Fixing the Wait List - Should BPS be More Like USAir?
One of the biggest problems parents have with the current student assignment process is wait lists. Many children get assigned to a school that's not their first choice and end up on one or more wait lists for higher choice schools. Even worse, many kindergarten children get no assignment and only get on wait lists. And wait lists often don't move until September when kids who are assigned to schools fail to show up for the first several days of school.
Another issue is the number of wait lists a child can be placed on. My memory is that BPS used to put children on an unlimited number of wait lists. If a kid got her 9th choice, she could be on the wait list for all eight of the schools she didn't get assigned to. Now BPS will only put a child on up to three wait lists. This causes parents to try to game an "un-gameable" system. The lottery algorithm in use since 2006 is supposed to encourage families to list schools in their true order of preference. But some families will put an under-chosen school in their top three just to increase their chances of at least getting a decent spot on a wait list.
BPS also used to limit the number of children on the wait list. The idea was not to have kids so far down wait lists that they have no chance to get in. Now a child will be placed on wait lists for their top three schools no matter how far down the list they are.
I don't understand the reasons behind these changes. I'd like to see a return to an unlimited number of wait lists but with a reasonable limit on how many kids would be on the list for each school.
The good news is that BPS is doing a couple of things already to try to improve things. One change is to allow families of kindergarten children to ask to be administratively assigned. This means that if the child doesn't get into any of his or her choices, BPS will assign the child to the closest school in the child's zone with an available seat. This means families won't have to go back to a family resource center if they don't receive an assignment. I haven't seen any official data on this, but I heard that an overwhelming majority of families are checking the box.
BPS hasn't given a lot of detail on the other initiative they're undertaking, but my understanding is that they're planning to call families who have been assigned to schools to confirm their child's attendance. BPS already sends letters to families each spring asking them to confirm whether or not their child will attend school in the fall. However, because of the nature of the district's population, BPS does not remove a child from a school's roster just because the letter isn't returned. Instead, they only remove children if parents confirm that the child will not be attending. By contacting parents by phone, the district may be able to remove some of those children before school starts and move the wait lists sooner.
In order to really address this problem, BPS has to improve quality and make sure schools are located where the students are. But in the short term there are ways to significantly reduce the problem.
I wrote about one change that could be made in a previous post. If charters and district schools were assigned through a single process, fewer students would be on waiting lists when they had already chosen a preferred school.
Another way that BPS could help families would be to implement the most loathed practice of a fairly unpopular industry: overbooking. Airlines routinely overbook. As passengers, we're only aware of it when we get bumped or when we hear them offering hundreds of dollars in vouchers to passengers willing to take another flight. In those cases it may cost the airlines, but in general they make a lot more money by overbooking and filling more seats.
Airlines do fairly sophisticated forecasts based on historical cancelation and no-show rates for various classes of service. BPS could do the same thing by looking at what percentage of kids usually show up at each school. If siblings show up at a higher rate (and I suspect they do), that could be factored in. BPS could then conservatively overbook each kindergarten class. If a class started the year with one or two extra kids, it wouldn't be the end of the world. In fact, it would probably be less disruptive than the musical chairs that takes place in September now when some kids don't show and others start at one school then move to another when they get in off the wait list.
The impact of this would be much larger than it may appear. For example, let's assume that the popular Thomas M. Menino Elementary has two kindergarten classes and each is overbooked by two kids. We would expect that to reduce each school's wait list by four kids. However, the impact could actually be much greater. There may be kids assigned to the Menino who put the Raymond L. Flynn K-8 as their first choice. By also overbooking the Flynn, two of those kids may be assigned to the Flynn, freeing up two more spaces for kids who would otherwise be on the Menino wait list. At the same time there may be kids who are on the wait for both the Menino and the Kevin H. White Early Learning Center who put the White ELC as their first choice. The White ELC also overbooks, so those kids get assigned to the White.
