The downside of the DC school lottery

[Anonymous]

Anonymous wrote:

Anonymous wrote:I do not understand why there are so many slots per entry. Does it increase or decrease your chances of getting in if you do not use all the slots? And why do people even put in slots that they don’t really want?

It increases your chances of getting into a school if you can list more schools. It will not affect how good your number is. But we matched with our #12 choice. If we could only list 10, we would have had no match. We did really want it, just not as much as we wanted the 11 above it. But we enrolled and it has been good so far.

I have a hard time believing that there are 12 schools that are that much better than an IB for a 3 year old when you take into account commutes. There aren’t 12 good public/charter MS options in the entire city. Six is a more reasonable number.

[Anonymous]

I'm not sure why this optimally condition is so sacred, except that the result is so clean for this set of very strong assumptions. If we are worried parents would want to switch, why not allow for a scheme where parents can swap their assignments? Honestly, once parents get an assignment, they will start to value it more than they might have abstractly conceived of it beforehand.anyway. I count 132 options for PK3 and I can only list 12... strategizing is already happening.

[Anonymous]

Anonymous wrote:

Anonymous wrote:Also, the "game" is repeated with high positive serial correlation whereas the DC Lottery implementation and associated proofs of efficiency assume only a single "game".

Please expand on this thought. (Formatting corrected)

The nice clean proof of the "optimality" of this algorithm rests on many debatable assumptions. One assumption is that this matching happens only once. Of course, the DC lottery happens every year. Ostensibly, the lottery result for this year only effects next year. However, this isn't true. It is a repeated game where prior results and future anticipated results effect what I should do. For example, if you don't like your in boundary middle school, it makes sense to consider giving yourself some options by giving ordinal preference to schools with future preference at other middle schools. Since you can always switch to your in boundary middle school later, this option seems very valuable to some folks. It may be by the time you realize this later, it is too difficult to get into these other paths. This is the kind of strategizing the algorithm purports to stop but it doesn't. Let me think about this more to give a more thorough discussion. I should add that I'm probably wrong but it seems clear they haven't done any modeling of this as the repeated game that this is. Yep, I know about the Nobel Prize, blah blah blah.

People want optionality and risk management not absolute feality to dubious ordinal preferences with a one year look ahead, especially since life is uncertain.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:I do not understand why there are so many slots per entry. Does it increase or decrease your chances of getting in if you do not use all the slots? And why do people even put in slots that they don’t really want?

It increases your chances of getting into a school if you can list more schools. It will not affect how good your number is. But we matched with our #12 choice. If we could only list 10, we would have had no match. We did really want it, just not as much as we wanted the 11 above it. But we enrolled and it has been good so far.

I have a hard time believing that there are 12 schools that are that much better than an IB for a 3 year old when you take into account commutes. There aren’t 12 good public/charter MS options in the entire city. Six is a more reasonable number.

In most parts of the city that's true, but in Ward 5 where we live, population density and plenty of charters mean it's not hard to make a list of 11 schools better than IB.

When we were lotterying, I was looking to change my job and we were renters open to the idea of moving, so I really appreciated being able to have a lot on the list.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:Also, the "game" is repeated with high positive serial correlation whereas the DC Lottery implementation and associated proofs of efficiency assume only a single "game".

Please expand on this thought. (Formatting corrected)

The nice clean proof of the "optimality" of this algorithm rests on many debatable assumptions. One assumption is that this matching happens only once. Of course, the DC lottery happens every year. Ostensibly, the lottery result for this year only effects next year. However, this isn't true. It is a repeated game where prior results and future anticipated results effect what I should do. For example, if you don't like your in boundary middle school, it makes sense to consider giving yourself some options by giving ordinal preference to schools with future preference at other middle schools. Since you can always switch to your in boundary middle school later, this option seems very valuable to some folks. It may be by the time you realize this later, it is too difficult to get into these other paths. This is the kind of strategizing the algorithm purports to stop but it doesn't. Let me think about this more to give a more thorough discussion. I should add that I'm probably wrong but it seems clear they haven't done any modeling of this as the repeated game that this is. Yep, I know about the Nobel Prize, blah blah blah.

People want optionality and risk management not absolute feality to dubious ordinal preferences with a one year look ahead, especially since life is uncertain.

There are lots of problems with feeder school rights, especially in the context of the lottery. For one it makes the stakes higher than they need to be. Lottery into a Deal feeder for pre-k, you're set for 14 years.

[Anonymous]

If anyone wants to write a paper, let me know. It would be fun to jump into academia again for a bit.. Kidding, sorta

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:Also, the "game" is repeated with high positive serial correlation whereas the DC Lottery implementation and associated proofs of efficiency assume only a single "game".

