are also breaking the law, so rarity vs. commonness / “someone else would have done it anyway” doesn’t absolve them either. So why should they not be blamed? Poverty is hardly an excuse for it. Plenty of people are poor and don’t litter.

That's not a uniform prior...? The events are not assumed to happen with equal chance, far from it.

It does makes a difference, unless you're actively looking to assign blame on an entire community for something that could easily have happened without them?

Reply 98

Original post
by SHallowvale
That's not a uniform prior...? The events are not assumed to happen with equal chance, far from it.


So then why are you assuming that, just because murder is rare in general, it means that the CEO specifically would not have died, had it not been for Mangione? What makes you think his personal probability of death would be anything like the general murder rate? Maybe someone else would have bean just as politically motivated to kill him. Maybe he would have died of an illness, or in some other way. You don’t know either way; it’s an extra assumption you’re having to make, that I don’t have to.

It does makes a difference, unless you're actively looking to assign blame on an entire community for something that could easily have happened without them?


No, it doesn’t make a difference. When I say “Person A did action X” I don’t mean “had it not been for person A, nobody else would have done X.” I simply mean “Person A did action X”. Sure, Person B could have done it instead had person A not been there, but in reality Person A did it. Person B has nothing to do with the truth of my statement.

You say that even if the murder rate was very high, it is still true to say that Mangione broke the law by murdering the CEO, and he is still blameworthy for having done so. I’m saying the same thing here for members of an immigrant community who litter.
(edited 11 months ago)

Reply 99

Original post
by tazarooni89
So then why are you assuming that, just because murder is rare in general, it means that the CEO specifically would not have died, had it not been for Mangione? What makes you think his personal probability of death would be anything like the general murder rate? Maybe someone else would have bean just as politically motivated to kill him. Maybe he would have died of an illness, or in some other way.
You don’t know that he wouldn’t have died anyway. It’s an extra assumption you’re having to make, that I don’t have to.
No, it doesn’t make a difference. When I say “Person A did action X” I don’t mean “had it not been for person A, nobody else would have done X. I simply mean “Person A did action X”. Sure, Person B could have done it instead, but in this instance Person A did it. Person B has nothing to do with the truth of my original statement.

There is a mountain of evidence on what the death rate at 50 is and on the general murder rate, so it's safe to assume that he would still be alive if not for Luigi. The CEO is part of the population this evidence comes from (not quite actually but accounting for this only makes his death less likely, being a wealthy CEO and all).

But you're not just blaming an individual for doing a crime, you're blaming an entire communtiy for "ruining" an area.

Quick Reply

How The Student Room is moderated

To keep The Student Room safe for everyone, we moderate posts that are added to the site.