The end result is that the Menino has a much smaller waiting list. More importantly, far fewer parents are waiting until the second week of September to get an assignment. And there would be a lot less movement after school starts which would benefit both families and schools.
What else could be done to reduce this problem? Make your suggestions in the comments.
In my postmortem thinking about this process I have often thought that one of the biggest issues that was never addressed was the ability of the BPS to set the size of the school at whatever they choose to, with no responsibility to try to accomidate people in school of their choice, within reasonable limitations.
If you look at the building capacity spreadsheet on this page (http://bostonschoolchoice.org/raw-data/raw-data-2/) a lot of the popular schools that I know about have a building capacity that is at least 50% greater then the current enrollment of the school. Not sure about the accuracy of these numbers, but to me any school that constantly has a long waitlist should be maxed out of physical capacity.
Lotto or not it would have probably made a bigger difference then anything else if there was some commitment in the new plan that popular schools would have to add some amount of seats every year (up to the capacity of the building). 99% of school districts have to provide this kind of flexibility because they do not have a lotto, so it can be done.
In reality you could have a lotto combined with a commitment to get the waitlists as short as possible, but it would require enrollment flexibility, as you state when talking about overbooking. Unfortunately the BPS is given the luxury of setting the number of seats at whatever number they want. I am sure they like it that way since it probably makes planning a management a whole lot easier, but it really ties your hands when talking about waitlists.
The capacity numbers are interesting
I was also curious about those capacity numbers. With enrollment rising, BPS is definitely looking for schools with extra space. And with the weighted student funding formula, principals now have incentive to maximize the number of students in their schools.
I don't think these numbers are really accurate. I suspect that part of the problem is that they're based on the number of students the schools were built for which is probably based on much larger class sizes than we have now. I'll try to find out how they were calculated.
Of course what we really need is a long-term facilities plan that takes into account projected future enrollment and actually aims to match supply to demand in every part of the city.
One thing I learned when we were on a waitlist is that the principal has a LOT of leeway with letting kids in. The K1 class at Hurley that year was underenrolled for most of the fall, but we did not get a seat until late October because the principal somehow managed to keep those seats open. What I heard through the grapevine was that there were a lot of kids who were having a lot of trouble adjusting to a classroom setting and so she did not want to fill it up right away. My takeaway is that the principals have some control over when empty seats are reported to enrollment.
I'm not sure what this adds - although it does make me think more more positively toward oversubscribing classes (we never would have been on the waitlist!). But I think you could run in to trouble if it was more than one or two seats in highly-desired schools….on the other hand, the last seat in that class was never filled and there were never more than 21 kids in it.
I've also heard that principals inform BPS of open spaces so they do control when kids are admitted. Open spaces reduce funding, so there is some incentive to fill those spots.
I think people should be able to subscribe to as many wait lists as possible. Maybe it could be part of the lottery questions. When a school is chosen it could be listed as a waitlist school or not. After the results of the lottery parents should be able to add and substract schools from their list.
This could move the wait lists faster. I know that I regret listing my third choice school. I would much rather have my fourth and fifth choice options be wait list choices. But I can't make the switch until the next round of the lottery. And then it seems like too much of a hassle.
I was wondering how applying to charter schools fits into this. A neighbor's daughter got into the Dudley School and about the same time got into the Eliot. She told BPS that she wanted to choose the Eliot but they said that the Dudley was her first choice so she couldn't take the Eliot spot. I would be interested to learn more how charter school admmissions fit into the process.
Different for K-1?
Is it different for K1? We had the Quincy school in our top 3, and didn't even get on the wait list. I assumed that they were cutting off the wait lists at some number, when there was no chance of getting in.
Not sure on K1
The BPS web site says nothing about differences between K1 and other grades. It also doesn't say anything explicitly about whether there are any limits on the lengths of wait lists. So as with many things at BPS, it's pretty hard to say how this really works.