Please expand on this thought. (Formatting corrected)

The nice clean proof of the "optimality" of this algorithm rests on many debatable assumptions. One assumption is that this matching happens only once. Of course, the DC lottery happens every year. Ostensibly, the lottery result for this year only effects next year. However, this isn't true. It is a repeated game where prior results and future anticipated results effect what I should do. For example, if you don't like your in boundary middle school, it makes sense to consider giving yourself some options by giving ordinal preference to schools with future preference at other middle schools. Since you can always switch to your in boundary middle school later, this option seems very valuable to some folks. It may be by the time you realize this later, it is too difficult to get into these other paths. This is the kind of strategizing the algorithm purports to stop but it doesn't. Let me think about this more to give a more thorough discussion. I should add that I'm probably wrong but it seems clear they haven't done any modeling of this as the repeated game that this is. Yep, I know about the Nobel Prize, blah blah blah.

People want optionality and risk management not absolute feality to dubious ordinal preferences with a one year look ahead, especially since life is uncertain.

There are lots of problems with feeder school rights, especially in the context of the lottery. For one it makes the stakes higher than they need to be. Lottery into a Deal feeder for pre-k, you're set for 14 years.

Exactly

[Anonymous]

Anonymous wrote:

Anonymous wrote:The system the OP is describing is (I think) pretty close to what DC used to do when each school ran its own lottery. it was a mess. The biggest issue from an economics point of view is that it led to a situation where there could have been a lot of mutually beneficial trades -- which means it was inefficient at allocating a scarce resource. For example under the old system it was entirely possible for the following scenario to take place:

KidA gets into MV and has a bad waitlist number for IT, his parents prefer IT

KidB gets into IT and has a bad waitlist number for MV, his parents prefer MV

Under the new system, that won't happen because the parents will rank their choices and if KidA has a good number, he will rank IT first and get in there. KidB would get into MV with a good number.

Ding ding ding ding! This is the correct answer. All the rest of you are wrong. OP, you’d need to argue against this suboptimal outcome. You can’t. You lose.

All the rest of you are also wrong.

OP here. You’re right.

If the algorithm uses more than one lottery number, this situation can occur. This is a “non-stable assignment”.

Ok, makes sense. It’s still bad that kids get one lottery number for all schools per year. Seems pretty sub-optimal.

It occurs to me that the algorithm is a deferred-assignment algorithm. So I think the problem you’ve mentioned above can be resolved by an additional round of swapping assignments to eliminate instability (of course, as in the current lottery system, this would all happen before any results are released.) Swapping could also occur for waitlist rankings to ensure stability. But that seems like a pretty complicated set of scenarios to try to ensure are non-game able.

Still, the one-number-per-year thing is a real bummer of the current system.

[Anonymous]

Anonymous wrote:I'm not sure why this optimally condition is so sacred, except that the result is so clean for this set of very strong assumptions. If we are worried parents would want to switch, why not allow for a scheme where parents can swap their assignments? Honestly, once parents get an assignment, they will start to value it more than they might have abstractly conceived of it beforehand.anyway. I count 132 options for PK3 and I can only list 12... strategizing is already happening.

The Nobel lecture online from Roth explains some of this. Allowing parents to switch puts on huge pressure to act quickly and also heavily advantages people who have a lot of information. The property that there are no swaps that benefit both parents is what prevents these failures.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:The system the OP is describing is (I think) pretty close to what DC used to do when each school ran its own lottery. it was a mess. The biggest issue from an economics point of view is that it led to a situation where there could have been a lot of mutually beneficial trades -- which means it was inefficient at allocating a scarce resource. For example under the old system it was entirely possible for the following scenario to take place:

KidA gets into MV and has a bad waitlist number for IT, his parents prefer IT

KidB gets into IT and has a bad waitlist number for MV, his parents prefer MV

Under the new system, that won't happen because the parents will rank their choices and if KidA has a good number, he will rank IT first and get in there. KidB would get into MV with a good number.

Ding ding ding ding! This is the correct answer. All the rest of you are wrong. OP, you’d need to argue against this suboptimal outcome. You can’t. You lose.

All the rest of you are also wrong.

OP here. You’re right.

If the algorithm uses more than one lottery number, this situation can occur. This is a “non-stable assignment”.

Ok, makes sense. It’s still bad that kids get one lottery number for all schools per year. Seems pretty sub-optimal.

It occurs to me that the algorithm is a deferred-assignment algorithm. So I think the problem you’ve mentioned above can be resolved by an additional round of swapping assignments to eliminate instability (of course, as in the current lottery system, this would all happen before any results are released.) Swapping could also occur for waitlist rankings to ensure stability. But that seems like a pretty complicated set of scenarios to try to ensure are non-game able.

Still, the one-number-per-year thing is a real bummer of the current system.

OMG. It is not the number of numbers that is causing some people to have a bad outcome! There are not enough high-quality seats and that is the problem. No lottery system is going to create more seats. No matter how the lottery works, someone is going to have a bad outcome and feel that the system is a bummer. Catch on already.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:The system the OP is describing is (I think) pretty close to what DC used to do when each school ran its own lottery. it was a mess. The biggest issue from an economics point of view is that it led to a situation where there could have been a lot of mutually beneficial trades -- which means it was inefficient at allocating a scarce resource. For example under the old system it was entirely possible for the following scenario to take place:

KidA gets into MV and has a bad waitlist number for IT, his parents prefer IT

KidB gets into IT and has a bad waitlist number for MV, his parents prefer MV

Under the new system, that won't happen because the parents will rank their choices and if KidA has a good number, he will rank IT first and get in there. KidB would get into MV with a good number.

Ding ding ding ding! This is the correct answer. All the rest of you are wrong. OP, you’d need to argue against this suboptimal outcome. You can’t. You lose.

All the rest of you are also wrong.

OP here. You’re right.

If the algorithm uses more than one lottery number, this situation can occur. This is a “non-stable assignment”.

Ok, makes sense. It’s still bad that kids get one lottery number for all schools per year. Seems pretty sub-optimal.

It occurs to me that the algorithm is a deferred-assignment algorithm. So I think the problem you’ve mentioned above can be resolved by an additional round of swapping assignments to eliminate instability (of course, as in the current lottery system, this would all happen before any results are released.) Swapping could also occur for waitlist rankings to ensure stability. But that seems like a pretty complicated set of scenarios to try to ensure are non-game able.

Still, the one-number-per-year thing is a real bummer of the current system.

OMG. It is not the number of numbers that is causing some people to have a bad outcome! There are not enough high-quality seats and that is the problem. No lottery system is going to create more seats. No matter how the lottery works, someone is going to have a bad outcome and feel that the system is a bummer. Catch on already.

Yes. Before common lottery we applied for a zillion charters and were shut out of all of them. In fact, people then were JUST as likely to not get a school they wanted as people now because its supply and demand. No matter the system there aren't enough spots at schools people want.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:The system the OP is describing is (I think) pretty close to what DC used to do when each school ran its own lottery. it was a mess. The biggest issue from an economics point of view is that it led to a situation where there could have been a lot of mutually beneficial trades -- which means it was inefficient at allocating a scarce resource. For example under the old system it was entirely possible for the following scenario to take place:

KidA gets into MV and has a bad waitlist number for IT, his parents prefer IT

KidB gets into IT and has a bad waitlist number for MV, his parents prefer MV

Under the new system, that won't happen because the parents will rank their choices and if KidA has a good number, he will rank IT first and get in there. KidB would get into MV with a good number.

Ding ding ding ding! This is the correct answer. All the rest of you are wrong. OP, you’d need to argue against this suboptimal outcome. You can’t. You lose.

All the rest of you are also wrong.

OP here. You’re right.

If the algorithm uses more than one lottery number, this situation can occur. This is a “non-stable assignment”.

Ok, makes sense. It’s still bad that kids get one lottery number for all schools per year. Seems pretty sub-optimal.

It occurs to me that the algorithm is a deferred-assignment algorithm. So I think the problem you’ve mentioned above can be resolved by an additional round of swapping assignments to eliminate instability (of course, as in the current lottery system, this would all happen before any results are released.) Swapping could also occur for waitlist rankings to ensure stability. But that seems like a pretty complicated set of scenarios to try to ensure are non-game able.

Still, the one-number-per-year thing is a real bummer of the current system.

OMG. It is not the number of numbers that is causing some people to have a bad outcome! There are not enough high-quality seats and that is the problem. No lottery system is going to create more seats. No matter how the lottery works, someone is going to have a bad outcome and feel that the system is a bummer. Catch on already.

Yes. Before common lottery we applied for a zillion charters and were shut out of all of them. In fact, people then were JUST as likely to not get a school they wanted as people now because its supply and demand. No matter the system there aren't enough spots at schools people want.

And, before the common lottery there was little visibility/process in how schools moved their waitlists. People who knew staff at the school could hop up the wait list because it wasn’t common/known across the system.

[Anonymous]

Anonymous wrote:

Still, the one-number-per-year thing is a real bummer of the current system.

Imagine there was only one other person in the lottery, but only one seat. How would you rather settle it: a single coin toss, or each of you flips a coin 100 times and whoever gets the most heads wins?

You do realize that having more flips doesn't change your chances of winning right? And it certainly doesn't change the fact that there is only one seat for two people.

[Anonymous]

Anonymous wrote:

Anonymous wrote:The problem isn't that the lottery is poorly designed. The problem is that high-quality education should not be a scarce resource that is allocated by lottery in a first-world country.

I don't want more choices. I just want nicer things.

Then move to the suburbs.

Good idea, that will improve the schools. /s

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:The problem isn't that the lottery is poorly designed. The problem is that high-quality education should not be a scarce resource that is allocated by lottery in a first-world country.

I don't want more choices. I just want nicer things.

Then move to the suburbs.

Good idea, that will improve the schools. /s

Having your family in the school is not the improvement you may think it is.