It's funny, because when I thought about this thing where you only get 3 waitlists I realized that it is not correct when they tell you the system is "un-gameable". The top 3 slots on your list have more value then the others, so if you waste one of your top 3 slots on a very popular school that is not in your walkzone you are losing the possibility of being on a shorter waitlist for a school that you might like almost as much closer to home.
In the west zone I did not even list Kilmer and Lyndon because I am not in the walk zone and getting in is very unlikely, but I have a feeling a fair amount of people spend one their valuable top 3 slots on those schools, which I suspect is a bad idea.
Implementing the new system
But isn't it interesting that you - as a presumably well-informed parent - believed "I did not even list Kilmer and Lyndon because I am not in the walk zone and getting in is very unlikely" - when in fact walk zone turns out not to matter that much.
We have a similar story - other parents (usually an excellent source of info) told us it made a difference in your chance of getting in to list a school as your first choice. When in fact the algorithm doesn't work that way. (Though maybe with principals getting you in it does …)
The point being - there really needs to be a very clear explanation of how the system works and what it means, especially in implementing the new assignment plan.
I don't think listing a
I don't think listing a school first makes a difference once you are looking to get in off the waitlist vis a vis the principal influence I mentioned earlier. I don't think the principal knows whether you listed a school first or last on your list when you registered. To be clear - the principal influence doesn't really have anything to do with getting a specific kid in, it's just that they are more of a gatekeeper for getting *bodies* in that I realized.
That stinks that people told you it matters which you listed first. :(
I have actually read in quite a bit of detail the arguments that the walkzone does not matter that much but I still stand by my decision to not list popular out of walkzone schools this year, for these reasons:
1) It was sort of confused by the end how much the "walkzone does not matter" argument was talking about the new system vs the old system. In the new system a lot of schools do really not have a zone much bigger then their walkzone, so it would probably not matter much, but right now I have to think that lots of outside the walkzone people apply to the especially well regarded schools. I know how they explained the math, but in the end of the day if a lot of out of walkzone people apply to a school their odds have to be less. That may not happen for a lot of schools, but if it happens for any it will be the really popular ones.
2) While I think it may very well be true that the walkzone does not matter that much as an average over the whole system, I would be quite surprised if it does not matter for the really popular schools. Walkzone could be quite significant for the handful of really popular schools, but not really show up that much when spread over the 90 schools or whatever in the system. Walkzone is going to be especially significant when a lot of out of walkzone people apply to the same school. Probably not many people are going to apply to a not very popular school out of their walkzone, but a lot of people may apply to popular ones.
3) The only people who got to look at this data (I asked several times for more detailed data and was not responded to since I am a critic of the new system) were people who wanted to get rid of the walkzone. So, I would not be surprised if they took the numbers from their study that minimized the walkzone impact. The most obvious way to do this being to look at a large average vs. its impact on the popular schools where it is most important.
Anyway, regardless of the impact of the walkzone, I still think one can "game" the system since if you use a top spot on a very popular school you may end up on a long waitlist. Meanwhile, you may only miss a less popular school by 1 spot, but if you have it as 4th on your list you will have no chance to get a seat, while if you had listed it 3rd at the expense of a small chance of getting into a really popular school you would basically be in.
There's lots of mythology out there
First choice was very important up until 2006 when the algorithm changed. I've heard all kinds of things. That principals can choose kids off the wait list, that race is used in the algorithm, etc. I think there's still a lot of mistrust and confusion which causes this bad information to spread. Transparency and simplification could help.
There are also quite a few exceptions built in to the system. The school committee should really review those in light of the new plan and eliminate any that are no longer needed.
I know from personal experience that principals used to have a lot of power to hold seats open but I also know from working for the BPS that upon roll out of the new Student Information System that principals lost that freedom- Court St now can see how many kids show up instantly. BPS also is much stricter about the showing within the first week of school rule and there are attendance codes that are used for those kids to alert Court St to the fact that there are possible seats.
Thanks, Maura. I was wondering if SIS might change this.